the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Improving the agricultural erosion management in Finland through high-resolution data
Abstract. Soil erosion reduces the sustainability of agricultural sector by loss of productive soil and through negative impacts on surface waters. In Finland, considerable efforts have been made to reduce soil erosion, but the suspended sediment loads to surface waters have not markedly reduced. A major limitation has been the lack of high-resolution data on erosion risk for efficient targeting of the erosion management efforts. In this study, by using the Revised Universal Soil Loss Equation (RUSLE) a two-meter resolution erosion risk data was developed and consequently the spatial distribution of the erosion risk of Finnish agricultural land was analysed. With agricultural management practices of 2019, the average erosion of agricultural land was estimated to be 430 kg ha−1 yr−1, and it varied at the municipality scale from 100 to 1290 kg ha−1 yr−1. At more local scales the erosion risk had even greater variability, and areas with high erosion risk were differently located in terms distances to water bodies. The results also suggest that the past erosion management efforts have not been well-targeted according to the actual erosion risk. Altogether, the results indicate that erosion mitigation measures can be improved by inclusion of high-resolution data in the planning and implementation of the measures, by considering the spatial variability of the erosion risk over multiple spatial scales, and by implementation of location specific erosion reduction measures.
- Preprint
(1470 KB) - Metadata XML
-
Supplement
(1141 KB) - BibTeX
- EndNote
Status: closed
-
EC1: 'Comment on hess-2021-457', Christian Stamm, 18 Oct 2021
\begin{document}
{\parindent0pt % disables indentation for all the text between { and }
Comments hess-2021-457\\\
Dear authors,\\\
I list a number of more general and detailed issues. \\\
General aspects:\\\
\begin{description}
\item[Language:] In general, the text is easy to read. Still, sometimes there are issues with the grammar such as missing articles (e.g., L. 19, 34 - 35).
\item[Typos:] There are are few instances with misspellings (e.g., L. 38, 218).
\end{description}Detailed comments:\\\
\begin{description}
\item[L. 48:] Strange sentence. What does \textit{implementation through natural constraints} mean?
\item[L. 132 - 133:] How was the LS factor linked to the actual parcels? How do upslope fields influence downslope parcels?
\item[L. 133:] What is the empirical basis for the claim that sink filling increases the errors? Sinks in a DEM can be real (and should be accounted for) or can be artifacts. Why did you not distinguish between the two situations?
\item[L. 162:] According to my knowledge, the RUSLE model does conceptually not account for sub-surface transport through tile-drains. Nevertheless, you compare RUSLE simulations to empirical data of the sum of surface and subsurface sediment transport. Should that not be reflected in a conceptual modification of the RUSLE model including a model parameter accounting for the split between surface and subsurface transport?
Additionally, the subsurface flow can induce mobilisation of soil particles also within the soil profile, especially in the vicinity of subsurface drains because of the disturbances of the soil profile due to the installation of the drains. How is this accounted for?
\item[L. 274 (Tab. 4):] Please provide the number of observation years and the standard deviation of the measured erosion.
\item[L. 303 - 304:] Are these novel findings?
\item[L. 318 - 319 (Fig. 4):] The high resolution DEM only affects the the LS factor, doesn't it? Hence, only this map should make any difference to previous estimates, shouldn't it?
\item[L. 349 (Fig. 6):] To which degree are these findings novel?
\item[L. 353:] Replace \textit{high field area} by \textit{areas with a large fraction of arable land} (or similar).
\item[L. 357 - 358:] Is that statement not trivial given the definition of the EMI index?
\item[L. 445 - 446:] Where is the evidence that it was indeed the lack of high resolution risk maps that prevented the implemented of targeted measures?
\item[L. 448 - 452:] The four bullet points seem rather similar to me. Can you more precisely explain what the differences are?
\item[L. 454 - 456:] Is this a novel result?
\item[L. 458 - 460:] This seems to be quite standard knowledge, or am I wrong?
\item[L. 471 - 472:] Given that you have access to actually crop management data, it should be straight forward to assess the effects such modification in practice, shouldn't it?
\item[L. 477 - 478:] Where is the evidence for that? It is a frequently used arguments by natural scientists that improved model will enhance management, but which evidence demonstrates the validity of the claim?
\item[L. 481:] The previous erosion risk estimates were rather similar (see L. 384 - 389). So in which sense has the understanding of erosion risk considerably been improved?
\item[L. 487 - 488:] What do you mean by considering erosion risk across multiple scales? What does it mean from a scientific point of view, what does it mean in practice?
\item[L. 489 - 490:] Which aspect provides new opportunities for analysing the P- and C cycle given the similarity of previous erosion estimates?
\item[L. 491 - 492:] Where can one see this demonstration? The manuscript does not compare how policies or planning has changed due to the new erosion risk map.
\end{description}Sincerely \\\
Dr. Christian Stamm
Editor HESS
}\end{document}
Citation: https://doi.org/10.5194/hess-2021-457-EC1 -
AC1: 'Reply on EC1', Timo Räsänen, 10 Dec 2021
Dear Christian Stamm (Editor),
Thank you for your thoughtful and constructive comments (EC) – they helped to improve the manuscript. Altogether, the comments from the editor and the two referees prompted major revisions, and we have revised the manuscript accordingly. The major revisions are:
- The country scale results on actual erosion and erosion management index were removed from the revised manuscript due to limitations in the used C-factor data, which was pointed out by Pedro Batista (Referee #2). Correction of the C factor data requires considerable work and we decided to leave it as future work to be published in another publication.
- The LS factor and consequent erosion data were recalculated to account for field borders.
- The results on RUSLE evaluation, potential erosion risk and potential erosion risk near water bodies were slightly restructured to accommodate the changes caused by removal of the two parts in mentioned above.
- New sensitivity analyses were added that provide estimates on
- the propagation of uncertainties from RUSLE factors to erosion estimates
- effects of location specific cropping and management practices and temporal rainfall erosivity distribution on C- factor values and the consequent erosion estimates
- The terms “potential erosion risk” and “actual erosion risk” were replaced by terms, “erosion susceptibility” and “actual erosion” to avoid the misuse of the term “risk”.
Thus, the new findings of the revised manuscript are:
- New high-resolution (two-meter) country scale erosion susceptibility data for Finland
- New evaluation of RUSLE and its performance in boreal conditions, which considers also different spatial scales and issues related to upscaling from field parcel to larger spatial scales
- Improved scientific understanding of agricultural erosion and its spatial distribution
These findings provide new opportunities for research and erosion management. In the following we provide our comments and responses (AC) to the editor comments (EC) point by point.
Comments and answers:
EC1: [Language:] In general, the text is easy to read. Still, sometimes there are issues with the grammar such as missing articles (e.g., L. 19, 34 - 35). [Typos:] There are few instances with misspellings (e.g., L. 38, 218).
AC1: We will proofread the manuscript and correct the grammar and misspelling issues.EC2: [L. 48:] Strange sentence. What does “implementation through natural constraints” mean?
AC2: This refers to the national system of paying subsidies for farmers for implementing environmental measures, such as the buffer zones. The discussion on this was removed from the manuscript as it was less relevant after the omissions in the manuscript.EC3: How was the LS factor linked to the actual parcels? How do upslope fields influence downslope parcels?
AC3: The LS factor was recalculated in the revised manuscript to account for the field parcel borders. We explain this in revised manuscript in a following way: “The LS factor was calculated from a two-meter resolution LiDAR-based digital elevation model (DEM) of Finland (National Land Survey of Finland, 2020), and by using the SAGA-GIS Module LS Factor (Conrad, 2013) and the method of Desmet and Govers (1996 ) with default settings. The LS calculation was performed in two-meter resolution for agricultural lands in 301 hydrological units that consisted of river basins, sub-basin groups (Finnish Environment Institute, 2010) and groups of islands (Fig. S1). The agricultural lands were defined according to the field parcel data from Finnish Food Authority, which contains over one million vectorized field parcels and accounts almost all agricultural land in Finland. The use of vectorized field parcels treated each field parcel as an isolated hydrological unit (in terms of overland flow) in the LS calculation to account for the effects of varying landcover on surface runoff, as recommended by Desmet and Govers (1996). The approach is considered justified as in Finland the fields are typically well drained and commonly surrounded by open ditches, which advocates for the hydrological isolation of the field parcels. However, adjacent field parcels that shared the same parcel border were treated in the calculation as a single field parcel, since it is common that uniform field areas are divided into separate field parcels for cropping and management purposes.” The evaluation of RUSLE at the seven field sites was not affected by the recalculation of the LS. Their LS factors were originally calculated considering the field borders.EC4: [L. 133:] What is the empirical basis for the claim that sink filling increases the errors? Sinks in a DEM can be real (and should be accounted for) or can be artifacts. Why did you not distinguish between the two situations?
AC4: We removed the statement that it increases errors. Our observation was that sink filling with a two-meter resolution DEM resulted in flattening and raising of the field surface levels to the levels of neighbouring landforms, such as embankments and roads. This significantly distorted the DEM and the original surface characteristics of the fields were lost. Also, as mentioned in the EC4, fields often have natural depressions, and breaching was observed to create artificial erosion areas. We are not aware of research that would provide suggestions for correct treatment of high-resolution DEM for LS calculation on agricultural lands. We revised the justification on treatment of DEM (or lack of) in the manuscript to be clearer.EC5: [L. 162:] According to my knowledge, the RUSLE model does conceptually not account for sub-surface transport through tile-drains. Nevertheless, you compare RUSLE simulations to empirical data of the sum of surface and subsurface sediment transport. Should that not be reflected in a conceptual modification of the RUSLE model including a model parameter accounting for the split between surface and subsurface transport? Additionally, the subsurface flow can induce mobilisation of soil particles also within the soil profile, especially in the vicinity of subsurface drains because of the disturbances of the soil profile due to the installation of the drains. How is this accounted for?
AC5: According to Renard et al. (1997) subsurface drainage is considered in the P factor. Bengtson and Sabbagh (1990) suggested an average P factor value of 0.6, which was also recognised by Renard et al. (1997). However, the research on how the subsurface drainage should be considered in the RUSLE is limited, and to our understanding there is no commonly accepted approach for this. Except, RUSLE2 (USDA, 2013) incorporates a method for this, which considers the changes in K factor due to the subsurface drainage, but in our case the data was too limited to consider this. Therefore, we chose to follow the original suggestions by Renard et al. (1997) and Bengtson and Sabbagh (1990). According to our literature review the subsurface drainage reduce erosion 8-90% (on average 38%) (Bengtson et al., 1988, 1984; Bengtson and Sabbagh, 1990; Formanek et al., 1987; Gilliam et al., 1999; Grazhdani et al., 1996; Istok et al., 1985; Maalim and Melesse, 2013; Skaggs et al., 1982), which results to average P value of 0.62. We used the P value of P 0.6 as suggested and used earlier in Finland by Lilja et al. (2017a). Also, the research in Finland showed that substituting of old drainage pipes with new ones reduced erosion up to 15% on a clay soil (Turtola and Paajanen, 1995). The use of sum of surface and subsurface sediment is justified also by research. In Finland, it is observed that up to 50-90 % of the erosion loading from clay soils occurs via subsurface drainage (Finnish Environment Institute, 2019; Turtola et al., 2007; Turunen et al., 2017; Warsta et al., 2014, 2013) and that erosion material in subsurface drainage flow from clay soils originates mainly from the surface soil (Uusitalo et al., 2001). In the Finnish research, the origin of the erosion material was determined by an analysis of Cesium-137 contents of soil layers and eroded soil material in subsurface drainage flow. Also, process-based modelling studies in Finland suggest that majority of the erosion material in subsurface drainage flow originates from the surface soil (Turunen et al., 2017). A study from Norway also reports that soil material in the drain flow originated most likely from the plough layer, and the soil material was transported to subsurface drains via cracks and macropores in the soil (Øygarden et al., 1997). However, we do acknowledge that the consideration of subsurface drainage in the P factor lumps a complex and poorly understood process into a single value and has therefore limitations and is a considerable source of uncertainty. In the revised manuscript we perform a sensitivity analysis which includes the uncertainty in the P value. This justification was added to the revised manuscript. Note also that our modelling approach tests the above assumptions against empirical data and the resulting performance can be considered reasonable.EC6: [L. 274 (Tab. 4):] Please provide the number of observation years and the standard deviation of the measured erosion.
AC6: These have been added to the revised manuscript.EC7: [L. 303 - 304:] Are these novel findings?
AC7: To our knowledge these areas have not been identified as high erosion areas earlier, and therefore these findings are novel.EC8: [L. 318 - 319 (Fig. 4):] The high-resolution DEM only affects the LS factor, doesn't it? Hence, only this map should make any difference to previous estimates, shouldn't it?
AC8: The R (Panagos et al., 2015) and K factor data (Lilja et al., 2017a, 2017b) are based on existing data, but the LS data is new and created in the manuscript. Thus, the LS factor data, and the erosion estimates calculated with the LS data are new. Erosion has not been estimated earlier over the whole Finland with RUSLE at two-meter resolution. Lilja et al. (2017b) started the two-meter resolution modelling work, but it was not finished and not published.EC9: [L. 349 (Fig. 6):] To which degree are these findings novel?
AC9: According to the Pedro Batista (referee #2), the results in Fig. 6 contain large uncertainties due to limitations in the C factor. After careful consideration and performed sensitivity analyses, we agree with his view. The C factors vary by location and this was not considered in the submitted manuscript, which caused regional biases in the erosion estimates. Therefore, the results in Fig. 6 were omitted from the revised manuscript, and their correction will be addressed in future research. However, the sensitivity analyses prompted by Pedro Batista’s thoughtful comments are a new addition to the revised manuscript. The issues related to the C factor are discussed in more detail in the Batista’s comments and in our answers to him. Despite the omissions, the revised manuscript provides still substantial new findings as explained in the beginning of our comments. Reporting the sensitivities in the northern conditions is considered to provide valuable information regarding erosion assessments.EC10: [L. 353:] Replace “high field area” by “areas with a large fraction of arable land” (or similar).
AC10: This section was removed from the revised manuscript.EC11: [L. 357 - 358:] Is that statement not trivial given the definition of the EMI index?
AC11: This section was removed from the revised manuscript.EC12: [L. 445 - 446:] Where is the evidence that it was indeed the lack of high-resolution risk maps that prevented the implemented of targeted measures?
AC12: We have removed this statement from the revised manuscript.EC13: [L. 448 - 452:] The four bullet points seem rather similar to me. Can you more precisely explain what the differences are?
AC13: These are affected by the omissions and therefore the policy and management implications were thus revised, and they are now as follows: “The developed erosion susceptibility data markedly improves the basis for analysing the agricultural erosion over multiple spatial scales and consequently provides new opportunities for planning the erosion management. For example, the current data showed large areas with high agricultural intensity and high erosion susceptibility (e.g., coastal areas in Southern Finland), and how erosion varies locally (e.g., Karjaanjoki and Paimionjoki basins), including areas where field parcels near water bodies and have high erosion susceptibility. Such information can be used to guide planning and allocation of erosion management efforts. The consideration of different spatial scales is also important as different scales were found to provide different insights to erosion, which can affect the conclusions drawn from the data and the choice of erosion management measures. Larger scales can provide indication of broader areas needing erosion management, and the local scales reveal more exact locations of high erosion field parcels and help in choosing appropriate erosion management measures. Altogether, the developed data can be used to improve erosion management from policy to actual management levels. However, the use of the data requires understanding of the related uncertainties that were also clarified in this research.”EC14: [L. 454 - 456:] Is this a novel result?
AC14: The results from evaluation of RUSLE are novel, and they provide improved understanding of performance of RUSLE in boreal conditions and provide new estimates on the effects of different crop and management practices on erosion.EC15: [L. 458 - 460:] This seems to be quite standard knowledge, or am I wrong?
AC15: Agreed. Our intention here was to address a broader audience and to underline an important issue, which is sometimes neglected in practical management. Therefore, we wish to keep this sentence in the manuscript.EC16: [L. 471 - 472:] Given that you have access to actually crop management data, it should be straight forward to assess the effects such modification in practice, shouldn't it?
EC16: Yes, it should be straightforward with appropriate C values, but this work is considered to be outside the scope of the current manuscript.EC17: [L. 477 - 478:] Where is the evidence for that? It is a frequently used arguments by natural scientists that improved model will enhance management, but which evidence demonstrates the validity of the claim?
AC17: In this particular case the argument originates not only from natural scientists, but from our personal communication with actors involved planning and implementation of the environmental measures in the agricultural sector, such as the Ministry of Agriculture and Forestry and the Finnish Food Authority who are responsible of allocation of environmental measures in Finland. Of course, there is never a guarantee that improved system understanding leads to improved management outcomes as the implementation depends on variety of factors.EC18: [L. 481:] The previous erosion risk estimates were rather similar (see L. 384 -389). So in which sense has the understanding of erosion risk considerably been improved?
AC18: The current research provides a new spatially explicit data and information on erosion in high-resolution over the whole Finland and such data and information has not existed before. The earlier work does not provide either the same spatial resolution or coverage as the current work, and the spatial distribution of erosion on country scale has not been well analysed and presented in earlier publications.EC19: [L. 487 - 488:] What do you mean by considering erosion risk across multiple scales? What does it mean from a scientific point of view, what does it mean in practice?
AC19: The manuscript shows how different scales reveal different spatial patterns in erosion. From scientific point of view this means that conclusions drawn from analyses will depend on the analysis scale, and analysis only in one scale provides a limited view on spatial distribution of erosion. From practical point of view this can affect how management of erosion is approached. Broader scales can reveal regions where greater erosion management effort is needed, and local scales can provide insights into efficient location-specific targeting of mitigation measures. We added following to the revised manuscript:” The consideration of different spatial scales is also important as different scales were found to provide different insights into spatial distribution of erosion, which can affect the conclusions drawn from the data and the choice of erosion management measures. Larger scales can provide indication of broader areas needing erosion management, and the local scales can reveal the locations of the high erosion areas within the field parcels, and consequently help in choosing appropriate erosion management measures for a given location.”EC20: [L. 489 - 490:] Which aspect provides new opportunities for analysing the P- and C cycle given the similarity of previous erosion estimates?
AC20: We have decided to simplify the discussion and remove this from the manuscript.EC21: [L. 491 - 492:] Where can one see this demonstration? The manuscript does not compare how policies or planning has changed due to the new erosion risk map.
AC21: We have removed this statement from the manuscript.References
Bengtson, R.L., Carter, C.E., Morris, H.F., Bartkiewicz, S.A., 1988. The Influence of Subsurface Drainage Practiceson Nitrogen and Phosphorus Losses in a Warm, Humid Climate. Transactions of the ASAE 31, 0729–0733. https://doi.org/10.13031/2013.30775
Bengtson, R.L., Carter, C.E., Morris, H.F., Kowalczuk, J.G., 1984. Reducing Water Pollution with Subsurface Drainage. Transactions of the ASAE 27, 0080–0083. https://doi.org/10.13031/2013.32739
Bengtson, R.L., Sabbagh, G., 1990. USLE P factors for subsurface drainage on low slopes in a hot, humid climate. Journal of Soil and Water Conservation 45, 480–482.
Conrad, O., 2013. Module LS-Factor, Field Based [WWW Document]. SAGA-GIS Module Library Documentation (v2.1.4). URL http://www.saga-gis.org/saga_tool_doc/2.1.4/ta_hydrology_25.html (accessed 5.29.20).
Desmet, P.J.J., Govers, G., 1996. A GIS procedure for automatically calculating the USLE LS factor on topographically complex landscape units. Journal of Soil and Water Conservation 51, 427–433.
Finnish Environment Institute, 2019. Sediment and nutrient loading to surface waters in 3 different scales [WWW Document]. Finnish Environment Institute (SYKE). URL https://metasiirto.ymparisto.fi:8443/geoportal/catalog/search/resource/details.page?uuid=%7B15893DD0-0193-40AD-9E21-452D271DB791%7D (accessed 1.25.21).
Finnish Environment Institute, 2010. Ranta10 - rantaviiva 1:10 000 - SYKE [WWW Document]. URL https://ckan.ymparisto.fi/dataset/%7BC40D8B4A-DC66-4822-AF27-7B382D89C8ED%7D (accessed 3.25.21).
Formanek, G.E., ROSS, E., Istok, J., 1987. Subsurface drainage for erosion reduction on croplands in northwestern Oregon. In: Irrigation Systems for the 21st Century, in: Proceedings of the Irrigation and Drainage Division Special Conference. American Society of Civil Engineers, New York, New York, pp. 25–31.
Gilliam, J. w., Baker, J. l., Reddy, K. r., 1999. Water Quality Effects of Drainage in Humid Regions, in: Agricultural Drainage. John Wiley & Sons, Ltd, pp. 801–830. https://doi.org/10.2134/agronmonogr38.c24
Grazhdani, S., Jacquin, F., Sulçe, S., 1996. Effect of subsurface drainage on nutrient pollution of surface waters in south eastern Albania. Science of The Total Environment 191, 15–21. https://doi.org/10.1016/0048-9697(96)05168-6
Istok, J.D., Boersma, L., Kling, G.F., 1985. Subsurface drainage: An erosion control practice for Western Oregon (No. 729), Special report. Agricultural Experiment Station, Oregon State University, Cornvallis.
Lilja, H., Hyväluoma, J., Puustinen, M., Uusi-Kämppä, J., Turtola, E., 2017a. Evaluation of RUSLE2015 erosion model for boreal conditions. Geoderma Regional 10, 77–84. https://doi.org/10.1016/j.geodrs.2017.05.003
Lilja, H., Puustinen, M., Turtola, E., Hyväluoma, J., 2017b. Suomen peltojen karttapohjainen eroosioluokitus (Map-based classificication of erosion in agricultural lands of Finland ). Natural Resources Institute Finland (Luke) 36.
Maalim, F.K., Melesse, A.M., 2013. Modelling the impacts of subsurface drainage on surface runoff and sediment yield in the Le Sueur Watershed, Minnesota, USA. Hydrological Sciences Journal 58, 570–586. https://doi.org/10.1080/02626667.2013.774088
Øygarden, L., Kværner, J., Jenssen, P.D., 1997. Soil erosion via preferential flow to drainage systems in clay soils. Geoderma 76, 65–86. https://doi.org/10.1016/S0016-7061(96)00099-7
Panagos, P., Ballabio, C., Borrelli, P., Meusburger, K., Klik, A., Rousseva, S., Tadić, M.P., Michaelides, S., Hrabalíková, M., Olsen, P., Aalto, J., Lakatos, M., Rymszewicz, A., Dumitrescu, A., Beguería, S., Alewell, C., 2015. Rainfall erosivity in Europe. Science of The Total Environment 511, 801–814. https://doi.org/10.1016/j.scitotenv.2015.01.008
Renard, K.G., Foster, G.R., Weesies, G.A., McCool, D.K., Yoder, D.C., 1997. Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE). Agricultural Handbook 703. US Department of Agriculture, Washington, DC, pp. 404.
Skaggs, R.W., Nassehzadeh-Tabrizi, A., Foster, G.R., 1982. Subsurface drainage effects on erosion. Journal of Soil and Water Conservation 37, 167–172.
Turtola, E., Alakukku, L., Uusitalo, R., 2007. Surface runoff, subsurface drainflow and soil erosion as affected by tillage in a clayey Finnish soil. AFSci 16, 332–351. https://doi.org/10.2137/145960607784125429
Turtola, E., Paajanen, A., 1995. Influence of improved subsurface drainage on phosphorus losses and nitrogen leaching from a heavy clay soil. Agricultural Water Management 28, 295–310. https://doi.org/10.1016/0378-3774(95)01180-3
Turunen, M., Warsta, L., Paasonen-Kivekäs, M., Koivusalo, H., 2017. Computational assessment of sediment balance and suspended sediment transport pathways in subsurface drained clayey soils. Soil and Tillage Research 174, 58–69. https://doi.org/10.1016/j.still.2017.06.002
USDA, 2013. Revised Universal Soil Loss Equation Version 2 (RUSLE2), Science documentation. USDA-Agricultural Research Service, Washington, D.C.
Uusitalo, R., Turtola, E., Kauppila, T., Lilja, T., 2001. Particulate Phosphorus and Sediment in Surface Runoff and Drainflow from Clayey Soils. Journal of Environmental Quality 30, 589–595. https://doi.org/10.2134/jeq2001.302589x
Warsta, L., Taskinen, A., Koivusalo, H., Paasonen-Kivekäs, M., Karvonen, T., 2013. Modelling soil erosion in a clayey, subsurface-drained agricultural field with a three-dimensional FLUSH model. Journal of Hydrology 498, 132–143. https://doi.org/10.1016/j.jhydrol.2013.06.020
Warsta, L., Taskinen, A., Paasonen-Kivekäs, M., Karvonen, T., Koivusalo, H., 2014. Spatially distributed simulation of water balance and sediment transport in an agricultural field. Soil and Tillage Research 143, 26–37. https://doi.org/10.1016/j.still.2014.05.008
Citation: https://doi.org/10.5194/hess-2021-457-AC1
-
AC1: 'Reply on EC1', Timo Räsänen, 10 Dec 2021
-
RC1: 'Comment on hess-2021-457', Anonymous Referee #1, 25 Oct 2021
This is a very interesting paper addressing soil erosion in a Nordic country. The interesting aspects are the modified approaches of a well known model and the calibration of different factors. This manuscript may help in better spatial planning and better decision making in agricultural sector. There are some issues that can be improved. I would suggest a moderate revision.
Abstract. L12-13: In terms distances??? Please correct this sentence.
Introduction
You have used some abbreviations which are not appropriate in many parts of the manuscript. E.g. L38 Fig.s…., VegeTab.s (l.218).
L41: it is not only the transfer of phosphorus and nutrients but also the transfer of heavy metals. Please add a sentence there with a proper reference.
L70: “was” ? better to put in present. In L72: You can say that the objective of this study is addressed by 1)…………
L89-91: Your reference is always the Fig.1. Please put (B, Fig.1).
L96: The results were analysed spatially (you do not need this sentence). I tis obvious.
For Equations 2 and 3 you refer to the relevant publication. However, please be more specific by providing the reference to the LANDUM model which estimates them.
Section 2.2. what is the difference between your high resolution LS factor (2m) and the European one (at 25m)?
In table 1, please add a column with the Spatial resolution of each dataset.
Somewhere in section 2, please provide a map with the Agricultural land of Finland, with a zoom also in the location of the seven monitoring sites, etc. Maybe can you include also the boarders of the 14 selected basins?
In the paragraph 175-185: there is a fourth reason which does not allow to compare RUSLE with sediment data. Sediments are the results of many processes: gully, wind erosion, harvest erosion, landslides (not only sheet and rill erosion as RUSLE predicts).
L214: attention 2,34 should be 2.34
In section 2.6: You shoud be more specific about Emax, Emin, Ei. What are they? How they are calculated?
In figure 4, important to see also the C-factor
Figure 6a: legend. The Field area is misleading. Please use the 6a) description in the legend. The same applies for 6d. not simly EMI but what is Erosion Management Index.
The last paragraph of the conclusion is very generic. I would expect something relevant to your findings.
Finally, it will be excellent to know the most effective practices to reduce erosion in Finland.
Section 4.1 can be renamed (earlier is not an appropriate term)
Citation: https://doi.org/10.5194/hess-2021-457-RC1 -
AC2: 'Reply on RC1', Timo Räsänen, 10 Dec 2021
Dear Referee #1,
Thank you for your thoughtful and constructive comments (RC) – they helped to improve the manuscript. Altogether, the comments from the editor and the two referees prompted major revisions, and we have revised the manuscript. The major revisions are:
- The country scale results on actual erosion and erosion management index were removed from the revised manuscript due to limitations in the used C-factor data, which was pointed out by Pedro Batista (Referee #2). Correction of the C factor data requires considerable research work and decision was made to leave it as future work to be published in another publication.
- The LS factor and consequent erosion data was recalculated to account for field borders.
- The results on RUSLE evaluation, erosion susceptibility and susceptibility near water bodies were slightly restructured to accommodate the changes caused by removal of the two parts in mentioned in the previous point.
- New sensitivity analyses were added that provide estimates on
- the propagation of uncertainties from RUSLE factors to erosion estimates
- effects of location specific cropping and management practices and temporal rainfall erosivity distribution on C- factor values and the consequent erosion estimates
- The terms “potential erosion risk” and “actual erosion risk” were replaced by terms, “erosion susceptibility” and “actual erosion” to avoid the misuse of the term “risk”.
Thus, the new findings of the revised manuscript are:
- New high-resolution (two-meter) country scale erosion susceptibility data for Finland
- New evaluation of RUSLE and its performance in boreal conditions, which considers also different spatial scales and issues relate to upscaling from field parcel to larger spatial scales
- Improved scientific understanding of agricultural erosion and its spatial distribution
These findings provide new opportunities for research and erosion management. In the following we provide our comments and responses (AC) to the referee comments (RC) point by point.
Comments:
RC1: This is a very interesting paper addressing soil erosion in a Nordic country. The interesting aspects are the modified approaches of a well-known model and the calibration of different factors. This manuscript may help in better spatial planning and better decision making in agricultural sector. There are some issues that can be improved. I would suggest a moderate revision.
AC1: Thank you for your positive comments and encouragement.RC2: Abstract. L12-13: In terms distances??? Please correct this sentence.
AC2: We have revised the sentence as follows: ” The developed data revealed spatially varying erosion patterns, which has implications on erosion management. For example, high erosion rates were found in intensive agricultural areas, and in several areas high erosion rates were concentrated near water bodies, where the eroded soil is more likely to cause negative off-site impacts.”RC3: You have used some abbreviations which are not appropriate in many parts of the manuscript. E.g. L38 Fig.s…., VegeTab.s (l.218).
AC3: We will perform a proof reading of the revised manuscript. Thank you for noting these.RC4: L41: it is not only the transfer of phosphorus and nutrients but also the transfer of heavy metals. Please add a sentence there with a proper reference.
AC4: We have mentioned heavy metals in the introduction together with appropriate reference (e.g. Shi et al., 2018) in the revised manuscript.RC5: L70: “was” ? better to put in present. In L72: You can say that the objective of this study is addressed by 1)…………
AC5: Corrected as suggested in the revised manuscript.RC6: L89-91: Your reference is always the Fig.1. Please put (B, Fig.1).
AC6: The whole section was revised, and this problem does not occur in the revised manuscript.RC7: L96: The results were analysed spatially (you do not need this sentence). I tis obvious.
AC7: Agreed. The sentence was removed from the revised manuscript.RC8: For Equations 2 and 3 you refer to the relevant publication. However, please be more specific by providing the reference to the LANDUM model which estimates them.
RC8: The manuscript has been revised regarding the C factor and we use now the original definition from Renard et al. (1997).RC9: Section 2.2. what is the difference between your high resolution LS factor (2m) and the European one (at 25m)?
AC9: We assume that you refer to the 25 meter resolution EU DEM, and to the European LS data calculated from the EU DEM by Panagos et al. (2015). We used a national a two-meter resolution LiDAR-based DEM (National Land Survey of Finland, 2020) and used the same calculation tool (Conrad, 2013) and method (Desmet and Govers, 1996) as Panagos et al. (2015). It is known that the resolution of the DEM influences the calculation of the LS and the erosion estimates (e.g. Chen et al., 2018; Beeson et al., 2014). For example, coarser resolution DEMs can result in larger estimates of L values and finer DEM’s on larger estimates of S values (Fu et al., 2015). To our knowledge there are no guidelines on how to account for the effect of DEM resolution, which consequently adds uncertainty in RUSLE estimates. We are considering to add the following sentence to the discussion section of the revised manuscript: ” According to an analysis conducted in Finland (Lilja et al. 2017b), the use of two-meter resolution DEM with a modified LS calculation method of Desmet and Govers (1996) resulted in 37-43% larger erosion estimates compared to the use of 25 m resolution DEM, but it is not clear how the modification of the LS calculation method affected these estimates compared to the original approach (Desmet and Govers, 1996) and whether the field parcels were considered hydrologically isolated in the calculation of the LS factor. This comparison is, however, available only in Finnish language, and it is not published in a peer-reviewed publication.RC10: In table 1, please add a column with the Spatial resolution of each dataset.
AC10: The Tab. 1 was removed from the revised manuscript due to major revisions resulting from Pedro Batista’s (Referee #2) comments. In the revised manuscript the summary of C and P factor data are not needed, and therefore, the need for data summary table is also reduced.RC11: Somewhere in section 2, please provide a map with the Agricultural land of Finland, with a zoom also in the location of the seven monitoring sites, etc. Maybe can you include also the boarders of the 14 selected basins?
AC11: A new map is provided in the revised manuscript that shows agricultural areas, the seven field sites, the small catchments, and large river basins.References
Beeson, P.C., Sadeghi, A.M., Lang, M.W., Tomer, M.D., Daughtry, C.S.T., 2014. Sediment Delivery Estimates in Water Quality Models Altered by Resolution and Source of Topographic Data. Journal of Environmental Quality 43, 26–36. https://doi.org/10.2134/jeq2012.0148
Chen, W., Li, D.-H., Yang, K.-J., Tsai, F., Seeboonruang, U., 2018. Identifying and comparing relatively high soil erosion sites with four DEMs. Ecological Engineering 120, 449–463. https://doi.org/10.1016/j.ecoleng.2018.06.025
Conrad, O., 2013. Module LS-Factor, Field Based [WWW Document]. SAGA-GIS Module Library Documentation (v2.1.4). URL http://www.saga-gis.org/saga_tool_doc/2.1.4/ta_hydrology_25.html (accessed 5.29.20).
Desmet, P.J.J., Govers, G., 1996. A GIS procedure for automatically calculating the USLE LS factor on topographically complex landscape units. Journal of Soil and Water Conservation 51, 427–433.
Fu, S., Cao, L., Liu, B., Wu, Z., Savabi, M.R., 2015. Effects of DEM grid size on predicting soil loss from small watersheds in China. Environ Earth Sci 73, 2141–2151. https://doi.org/10.1007/s12665-014-3564-3
Lilja, H., Puustinen, M., Turtola, E., Hyväluoma, J., 2017. Suomen peltojen karttapohjainen eroosioluokitus (Map-based classificication of erosion in agricultural lands of Finland ). Natural Resources Institute Finland (Luke) 36.
National Land Survey of Finland, 2020. Elevation model 2 m [WWW Document]. URL https://www.maanmittauslaitos.fi/en/maps-and-spatial-data/expert-users/product-descriptions/elevation-model-2-m (accessed 5.29.20).
Panagos, P., Borrelli, P., Meusburger, K., 2015. A New European Slope Length and Steepness Factor (LS-Factor) for Modeling Soil Erosion by Water. Geosciences 5, 117–126. https://doi.org/10.3390/geosciences5020117
Renard, K.G., Foster, G.R., Weesies, G.A., McCool, D.K., Yoder, D.C., 1997. Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE). Agricultural Handbook 703. US Department of Agriculture, Washington, DC, pp. 404.
Shi, T., Ma, J., Wu, X., Ju, T., Lin, X., Zhang, Y., Li, X., Gong, Y., Hou, H., Zhao, L., Wu, F., 2018. Inventories of heavy metal inputs and outputs to and from agricultural soils: A review. Ecotoxicology and Environmental Safety 164, 118–124. https://doi.org/10.1016/j.ecoenv.2018.08.016
Citation: https://doi.org/10.5194/hess-2021-457-AC2
-
AC2: 'Reply on RC1', Timo Räsänen, 10 Dec 2021
-
RC2: 'Comment on hess-2021-457', Pedro Batista, 01 Nov 2021
Dear authors,
I enjoyed reading your manuscript “Improving the agricultural erosion management in Finland through high-resolution data”. I appreciated the use of the high-resolution DEM and the field-parcel data for model parameterisation. However, there are some issues, which, in my opinion, need to be addressed before the manuscript can be considered for publication.
First, you state that the goal of your study is to produce ‘erosion risk data for agricultural lands in Finland’. However, what you produced are soil loss maps, which do not translate into erosion risk assessments. I understand this is a common misconception in erosion modelling research, but the manuscript should not add to the confusion. For instance, you assume risk is the modelled erosion value for a given location, without stating the assets at risk, what negative consequences erosion could bring to these assets, and what are the probabilities of these consequences occurring.
Second, there are serious problems with methodology used for calibrating the C factor for the RUSLE. From what I understood, your approach considerably deviates from the original USLE or RUSLE methodology, neglects the influence of rainfall erosivity and crop stages, and relies solely on a deterministic parameter optimisation procedure, without considering the uncertainty in the input data and the calibration methodology. These issues are described in detail below, and please correct me if I am wrong.
Third, there is no uncertainty analysis. Although you dedicate a large amount of text to pointing out the uncertainties in the model, you did not attempt to quantify them. In my opinion, if you wish public policy to be guided by your results, you should at least provide a forward error assessment to quantify the uncertainty associated to the model parameterisation. For instance, you state that your study provides a generalisation of the effects of management practices on erosion. Do you believe it is sound to provide such a generalisation, based on a limited number of observations, without a measure of uncertainty?
As pointed out by Christian Stamm, I also have some concerns regarding how tile drainage and field borders were incorporated (or not) into the modelling.
I believe you can address these issues in a new manuscript, but that would require a different submission, in my opinion. I hope these comments are at all useful and I wish the best of luck with your research.
All the best,
Pedro Batista
Detailed comments
L26: I did not understand what you meant with “the key process causing erosion is hydrological”.
L34: Do you mean erosion is affected by the short growing period?
L36-37: I think superscripts are missing here.
L64: In my opinion, acquiring spatial data for parameterisation and calibration is more of a challenge than computational power.
L68: Could you also state some of the limitations of the USLE-family models here? For instance, you cite our paper to corroborate the ability of the USLE to simulate annual loads – I imagine you mean at the erosion plot scale. However, our review also shows how spatially distributed erosion rates compare poorly to independent measurements.
L81: Wouldn’t the RUSLE require longer time series for estimating the R factor?
L82-87: How are you defining risk? Can risk be expressed in mass area-1 time-1? It seems to me you are calculating erosion rates, which of course can be a threat to multiple assets (e.g. the soil itself, downstream infrastructure, etc). However, threats, assets, and potential consequences need to be identified in order to produce an actual risk assessment. This is a common misconception in model-based erosion risk assessments, in my opinion.
L96-97: With a 2 m resolution, couldn’t you assess risk at field-block scale?
L112: These are not the sub-factors defined in the RUSLE (see Renard et al. 1997), correct? If so, please make it clearer you are using an adaptation.
L128-130: What is the ICECREAM model? In general I could not understand how the K factor was calculated. Are you taking single K factor values for mapping units in a soil map? This can introduce large errors to model outputs (see Van Rompaey and Govers, 2002).
L134-135: This is an interesting point about the sink filling. Have you made any tests with and without it?
L150: I have some questions about this calibration. Usually, we would calculate soil loss ratios for different crop management systems/crop rotations by use of erosion plot data and/or plant, soil, and residue measurements. These ratios would then be weighted with rainfall erosivity to calculate the C factors for specific locations. However, here you are using an optimisation approach – could you explain why? Moreover, did you perform any kind of split-off test, in which part of the data is used for calibration and another for testing? Would you agree that parameter calibration is necessarily conditional and that different parameter values can produce acceptable model responses? If not, why? If so, shouldn’t you use a range of behavioural parameter values to estimate the uncertainty in your model outputs? Moreover, it seems like you calibrated the sub-factor Ccrop, not the C factor. Did I understand this correctly?
L175-176: I agree model evaluation is difficult. However, particularly when models are being used to influence public policy or to guide decision-making, model testing is a necessary step. Our point in the paper you are citing was not to say it is okay not to evaluate models because it is difficult and rarely done. Instead, we wanted to incentivise the erosion modelling community to improve how we perform model testing and uncertainty analysis.
L180-187: I agree, and I appreciate how you are open about the limitations of your testing data.
L210-221: Why was rainfall erosivity not considered in any of the C factor calculations? This is crucial in USLE-type models, since erosion rates will vary largely for a same crop if the time in which the soil is exposed coincides with the time in which there is greater rainfall erosivitiy.
L235-242: Have you considered using a sediment routing model, which would allow you to deal with these issues?
L267: “Reasonable, with limitations” seems subjective. I suggest sticking to the numbers at this point of the results.
L270: The mean error shows you if the model is biased or not, but this metric is affected by cancelation. Could you also provide the RMSE, NSE, or such? Moreover, are you comparing the average soil loss from the entire period or individual annual losses?
L287-290: Would you not expect the (mean?) estimated erosion rates per catchment to deviate from the TSS measurements anyhow? These are two different things. I agree that if there is a correlation between them, there is an indication that the model might be consistently identifying the catchments where there is greater sediment production. However, since the RUSLE does not quantify sediment delivery to water courses and only represents rill and interrill erosion, do you believe there is any chance this correlation does not amount to causation?
L290-298: Do you believe mean values per catchment are good descriptors of your data here?
L313-315: Do you mean highly erodible soils?
Fig.4: By looking at your R factor map, I imagine the spatial variability of rainfall erosivity to have a large influence on the C factors for croplands.
L324-325: There seems to be a concern here regarding connectivity and sediment delivery to water bodies. Do you think the RUSLE is an appropriate model for looking at these issues?
L390-395: If you are using the same data for calibration and testing, considering average annual soil losses, and with a +- 50% limit of acceptability, how rigorous do you believe your model evaluation procedure to be? Moreover, this methodology for incorporating the effects of sub-surface drainage into the C factor needs more explanation, in my opinion. Is (tile?) drainage affecting rill and interrill erosion or the sediment export from agricultural plots?
L400-420: Considering all these uncertainties, do you believe it is acceptable to provide a single, deterministic, model output?
L403-404: Yes, quantifying uncertainty is challenging, but shouldn’t we step up to the challenge? We have published a simple, open access, code to propagate the errors in the RUSLE (Batista et al., 2021b). Similar code using high-resolution data is available in Batista et al. (2021a). If these approaches are too computationally intensive considering the scale of the model application, I suggest using the sub-catchments where you have measured TSS data as case studies (see Tetzlaff et al., 2013).
L426-435: With a 2 m resolution, isn’t there any other way you could incorporate the influence of the buffer zones in your model?
L436: How do you define reasonable quality?
L436-441: I am sorry, but I disagree that the calibration of the C factor provided a good basis for modelling. The main reasons being:
- The inclusion of sub-surface drainage into the soil loss estimates used for calibration is not well justified;
- The methodology for calculating the C factor is not in accordance with the RUSLE handbook (Renard et al., 1997);
- Rainfall erosivity is not considered in the C factor calculations;
- The parameter optimisation procedure does not consider the uncertainty in the data, and a single parameter value per crop type is used.
References:
Batista, P., Fiener, P., Scheper, S. and Alewell, C.: A conceptual model-based sediment connectivity assessment for patchy agricultural catchments, Hydrol. Earth Syst. Sci. Discuss., (May), 1–32, doi:10.5194/hess-2021-231, 2021a.
Batista, P. V. G., Laceby, J. P., Davies, J., Carvalho, T. S., Tassinari, D., Silva, M. L. N., Curi, N. and Quinton, J. N.: A framework for testing large-scale distributed soil erosion and sediment delivery modelsâ¯: Dealing with uncertainty in models and the observational data, Environ. Model. Softw., 137, doi:10.1016/j.envsoft.2021.104961, 2021b.
Renard, K. ., Foster, G. R., Weesies, G. A., McCool, D. K. and Yoder, D. C.: Predicting Soil Erosion by Water: A Guide to Conservation Planning With the Revised Universal Soil Loss Equation (RUSLE), 1997.
Van Rompaey, A. J. J. and Govers, G.: Data quality and model complexity for regional scale soil erosion prediction, Int. J. Geogr. Inf. Sci., 16(7), 663–680, doi:10.1080/13658810210148561, 2002.
Tetzlaff, B., Friedrich, K., Vorderbrügge, T., Vereecken, H. and Wendland, F.: Distributed modelling of mean annual soil erosion and sediment delivery rates to surface waters, Catena, 102, 13–20, doi:10.1016/j.catena.2011.08.001, 2013.
Citation: https://doi.org/10.5194/hess-2021-457-RC2 -
AC3: 'Reply on RC2', Timo Räsänen, 10 Dec 2021
Dear Pedro Batista,
Thank you for your thoughtful and constructive comments (EC). They made us to rethink our analyses and they provided a great learning opportunity. Most of all, they help to improve the manuscript. Altogether, the comments from the editor and the two referees prompted major revisions, and we have revised the manuscript. The major revisions are:
- The country scale results on actual erosion and erosion management index were removed from the revised manuscript due to limitations in the used C-factor data, which was pointed out by Pedro Batista (Referee #2). Correction of the C factor data requires considerable research work and decision was made to leave it as future work to be published in another publication.
- The LS factor and consequent erosion data were recalculated to account for field borders.
- The results on RUSLE evaluation, erosion susceptibility and susceptibility near water bodies were slightly restructured to accommodate the changes caused by removal of the two parts in mentioned in the previous point.
- New sensitivity analyses were added that provide estimates on
- the propagation of uncertainties from RUSLE factors to erosion estimates
- effects of location specific cropping and management practices and temporal rainfall erosivity distribution on C- factor values and the consequent erosion estimates
- The terms “potential erosion risk” and “actual erosion risk” of the original manuscript were replaced by new terms, “erosion susceptibility” and “actual erosion” as proposed by the referee to avoid the misuse of the term “risk”.
Thus, the new findings of the revised manuscript are:
- New high-resolution (two-meter) country scale erosion susceptibility data for all agricultural land in Finland
- New evaluation of RUSLE and its performance in boreal conditions, which considers also different spatial scales and issues relate to upscaling from field parcel to larger spatial scales
- Improved scientific understanding of agricultural erosion and its spatial distribution
These findings provide new opportunities for research and erosion management. In the following we provide our comments and responses (AC) to the referee comments (RC) point by point.
Comments:
RC1: I enjoyed reading your manuscript “Improving the agricultural erosion management in Finland through high-resolution data”. I appreciated the use of the high-resolution DEM and the field-parcel data for model parameterisation. However, there are some issues, which, in my opinion, need to be addressed before the manuscript can be considered for publication.
AC1: Happy to hear that you found our manuscript interesting. Thank you.RC2: First, you state that the goal of your study is to produce ‘erosion risk data for agricultural lands in Finland’. However, what you produced are soil loss maps, which do not translate into erosion risk assessments. I understand this is a common misconception in erosion modelling research, but the manuscript should not add to the confusion. For instance, you assume risk is the modelled erosion value for a given location, without stating the assets at risk, what negative consequences erosion could bring to these assets, and what are the probabilities of these consequences occurring.
AC2: We agree. This is an unfortunate blunder in terminology from our side. We have removed the word “risk” from the revised manuscript and used terms “erosion susceptibility” and “actual erosion”.RC3: Second, there are serious problems with methodology used for calibrating the C factor for the RUSLE. From what I understood, your approach considerably deviates from the original USLE or RUSLE methodology, neglects the influence of rainfall erosivity and crop stages, and relies solely on a deterministic parameter optimisation procedure, without considering the uncertainty in the input data and the calibration methodology. These issues are described in detail below, and please correct me if I am wrong. Third, there is no uncertainty analysis. Although you dedicate a large amount of text to pointing out the uncertainties in the model, you did not attempt to quantify them. In my opinion, if you wish public policy to be guided by your results, you should at least provide a forward error assessment to quantify the uncertainty associated to the model parameterisation. For instance, you state that your study provides a generalisation of the effects of management practices on erosion. Do you believe it is sound to provide such a generalisation, based on a limited number of observations, without a measure of uncertainty?
AC3: We agree with the comments here, and we have revised to the manuscript accordingly. In the revised manuscript we use the original definition of C from Renard et al. (1997), perform sensitivity analysis of forward propagation of uncertainties, and make a preliminary analysis on the effects of location specific cropping and management practices and temporal distribution of rainfall erosivity on C factor values and erosion estimates. We exclude the country scale estimation of actual erosion (A=RKLSCP) from the revised manuscript given the spatial variability C and aim to publish it later in the future, as explained in the beginning of our comment. These revisions are explained also in more detail in the responses to your specific comments.RC4: As pointed out by Christian Stamm, I also have some concerns regarding how tile drainage and field borders were incorporated (or not) into the modelling.
AC4: We have justified the incorporation of subsurface drainage in a following way in the revised manuscript. According to Renard et al. (1997) subsurface drainage is considered in the P factor. Bengtson and Sabbagh (1990) suggested an average P factor value of 0.6, which was also recognised by Renard et al. (1997). However, the research on how the subsurface drainage should be considered in the RUSLE is limited, and to our understanding there is no commonly accepted approach for this. Except, RUSLE2 (USDA, 2013) incorporates a method for this, which considers the changes in K factor due to the subsurface drainage, but in our case the data was too limited to consider this. Therefore, we chose to follow the original suggestions by Renard et al. (1997) and Bengtson and Sabbagh (1990). According to our literature review the subsurface drainage reduces erosion 8-90% (on average 38%) (Bengtson et al., 1988, 1984; Bengtson and Sabbagh, 1990; Formanek et al., 1987; Gilliam et al., 1999; Grazhdani et al., 1996; Istok et al., 1985; Maalim and Melesse, 2013; Skaggs et al., 1982), which results to average P value of 0.62. We used the P value of P 0.6 as suggested and used earlier in Finland by Lilja et al. (2017a). Also, the research in Finland showed that substituting of old drainage pipes and drain trenches with new ones reduced erosion up to 15% on a clay soil (Turtola and Paajanen, 1995). The use of sum of surface and subsurface sediment is justified also by research. In Finland, it is observed that up to 50-90 % of the erosion loading at clay soils occurs via subsurface drainage (Finnish Environment Institute, 2019; Turtola et al., 2007; Turunen et al., 2017; Warsta et al., 2014, 2013) and that in clay soils erosion material in subsurface drainage flow seems to originate mainly from the surface soil (Uusitalo et al., 2001). In this Finnish research the origin of the erosion material was determined by an analysis of Cesium-137 contents of soil layers and eroded soil material in the subsurface flow. The modelling studies in Finland also support this finding (Turunen et al., 2017). A study from Norway also reports that soil material in the drain flow originated most likely from the plough layer and, and the soil material was transported to subsurface drains via cracks and macropores in the soil (Øygarden et al., 1997). However, we do acknowledge that the consideration of subsurface drainage in the P factor lumps a complex and poorly understood process into single value and has therefore limitations and is a considerable source of uncertainty. In the revised manuscript we perform a sensitivity analysis which also includes the uncertainty in the P value. This justification was added to the revised manuscript. Note also that our modelling approach tests the above assumptions against empirical data and the resulting performance can be considered reasonable.Regarding the field borders, the LS factor was recalculated in the revised manuscript to account for the field parcel borders. We explain this in revised manuscript in a following way: “The LS factor was calculated from a two-meter resolution LiDAR-based digital elevation model (DEM) of Finland (National Land Survey of Finland, 2020), and by using the SAGA-GIS Module LS Factor (Conrad, 2013) and the method of Desmet and Govers (1996 ) with default settings. The LS calculation was performed in two-meter resolution for agricultural lands in 301 hydrological units that consisted of river basins, sub-basin groups (Finnish Environment Institute, 2010) and groups of islands (Fig. S1). The agricultural lands were defined according to the field parcel data from Finnish Food Authority, which contains over one million vectorized field parcels and it accounts almost all agricultural land in Finland. The use of vectorized field parcels treated each field parcel as an isolated hydrological unit (in terms of overland flow) in the LS calculation to account for the effects of varying landcover on surface runoff, as recommended by Desmet and Govers (1996). The approach is considered adequate as in Finland the fields are well drained and commonly surrounded by ditches, which advocates for the hydrological isolation of the field parcels. However, adjacent field parcels that shared the same parcel border were treated in the calculation as a single field parcel, since it is common that uniform field areas are divided into separate field parcels, as reported in the data of Finnish Food Authority, for annual cropping and management purposes.” The evaluation of RUSLE at the seven field sites was not affected by the recalculation of the LS. Their LS factors were originally calculated using field borders.
RC5: L26: I did not understand what you meant with “the key process causing erosion is hydrological”.
AC5: This is just poor language from our side. The purpose was to say that soil erosion by water is a hydrologically driven process. We have revised this in the revised manuscript.RC6: L34: Do you mean erosion is affected by the short growing period?
AC6: The word “process” was removed. Thank you for noting.RC7: L36-37: I think superscripts are missing here.
AC7: Corrected in the revised manuscript.RC8: L64: In my opinion, acquiring spatial data for parameterisation and calibration is more of a challenge than computational power.
AC8: We have added the issue of parameterisation into the discussion.RC9: L68: Could you also state some of the limitations of the USLE-family models here? For instance, you cite our paper to corroborate the ability of the USLE to simulate annual loads – I imagine you mean at the erosion plot scale. However, our review also shows how spatially distributed erosion rates compare poorly to independent measurements.
AC9: We have clarified that the sentence refers to the plot scale and we added the following sentences to the revised manuscript: ”The USLE type models are also commonly used over large spatial scales with varying spatial resolutions, but the success of these applications varies more than on plot scale. It is however, observed that the spatially distributed approaches are often able to rank erosion-prone fields if high quality data are available for parameterization, but the actual erosion rates compare poorly to independent measurements (Batista et al., 2019)”RC10: L81: Wouldn’t the RUSLE require longer time series for estimating the R factor?
AC10: The short data period can indeed cause uncertainties in the R factor, since in short timeseries for example exceptional events can have a greater effect on the average R estimates. This may lead to uncertainties in the magnitude and the spatial distribution of the R. Following sentence was added to the discussion section of the revised manuscript: “The R factor was based on European data, which in the case of Finland was based on measurement data from 60 stations from the period 2007-2013. This data period is relatively short and may therefore cause inaccuracies in the average level and spatial patterns of rainfall erosivity.” However, the used data is currently the best available, and improvements can be made by estimating new R factor data, but this is beyond the scope of the manuscript.RC11: L82-87: How are you defining risk? Can risk be expressed in mass area-1 time-1? It seems to me you are calculating erosion rates, which of course can be a threat to multiple assets (e.g. the soil itself, downstream infrastructure, etc). However, threats, assets, and potential consequences need to be identified in order to produce an actual risk assessment. This is a common misconception in model-based erosion risk assessments, in my opinion.
AC11: We have removed the word “risk” in the revised manuscript and as explained earlier. We will use terms “erosion susceptibility” and “actual erosion”.RC12: L96-97: With a 2 m resolution, couldn’t you assess risk at field-block scale?
AC12: Yes. The developed two-meter resolution erosion data, which will be publicly available, allows field parcel and within-field parcel scale analyses. In the manuscript we however, considered that it is more useful to introduce the developed data and the spatial patterns of erosion primarily on a broader scale, since the development of country scale data is the key novelty of the research.RC13: L112: These are not the sub-factors defined in the RUSLE (see Renard et al. 1997), correct? If so, please make it clearer you are using an adaptation.
AC13: In the revised manuscript we use the original definition of C by Renard et al. (1997).RC14: L128-130: What is the ICECREAM model? In general I could not understand how the K factor was calculated. Are you taking single K factor values for mapping units in a soil map? This can introduce large errors to model outputs (see Van Rompaey and Govers, 2002).
AC14: ICECREAM is a version of CREAMS model that has been adapted to boreal winter conditions (Rekolainen and Posch, 1993). The spatial K factor data was not calculated in this study, and it was developed earlier by Lilja et al. (2017a, 2017b) , by using the K values that have been estimated and evaluated in national research (e.g., Rekolainen and Posch, 1993; Bärlund and Tattari, 2001; Bärlund et al., 2009; Huttunen et al., 2016; Rankinen et al., 2001). The soil class specific K factor values are shown in the supplement. The Van Rompaey and Govers (2002) was very interesting reading, and we have taken a note on this and also included it in the discussion section of the manuscript. However, we did not follow their idea of simplifying K factor data, given that their findings are application and location specific and therefore they don’t recommend direct extrapolation of their findings. However, as a sensitivity analysis such simplification would be an interesting, but we decided to leave it for future work. Also, the simplification of K factor data would have resulted in loss of known soil regions in Finland that affect the regional distribution of erosion. These soil regions include, for example clay soils areas in Southern Finland, and the highly erodible soils in the river valleys in the western coast. We have clarified the nature of the K factor data in the revised manuscript.RC15: L134-135: This is an interesting point about the sink filling. Have you made any tests with and without it?
AC15: Unfortunately, we have not made tests on this, and it is a potential future work.RC16: L150: I have some questions about this calibration. Usually, we would calculate soil loss ratios for different crop management systems/crop rotations by use of erosion plot data and/or plant, soil, and residue measurements. These ratios would then be weighted with rainfall erosivity to calculate the C factors for specific locations. However, here you are using an optimisation approach – could you explain why? Moreover, did you perform any kind of split-off test, in which part of the data is used for calibration and another for testing? Would you agree that parameter calibration is necessarily conditional and that different parameter values can produce acceptable model responses? If not, why? If so, shouldn’t you use a range of behavioural parameter values to estimate the uncertainty in your model outputs? Moreover, it seems like you calibrated the sub-factor Ccrop, not the C factor. Did I understand this correctly?
AC16: The optimisation approach was used simply because of the lack of measurement data that would allow calculation of soil loss ratios. The optimisation was intended to provide rough estimates of the general magnitude of the C factors, which can be improved later by future measurements. Note that the approach has been applied previously by Lilja et al. (2017a). The erosion measurement data was also limited, and it did not allow splitting of the data to calibration and validation sets. We are also very aware that different parameter combinations can provide similar outcomes, which is common feature in modelling. However, we did our best to parameterise the RUSLE with realistic factor values, given the data limitations. In the revised manuscript the C factor is defined according to the Renard et al. (1997), and the optimisation was done for the C factor of each cropping and management practice separately. Subfactor thinking of Panagos et al. (2015) is not included in the manuscript anymore.RC17: L175-176: I agree model evaluation is difficult. However, particularly when models are being used to influence public policy or to guide decision-making, model testing is a necessary step. Our point in the paper you are citing was not to say it is okay not to evaluate models because it is difficult and rarely done. Instead, we wanted to incentivise the erosion modelling community to improve how we perform model testing and uncertainty analysis.
AC17: We have removed this sentence to avoid the misinterpretation.RC18: L180-187: I agree, and I appreciate how you are open about the limitations of your testing data.
AC18: In this case, we considered it is important to be very specific on the nature of the analysis to avoid misinterpretations.RC19: L210-221: Why was rainfall erosivity not considered in any of the C factor calculations? This is crucial in USLE-type models, since erosion rates will vary largely for a same crop if the time in which the soil is exposed coincides with the time in which there is greater rainfall erosivitiy.
AC19: In the revised manuscript we have made an analysis of the effects of location specific cropping and management practices and temporal distribution on rainfall erosivity on the C factor values and erosion estimates using the seven field sites as analysis locations. This was done by estimating the average annual C values at the seven field sites on monthly basis according to Renard (1997). The monthly R data was taken from Ballabio et al. (2017) and two sets of crop sequence specific soil loss ratio values were taken from Wischmeier and Smith (1978) as there were no measurement based soil loss ratio data available from Finland. The selected soil loss ratios were intended to parameterise cereals with normal autumn ploughing and spring cereals with reduced autumn tillage, and the soil loss ratio values were adapted to Finnish location specific cropping schedules. The main difference between the Southern, Northern and Eastern field sites is that in the two latter the spring field preparation and sowing occurs on average two weeks later. The aim of this preliminary analysis was to provide indication of the variability C values in Finland and its effect on erosion estimates, and it was not aimed to provide accurate estimates of the C factors in Finland, as described in more detail in the revised manuscript. The outcomes of the analysis provided interesting insights into the variation of C and erosion estimate by location, and to our knowledge the findings are novel in Finland.RC20: L235-242: Have you considered using a sediment routing model, which would allow you to deal with these issues?
AC20: At the current development stage of RUSLE in Finland, the inclusion of routing model is too early, but we consider it as an important future research direction.RC21: L267: “Reasonable, with limitations” seems subjective. I suggest sticking to the numbers at this point of the results.
AC21: Agreed and these remarks were removed.RC22: L270: The mean error shows you if the model is biased or not, but this metric is affected by cancelation. Could you also provide the RMSE, NSE, or such? Moreover, are you comparing the average soil loss from the entire period or individual annual losses?
AC22: We have added RMSE value to the revised manuscript.RC23: L287-290: Would you not expect the (mean?) estimated erosion rates per catchment to deviate from the TSS measurements anyhow? These are two different things. I agree that if there is a correlation between them, there is an indication that the model might be consistently identifying the catchments where there is greater sediment production. However, since the RUSLE does not quantify sediment delivery to water courses and only represents rill and interrill erosion, do you believe there is any chance this correlation does not amount to causation?
AC23: Yes. Erosion estimates of RUSLE and measured TSS from streams and rivers are different aspects of the erosion, transport, and deposition process, and represent different things. We have tried to be very open on the nature of this analysis and its limitations in the manuscript, and we do acknowledge that the relationship of correlation and causation may contain uncertainties in this case. Our thinking was that it still better to include this analysis to the manuscript than to exclude it, since the information provides some indication of the performance of spatially distributed RUSLE, given their known uncertainties (Batista et al., 2019).RC24: L290-298: Do you believe mean values per catchment are good descriptors of your data here?
AC24: We assume that you are referring to the slope values. The presented slope ranges describe how the average slope of field parcel varies by sub-catchment. We are not performing any quantitative analyses based on these values, and they are only for illustrating the differences of the two catchments. Therefore, we think average slopes are adequate for this purpose. We clarified in the revised manuscript that these values refer to the average slopes of field parcels by subcatchments.RC25: L313-315: Do you mean highly erodible soils?
AC25: Yes, thank you for noting this. Corrected in the revised manuscript.RC26: Fig.4: By looking at your R factor map, I imagine the spatial variability of rainfall erosivity to have a large influence on the C factors for croplands.
AC26: This was analysed in the revised manuscript. See author comment (AC19)RC27: L324-325: There seems to be a concern here regarding connectivity and sediment delivery to water bodies. Do you think the RUSLE is an appropriate model for looking at these issues?
AC27: RUSLE does not incorporate sediment transport, nor did we account for connectivity, but we do think that the RUSLE provides useful information on the variability of erosion near water bodies, both from the scientific and management point of view. However, RUSLE is not able to indicate which of the identified high erosion areas near water bodies will lead to sediment delivery to rivers and streams, and at this point such evaluation from the developed data requires judgement from the data user.RC28: L390-395: If you are using the same data for calibration and testing, considering average annual soil losses, and with a +- 50% limit of acceptability, how rigorous do you believe your model evaluation procedure to be?
AC28: This model evaluation was not designed in our manuscript, but it was used by Lilja et al. (2017a). Our intention was to make our results comparable within the evaluation framework of Lilja et al. (2017a). We have clarified this in the revised manuscript. Note that typically the error was lower, as explained in Section 3.2. of the manuscript.RC29: Moreover, this methodology for incorporating the effects of sub-surface drainage into the C factor needs more explanation, in my opinion. Is (tile?) drainage affecting rill and interrill erosion or the sediment export from agricultural plots?
AC29: We have justified the incorporation of subsurface drainage in the P factor in the revised manuscript in a following way. According to Renard et al. (1997) subsurface drainage is considered in the P factor. Bengtson and Sabbagh (1990) suggested an average P factor value of 0.6, which was also recognised by Renard et al. (1997). However, the research on how the subsurface drainage should be considered in the RUSLE is limited, and to our understanding there is no commonly accepted approach for this. Except, RUSLE2 (USDA, 2013) incorporates a method for this, which considers the changes in K factor due to the subsurface drainage, but in our case the data was too limited to consider this. Therefore, we chose to follow the original suggestions by Renard et al. (1997) and Bengtson and Sabbagh (1990). According to our literature review the subsurface drainage reduce erosion 8-90% and on average 38% (Bengtson et al., 1988, 1984; Bengtson and Sabbagh, 1990; Formanek et al., 1987; Gilliam et al., 1999; Grazhdani et al., 1996; Istok et al., 1985; Maalim and Melesse, 2013; Skaggs et al., 1982), which results to average P value of 0.62. We used the P value of P 0.6 as suggested and used earlier in Finland by Lilja et al. (2017a). Also, the research in Finland showed that substituting of old drainage pipes with new ones reduced erosion up to 15% on a clay soil (Turtola and Paajanen, 1995). According to the research cited above, the main mechanisms how subsurface drainage reduces rill and inter-rill erosion is via reduced surface runoff, increased soil permeability and increased crop yield.The use of sum of surface and subsurface sediment is justified also by research. In Finland, it is observed that up to 50-90 % of the erosion loading occurs via subsurface drainage (Finnish Environment Institute, 2019; Turtola et al., 2007; Turunen et al., 2017; Warsta et al., 2014, 2013) and that erosion material in subsurface drainage flow originates from the surface soil (Uusitalo et al., 2001). The origin of the erosion material was determined by an analysis of Cesium-137 contents of soil layers and eroded soil material, and the findings are in agreement with modelling results (Turunen et al., 2017). A study from Norway also reports that soil material in the drain flow originated most likely from the plough layer and, and the soil material was transported to subsurface drains via cracks and macropores in the soil (Øygarden et al., 1997). However, we do acknowledge that the consideration of subsurface drainage in the P factor lumps a complex and poorly understood process into single value and has therefore limitations and is a considerable source of uncertainty. In the revised manuscript we perform a sensitivity analysis which includes also the uncertainty in the P value.
References
Ballabio, C., Borrelli, P., Spinoni, J., Meusburger, K., Michaelides, S., Beguería, S., Klik, A., Petan, S., Janeček, M., Olsen, P., Aalto, J., Lakatos, M., Rymszewicz, A., Dumitrescu, A., Tadić, M.P., Diodato, N., Kostalova, J., Rousseva, S., Banasik, K., Alewell, C., Panagos, P., 2017. Mapping monthly rainfall erosivity in Europe. Science of The Total Environment 579, 1298–1315. https://doi.org/10.1016/j.scitotenv.2016.11.123
Bärlund, I., Tattari, S., 2001. Ranking of parameters on the basis of their contribution to model uncertainty. Ecological Modelling 142, 11–23. https://doi.org/10.1016/S0304-3800(01)00246-0
Bärlund, I., Tattari, S., Puustinen, M., 2009. Soil parameter variability affecting simulated fieldscale water balance, erosion and phosphorus losses. Agricultural and Food Science 18, 402–416. https://doi.org/10.23986/afsci.5949
Batista, P.V.G., Davies, J., Silva, M.L.N., Quinton, J.N., 2019. On the evaluation of soil erosion models: Are we doing enough? Earth-Science Reviews 197, 102898. https://doi.org/10.1016/j.earscirev.2019.102898
Bengtson, R.L., Carter, C.E., Morris, H.F., Bartkiewicz, S.A., 1988. The Influence of Subsurface Drainage Practiceson Nitrogen and Phosphorus Losses in a Warm, Humid Climate. Transactions of the ASAE 31, 0729–0733. https://doi.org/10.13031/2013.30775
Bengtson, R.L., Carter, C.E., Morris, H.F., Kowalczuk, J.G., 1984. Reducing Water Pollution with Subsurface Drainage. Transactions of the ASAE 27, 0080–0083. https://doi.org/10.13031/2013.32739
Bengtson, R.L., Sabbagh, G., 1990. USLE P factors for subsurface drainage on low slopes in a hot, humid climate. Journal of Soil and Water Conservation 45, 480–482.
Finnish Environment Institute, 2019. Sediment and nutrient loading to surface waters in 3 different scales [WWW Document]. Finnish Environment Institute (SYKE). URL https://metasiirto.ymparisto.fi:8443/geoportal/catalog/search/resource/details.page?uuid=%7B15893DD0-0193-40AD-9E21-452D271DB791%7D (accessed 1.25.21).
Formanek, G.E., ROSS, E., Istok, J., 1987. Subsurface drainage for erosion reduction on croplands in northwestern Oregon. In: Irrigation Systems for the 21st Century, in: Proceedings of the Irrigation and Drainage Division Special Conference. American Society of Civil Engineers, New York, New York, pp. 25–31.
Gilliam, J. w., Baker, J. l., Reddy, K. r., 1999. Water Quality Effects of Drainage in Humid Regions, in: Agricultural Drainage. John Wiley & Sons, Ltd, pp. 801–830. https://doi.org/10.2134/agronmonogr38.c24
Grazhdani, S., Jacquin, F., Sulçe, S., 1996. Effect of subsurface drainage on nutrient pollution of surface waters in south eastern Albania. Science of The Total Environment 191, 15–21. https://doi.org/10.1016/0048-9697(96)05168-6
Huttunen, I., Huttunen, M., Piirainen, V., Korppoo, M., Lepistö, A., Räike, A., Tattari, S., Vehviläinen, B., 2016. A National-Scale Nutrient Loading Model for Finnish Watersheds—VEMALA. Environ Model Assess 21, 83–109. https://doi.org/10.1007/s10666-015-9470-6
Istok, J.D., Boersma, L., Kling, G.F., 1985. Subsurface drainage: An erosion control practice for Western Oregon (No. 729), Special report. Agricultural Experiment Station, Oregon State University, Cornvallis.
Lilja, H., Hyväluoma, J., Puustinen, M., Uusi-Kämppä, J., Turtola, E., 2017a. Evaluation of RUSLE2015 erosion model for boreal conditions. Geoderma Regional 10, 77–84. https://doi.org/10.1016/j.geodrs.2017.05.003
Lilja, H., Puustinen, M., Turtola, E., Hyväluoma, J., 2017b. Suomen peltojen karttapohjainen eroosioluokitus (Map-based classificication of erosion in agricultural lands of Finland ). Natural Resources Institute Finland (Luke) 36.
Maalim, F.K., Melesse, A.M., 2013. Modelling the impacts of subsurface drainage on surface runoff and sediment yield in the Le Sueur Watershed, Minnesota, USA. Hydrological Sciences Journal 58, 570–586. https://doi.org/10.1080/02626667.2013.774088
Øygarden, L., Kværner, J., Jenssen, P.D., 1997. Soil erosion via preferential flow to drainage systems in clay soils. Geoderma 76, 65–86. https://doi.org/10.1016/S0016-7061(96)00099-7
Panagos, P., Borrelli, P., Meusburger, K., Alewell, C., Lugato, E., Montanarella, L., 2015. Estimating the soil erosion cover-management factor at the European scale. Land Use Policy 48, 38–50. https://doi.org/10.1016/j.landusepol.2015.05.021
Rankinen, K., Tattari, S., Rekolainen, S., 2001. Modelling of vegetative filter strips in catchment scale erosion control. Agricultural and Food Science 10, 99–112. https://doi.org/10.23986/afsci.5684
Rekolainen, S., Posch, M., 1993. Adapting the CREAMS Model for Finnish Conditions. Hydrology Research 24, 309–322. https://doi.org/10.2166/nh.1993.10
Renard, K.G., Foster, G.R., Weesies, G.A., McCool, D.K., Yoder, D.C., 1997. Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE). Agricultural Handbook 703. US Department of Agriculture, Washington, DC, pp. 404.
Rompaey, A.J.J.V., Govers, G., 2002. Data quality and model complexity for regional scale soil erosion prediction. International Journal of Geographical Information Science 16, 663–680. https://doi.org/10.1080/13658810210148561
Skaggs, R.W., Nassehzadeh-Tabrizi, A., Foster, G.R., 1982. Subsurface drainage effects on erosion. Journal of Soil and Water Conservation 37, 167–172.
Turtola, E., Alakukku, L., Uusitalo, R., 2007. Surface runoff, subsurface drainflow and soil erosion as affected by tillage in a clayey Finnish soil. AFSci 16, 332–351. https://doi.org/10.2137/145960607784125429
Turtola, E., Paajanen, A., 1995. Influence of improved subsurface drainage on phosphorus losses and nitrogen leaching from a heavy clay soil. Agricultural Water Management 28, 295–310. https://doi.org/10.1016/0378-3774(95)01180-3
Turunen, M., Warsta, L., Paasonen-Kivekäs, M., Koivusalo, H., 2017. Computational assessment of sediment balance and suspended sediment transport pathways in subsurface drained clayey soils. Soil and Tillage Research 174, 58–69. https://doi.org/10.1016/j.still.2017.06.002
USDA, 2013. Revised Universal Soil Loss Equation Version 2 (RUSLE2), Science documentation. USDA-Agricultural Research Service, Washington, D.C.
Uusitalo, R., Turtola, E., Kauppila, T., Lilja, T., 2001. Particulate Phosphorus and Sediment in Surface Runoff and Drainflow from Clayey Soils. Journal of Environmental Quality 30, 589–595. https://doi.org/10.2134/jeq2001.302589x
Warsta, L., Taskinen, A., Koivusalo, H., Paasonen-Kivekäs, M., Karvonen, T., 2013. Modelling soil erosion in a clayey, subsurface-drained agricultural field with a three-dimensional FLUSH model. Journal of Hydrology 498, 132–143. https://doi.org/10.1016/j.jhydrol.2013.06.020
Warsta, L., Taskinen, A., Paasonen-Kivekäs, M., Karvonen, T., Koivusalo, H., 2014. Spatially distributed simulation of water balance and sediment transport in an agricultural field. Soil and Tillage Research 143, 26–37. https://doi.org/10.1016/j.still.2014.05.008
Wischmeier, W., Smith, D., 1978. Predicting rainfall erosion losses: a guide to conservation planning. Agricultural Handbook No. 537. Washington DC, USA: U.S. Department of Agriculture.
Citation: https://doi.org/10.5194/hess-2021-457-AC3
Status: closed
-
EC1: 'Comment on hess-2021-457', Christian Stamm, 18 Oct 2021
\begin{document}
{\parindent0pt % disables indentation for all the text between { and }
Comments hess-2021-457\\\
Dear authors,\\\
I list a number of more general and detailed issues. \\\
General aspects:\\\
\begin{description}
\item[Language:] In general, the text is easy to read. Still, sometimes there are issues with the grammar such as missing articles (e.g., L. 19, 34 - 35).
\item[Typos:] There are are few instances with misspellings (e.g., L. 38, 218).
\end{description}Detailed comments:\\\
\begin{description}
\item[L. 48:] Strange sentence. What does \textit{implementation through natural constraints} mean?
\item[L. 132 - 133:] How was the LS factor linked to the actual parcels? How do upslope fields influence downslope parcels?
\item[L. 133:] What is the empirical basis for the claim that sink filling increases the errors? Sinks in a DEM can be real (and should be accounted for) or can be artifacts. Why did you not distinguish between the two situations?
\item[L. 162:] According to my knowledge, the RUSLE model does conceptually not account for sub-surface transport through tile-drains. Nevertheless, you compare RUSLE simulations to empirical data of the sum of surface and subsurface sediment transport. Should that not be reflected in a conceptual modification of the RUSLE model including a model parameter accounting for the split between surface and subsurface transport?
Additionally, the subsurface flow can induce mobilisation of soil particles also within the soil profile, especially in the vicinity of subsurface drains because of the disturbances of the soil profile due to the installation of the drains. How is this accounted for?
\item[L. 274 (Tab. 4):] Please provide the number of observation years and the standard deviation of the measured erosion.
\item[L. 303 - 304:] Are these novel findings?
\item[L. 318 - 319 (Fig. 4):] The high resolution DEM only affects the the LS factor, doesn't it? Hence, only this map should make any difference to previous estimates, shouldn't it?
\item[L. 349 (Fig. 6):] To which degree are these findings novel?
\item[L. 353:] Replace \textit{high field area} by \textit{areas with a large fraction of arable land} (or similar).
\item[L. 357 - 358:] Is that statement not trivial given the definition of the EMI index?
\item[L. 445 - 446:] Where is the evidence that it was indeed the lack of high resolution risk maps that prevented the implemented of targeted measures?
\item[L. 448 - 452:] The four bullet points seem rather similar to me. Can you more precisely explain what the differences are?
\item[L. 454 - 456:] Is this a novel result?
\item[L. 458 - 460:] This seems to be quite standard knowledge, or am I wrong?
\item[L. 471 - 472:] Given that you have access to actually crop management data, it should be straight forward to assess the effects such modification in practice, shouldn't it?
\item[L. 477 - 478:] Where is the evidence for that? It is a frequently used arguments by natural scientists that improved model will enhance management, but which evidence demonstrates the validity of the claim?
\item[L. 481:] The previous erosion risk estimates were rather similar (see L. 384 - 389). So in which sense has the understanding of erosion risk considerably been improved?
\item[L. 487 - 488:] What do you mean by considering erosion risk across multiple scales? What does it mean from a scientific point of view, what does it mean in practice?
\item[L. 489 - 490:] Which aspect provides new opportunities for analysing the P- and C cycle given the similarity of previous erosion estimates?
\item[L. 491 - 492:] Where can one see this demonstration? The manuscript does not compare how policies or planning has changed due to the new erosion risk map.
\end{description}Sincerely \\\
Dr. Christian Stamm
Editor HESS
}\end{document}
Citation: https://doi.org/10.5194/hess-2021-457-EC1 -
AC1: 'Reply on EC1', Timo Räsänen, 10 Dec 2021
Dear Christian Stamm (Editor),
Thank you for your thoughtful and constructive comments (EC) – they helped to improve the manuscript. Altogether, the comments from the editor and the two referees prompted major revisions, and we have revised the manuscript accordingly. The major revisions are:
- The country scale results on actual erosion and erosion management index were removed from the revised manuscript due to limitations in the used C-factor data, which was pointed out by Pedro Batista (Referee #2). Correction of the C factor data requires considerable work and we decided to leave it as future work to be published in another publication.
- The LS factor and consequent erosion data were recalculated to account for field borders.
- The results on RUSLE evaluation, potential erosion risk and potential erosion risk near water bodies were slightly restructured to accommodate the changes caused by removal of the two parts in mentioned above.
- New sensitivity analyses were added that provide estimates on
- the propagation of uncertainties from RUSLE factors to erosion estimates
- effects of location specific cropping and management practices and temporal rainfall erosivity distribution on C- factor values and the consequent erosion estimates
- The terms “potential erosion risk” and “actual erosion risk” were replaced by terms, “erosion susceptibility” and “actual erosion” to avoid the misuse of the term “risk”.
Thus, the new findings of the revised manuscript are:
- New high-resolution (two-meter) country scale erosion susceptibility data for Finland
- New evaluation of RUSLE and its performance in boreal conditions, which considers also different spatial scales and issues related to upscaling from field parcel to larger spatial scales
- Improved scientific understanding of agricultural erosion and its spatial distribution
These findings provide new opportunities for research and erosion management. In the following we provide our comments and responses (AC) to the editor comments (EC) point by point.
Comments and answers:
EC1: [Language:] In general, the text is easy to read. Still, sometimes there are issues with the grammar such as missing articles (e.g., L. 19, 34 - 35). [Typos:] There are few instances with misspellings (e.g., L. 38, 218).
AC1: We will proofread the manuscript and correct the grammar and misspelling issues.EC2: [L. 48:] Strange sentence. What does “implementation through natural constraints” mean?
AC2: This refers to the national system of paying subsidies for farmers for implementing environmental measures, such as the buffer zones. The discussion on this was removed from the manuscript as it was less relevant after the omissions in the manuscript.EC3: How was the LS factor linked to the actual parcels? How do upslope fields influence downslope parcels?
AC3: The LS factor was recalculated in the revised manuscript to account for the field parcel borders. We explain this in revised manuscript in a following way: “The LS factor was calculated from a two-meter resolution LiDAR-based digital elevation model (DEM) of Finland (National Land Survey of Finland, 2020), and by using the SAGA-GIS Module LS Factor (Conrad, 2013) and the method of Desmet and Govers (1996 ) with default settings. The LS calculation was performed in two-meter resolution for agricultural lands in 301 hydrological units that consisted of river basins, sub-basin groups (Finnish Environment Institute, 2010) and groups of islands (Fig. S1). The agricultural lands were defined according to the field parcel data from Finnish Food Authority, which contains over one million vectorized field parcels and accounts almost all agricultural land in Finland. The use of vectorized field parcels treated each field parcel as an isolated hydrological unit (in terms of overland flow) in the LS calculation to account for the effects of varying landcover on surface runoff, as recommended by Desmet and Govers (1996). The approach is considered justified as in Finland the fields are typically well drained and commonly surrounded by open ditches, which advocates for the hydrological isolation of the field parcels. However, adjacent field parcels that shared the same parcel border were treated in the calculation as a single field parcel, since it is common that uniform field areas are divided into separate field parcels for cropping and management purposes.” The evaluation of RUSLE at the seven field sites was not affected by the recalculation of the LS. Their LS factors were originally calculated considering the field borders.EC4: [L. 133:] What is the empirical basis for the claim that sink filling increases the errors? Sinks in a DEM can be real (and should be accounted for) or can be artifacts. Why did you not distinguish between the two situations?
AC4: We removed the statement that it increases errors. Our observation was that sink filling with a two-meter resolution DEM resulted in flattening and raising of the field surface levels to the levels of neighbouring landforms, such as embankments and roads. This significantly distorted the DEM and the original surface characteristics of the fields were lost. Also, as mentioned in the EC4, fields often have natural depressions, and breaching was observed to create artificial erosion areas. We are not aware of research that would provide suggestions for correct treatment of high-resolution DEM for LS calculation on agricultural lands. We revised the justification on treatment of DEM (or lack of) in the manuscript to be clearer.EC5: [L. 162:] According to my knowledge, the RUSLE model does conceptually not account for sub-surface transport through tile-drains. Nevertheless, you compare RUSLE simulations to empirical data of the sum of surface and subsurface sediment transport. Should that not be reflected in a conceptual modification of the RUSLE model including a model parameter accounting for the split between surface and subsurface transport? Additionally, the subsurface flow can induce mobilisation of soil particles also within the soil profile, especially in the vicinity of subsurface drains because of the disturbances of the soil profile due to the installation of the drains. How is this accounted for?
AC5: According to Renard et al. (1997) subsurface drainage is considered in the P factor. Bengtson and Sabbagh (1990) suggested an average P factor value of 0.6, which was also recognised by Renard et al. (1997). However, the research on how the subsurface drainage should be considered in the RUSLE is limited, and to our understanding there is no commonly accepted approach for this. Except, RUSLE2 (USDA, 2013) incorporates a method for this, which considers the changes in K factor due to the subsurface drainage, but in our case the data was too limited to consider this. Therefore, we chose to follow the original suggestions by Renard et al. (1997) and Bengtson and Sabbagh (1990). According to our literature review the subsurface drainage reduce erosion 8-90% (on average 38%) (Bengtson et al., 1988, 1984; Bengtson and Sabbagh, 1990; Formanek et al., 1987; Gilliam et al., 1999; Grazhdani et al., 1996; Istok et al., 1985; Maalim and Melesse, 2013; Skaggs et al., 1982), which results to average P value of 0.62. We used the P value of P 0.6 as suggested and used earlier in Finland by Lilja et al. (2017a). Also, the research in Finland showed that substituting of old drainage pipes with new ones reduced erosion up to 15% on a clay soil (Turtola and Paajanen, 1995). The use of sum of surface and subsurface sediment is justified also by research. In Finland, it is observed that up to 50-90 % of the erosion loading from clay soils occurs via subsurface drainage (Finnish Environment Institute, 2019; Turtola et al., 2007; Turunen et al., 2017; Warsta et al., 2014, 2013) and that erosion material in subsurface drainage flow from clay soils originates mainly from the surface soil (Uusitalo et al., 2001). In the Finnish research, the origin of the erosion material was determined by an analysis of Cesium-137 contents of soil layers and eroded soil material in subsurface drainage flow. Also, process-based modelling studies in Finland suggest that majority of the erosion material in subsurface drainage flow originates from the surface soil (Turunen et al., 2017). A study from Norway also reports that soil material in the drain flow originated most likely from the plough layer, and the soil material was transported to subsurface drains via cracks and macropores in the soil (Øygarden et al., 1997). However, we do acknowledge that the consideration of subsurface drainage in the P factor lumps a complex and poorly understood process into a single value and has therefore limitations and is a considerable source of uncertainty. In the revised manuscript we perform a sensitivity analysis which includes the uncertainty in the P value. This justification was added to the revised manuscript. Note also that our modelling approach tests the above assumptions against empirical data and the resulting performance can be considered reasonable.EC6: [L. 274 (Tab. 4):] Please provide the number of observation years and the standard deviation of the measured erosion.
AC6: These have been added to the revised manuscript.EC7: [L. 303 - 304:] Are these novel findings?
AC7: To our knowledge these areas have not been identified as high erosion areas earlier, and therefore these findings are novel.EC8: [L. 318 - 319 (Fig. 4):] The high-resolution DEM only affects the LS factor, doesn't it? Hence, only this map should make any difference to previous estimates, shouldn't it?
AC8: The R (Panagos et al., 2015) and K factor data (Lilja et al., 2017a, 2017b) are based on existing data, but the LS data is new and created in the manuscript. Thus, the LS factor data, and the erosion estimates calculated with the LS data are new. Erosion has not been estimated earlier over the whole Finland with RUSLE at two-meter resolution. Lilja et al. (2017b) started the two-meter resolution modelling work, but it was not finished and not published.EC9: [L. 349 (Fig. 6):] To which degree are these findings novel?
AC9: According to the Pedro Batista (referee #2), the results in Fig. 6 contain large uncertainties due to limitations in the C factor. After careful consideration and performed sensitivity analyses, we agree with his view. The C factors vary by location and this was not considered in the submitted manuscript, which caused regional biases in the erosion estimates. Therefore, the results in Fig. 6 were omitted from the revised manuscript, and their correction will be addressed in future research. However, the sensitivity analyses prompted by Pedro Batista’s thoughtful comments are a new addition to the revised manuscript. The issues related to the C factor are discussed in more detail in the Batista’s comments and in our answers to him. Despite the omissions, the revised manuscript provides still substantial new findings as explained in the beginning of our comments. Reporting the sensitivities in the northern conditions is considered to provide valuable information regarding erosion assessments.EC10: [L. 353:] Replace “high field area” by “areas with a large fraction of arable land” (or similar).
AC10: This section was removed from the revised manuscript.EC11: [L. 357 - 358:] Is that statement not trivial given the definition of the EMI index?
AC11: This section was removed from the revised manuscript.EC12: [L. 445 - 446:] Where is the evidence that it was indeed the lack of high-resolution risk maps that prevented the implemented of targeted measures?
AC12: We have removed this statement from the revised manuscript.EC13: [L. 448 - 452:] The four bullet points seem rather similar to me. Can you more precisely explain what the differences are?
AC13: These are affected by the omissions and therefore the policy and management implications were thus revised, and they are now as follows: “The developed erosion susceptibility data markedly improves the basis for analysing the agricultural erosion over multiple spatial scales and consequently provides new opportunities for planning the erosion management. For example, the current data showed large areas with high agricultural intensity and high erosion susceptibility (e.g., coastal areas in Southern Finland), and how erosion varies locally (e.g., Karjaanjoki and Paimionjoki basins), including areas where field parcels near water bodies and have high erosion susceptibility. Such information can be used to guide planning and allocation of erosion management efforts. The consideration of different spatial scales is also important as different scales were found to provide different insights to erosion, which can affect the conclusions drawn from the data and the choice of erosion management measures. Larger scales can provide indication of broader areas needing erosion management, and the local scales reveal more exact locations of high erosion field parcels and help in choosing appropriate erosion management measures. Altogether, the developed data can be used to improve erosion management from policy to actual management levels. However, the use of the data requires understanding of the related uncertainties that were also clarified in this research.”EC14: [L. 454 - 456:] Is this a novel result?
AC14: The results from evaluation of RUSLE are novel, and they provide improved understanding of performance of RUSLE in boreal conditions and provide new estimates on the effects of different crop and management practices on erosion.EC15: [L. 458 - 460:] This seems to be quite standard knowledge, or am I wrong?
AC15: Agreed. Our intention here was to address a broader audience and to underline an important issue, which is sometimes neglected in practical management. Therefore, we wish to keep this sentence in the manuscript.EC16: [L. 471 - 472:] Given that you have access to actually crop management data, it should be straight forward to assess the effects such modification in practice, shouldn't it?
EC16: Yes, it should be straightforward with appropriate C values, but this work is considered to be outside the scope of the current manuscript.EC17: [L. 477 - 478:] Where is the evidence for that? It is a frequently used arguments by natural scientists that improved model will enhance management, but which evidence demonstrates the validity of the claim?
AC17: In this particular case the argument originates not only from natural scientists, but from our personal communication with actors involved planning and implementation of the environmental measures in the agricultural sector, such as the Ministry of Agriculture and Forestry and the Finnish Food Authority who are responsible of allocation of environmental measures in Finland. Of course, there is never a guarantee that improved system understanding leads to improved management outcomes as the implementation depends on variety of factors.EC18: [L. 481:] The previous erosion risk estimates were rather similar (see L. 384 -389). So in which sense has the understanding of erosion risk considerably been improved?
AC18: The current research provides a new spatially explicit data and information on erosion in high-resolution over the whole Finland and such data and information has not existed before. The earlier work does not provide either the same spatial resolution or coverage as the current work, and the spatial distribution of erosion on country scale has not been well analysed and presented in earlier publications.EC19: [L. 487 - 488:] What do you mean by considering erosion risk across multiple scales? What does it mean from a scientific point of view, what does it mean in practice?
AC19: The manuscript shows how different scales reveal different spatial patterns in erosion. From scientific point of view this means that conclusions drawn from analyses will depend on the analysis scale, and analysis only in one scale provides a limited view on spatial distribution of erosion. From practical point of view this can affect how management of erosion is approached. Broader scales can reveal regions where greater erosion management effort is needed, and local scales can provide insights into efficient location-specific targeting of mitigation measures. We added following to the revised manuscript:” The consideration of different spatial scales is also important as different scales were found to provide different insights into spatial distribution of erosion, which can affect the conclusions drawn from the data and the choice of erosion management measures. Larger scales can provide indication of broader areas needing erosion management, and the local scales can reveal the locations of the high erosion areas within the field parcels, and consequently help in choosing appropriate erosion management measures for a given location.”EC20: [L. 489 - 490:] Which aspect provides new opportunities for analysing the P- and C cycle given the similarity of previous erosion estimates?
AC20: We have decided to simplify the discussion and remove this from the manuscript.EC21: [L. 491 - 492:] Where can one see this demonstration? The manuscript does not compare how policies or planning has changed due to the new erosion risk map.
AC21: We have removed this statement from the manuscript.References
Bengtson, R.L., Carter, C.E., Morris, H.F., Bartkiewicz, S.A., 1988. The Influence of Subsurface Drainage Practiceson Nitrogen and Phosphorus Losses in a Warm, Humid Climate. Transactions of the ASAE 31, 0729–0733. https://doi.org/10.13031/2013.30775
Bengtson, R.L., Carter, C.E., Morris, H.F., Kowalczuk, J.G., 1984. Reducing Water Pollution with Subsurface Drainage. Transactions of the ASAE 27, 0080–0083. https://doi.org/10.13031/2013.32739
Bengtson, R.L., Sabbagh, G., 1990. USLE P factors for subsurface drainage on low slopes in a hot, humid climate. Journal of Soil and Water Conservation 45, 480–482.
Conrad, O., 2013. Module LS-Factor, Field Based [WWW Document]. SAGA-GIS Module Library Documentation (v2.1.4). URL http://www.saga-gis.org/saga_tool_doc/2.1.4/ta_hydrology_25.html (accessed 5.29.20).
Desmet, P.J.J., Govers, G., 1996. A GIS procedure for automatically calculating the USLE LS factor on topographically complex landscape units. Journal of Soil and Water Conservation 51, 427–433.
Finnish Environment Institute, 2019. Sediment and nutrient loading to surface waters in 3 different scales [WWW Document]. Finnish Environment Institute (SYKE). URL https://metasiirto.ymparisto.fi:8443/geoportal/catalog/search/resource/details.page?uuid=%7B15893DD0-0193-40AD-9E21-452D271DB791%7D (accessed 1.25.21).
Finnish Environment Institute, 2010. Ranta10 - rantaviiva 1:10 000 - SYKE [WWW Document]. URL https://ckan.ymparisto.fi/dataset/%7BC40D8B4A-DC66-4822-AF27-7B382D89C8ED%7D (accessed 3.25.21).
Formanek, G.E., ROSS, E., Istok, J., 1987. Subsurface drainage for erosion reduction on croplands in northwestern Oregon. In: Irrigation Systems for the 21st Century, in: Proceedings of the Irrigation and Drainage Division Special Conference. American Society of Civil Engineers, New York, New York, pp. 25–31.
Gilliam, J. w., Baker, J. l., Reddy, K. r., 1999. Water Quality Effects of Drainage in Humid Regions, in: Agricultural Drainage. John Wiley & Sons, Ltd, pp. 801–830. https://doi.org/10.2134/agronmonogr38.c24
Grazhdani, S., Jacquin, F., Sulçe, S., 1996. Effect of subsurface drainage on nutrient pollution of surface waters in south eastern Albania. Science of The Total Environment 191, 15–21. https://doi.org/10.1016/0048-9697(96)05168-6
Istok, J.D., Boersma, L., Kling, G.F., 1985. Subsurface drainage: An erosion control practice for Western Oregon (No. 729), Special report. Agricultural Experiment Station, Oregon State University, Cornvallis.
Lilja, H., Hyväluoma, J., Puustinen, M., Uusi-Kämppä, J., Turtola, E., 2017a. Evaluation of RUSLE2015 erosion model for boreal conditions. Geoderma Regional 10, 77–84. https://doi.org/10.1016/j.geodrs.2017.05.003
Lilja, H., Puustinen, M., Turtola, E., Hyväluoma, J., 2017b. Suomen peltojen karttapohjainen eroosioluokitus (Map-based classificication of erosion in agricultural lands of Finland ). Natural Resources Institute Finland (Luke) 36.
Maalim, F.K., Melesse, A.M., 2013. Modelling the impacts of subsurface drainage on surface runoff and sediment yield in the Le Sueur Watershed, Minnesota, USA. Hydrological Sciences Journal 58, 570–586. https://doi.org/10.1080/02626667.2013.774088
Øygarden, L., Kværner, J., Jenssen, P.D., 1997. Soil erosion via preferential flow to drainage systems in clay soils. Geoderma 76, 65–86. https://doi.org/10.1016/S0016-7061(96)00099-7
Panagos, P., Ballabio, C., Borrelli, P., Meusburger, K., Klik, A., Rousseva, S., Tadić, M.P., Michaelides, S., Hrabalíková, M., Olsen, P., Aalto, J., Lakatos, M., Rymszewicz, A., Dumitrescu, A., Beguería, S., Alewell, C., 2015. Rainfall erosivity in Europe. Science of The Total Environment 511, 801–814. https://doi.org/10.1016/j.scitotenv.2015.01.008
Renard, K.G., Foster, G.R., Weesies, G.A., McCool, D.K., Yoder, D.C., 1997. Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE). Agricultural Handbook 703. US Department of Agriculture, Washington, DC, pp. 404.
Skaggs, R.W., Nassehzadeh-Tabrizi, A., Foster, G.R., 1982. Subsurface drainage effects on erosion. Journal of Soil and Water Conservation 37, 167–172.
Turtola, E., Alakukku, L., Uusitalo, R., 2007. Surface runoff, subsurface drainflow and soil erosion as affected by tillage in a clayey Finnish soil. AFSci 16, 332–351. https://doi.org/10.2137/145960607784125429
Turtola, E., Paajanen, A., 1995. Influence of improved subsurface drainage on phosphorus losses and nitrogen leaching from a heavy clay soil. Agricultural Water Management 28, 295–310. https://doi.org/10.1016/0378-3774(95)01180-3
Turunen, M., Warsta, L., Paasonen-Kivekäs, M., Koivusalo, H., 2017. Computational assessment of sediment balance and suspended sediment transport pathways in subsurface drained clayey soils. Soil and Tillage Research 174, 58–69. https://doi.org/10.1016/j.still.2017.06.002
USDA, 2013. Revised Universal Soil Loss Equation Version 2 (RUSLE2), Science documentation. USDA-Agricultural Research Service, Washington, D.C.
Uusitalo, R., Turtola, E., Kauppila, T., Lilja, T., 2001. Particulate Phosphorus and Sediment in Surface Runoff and Drainflow from Clayey Soils. Journal of Environmental Quality 30, 589–595. https://doi.org/10.2134/jeq2001.302589x
Warsta, L., Taskinen, A., Koivusalo, H., Paasonen-Kivekäs, M., Karvonen, T., 2013. Modelling soil erosion in a clayey, subsurface-drained agricultural field with a three-dimensional FLUSH model. Journal of Hydrology 498, 132–143. https://doi.org/10.1016/j.jhydrol.2013.06.020
Warsta, L., Taskinen, A., Paasonen-Kivekäs, M., Karvonen, T., Koivusalo, H., 2014. Spatially distributed simulation of water balance and sediment transport in an agricultural field. Soil and Tillage Research 143, 26–37. https://doi.org/10.1016/j.still.2014.05.008
Citation: https://doi.org/10.5194/hess-2021-457-AC1
-
AC1: 'Reply on EC1', Timo Räsänen, 10 Dec 2021
-
RC1: 'Comment on hess-2021-457', Anonymous Referee #1, 25 Oct 2021
This is a very interesting paper addressing soil erosion in a Nordic country. The interesting aspects are the modified approaches of a well known model and the calibration of different factors. This manuscript may help in better spatial planning and better decision making in agricultural sector. There are some issues that can be improved. I would suggest a moderate revision.
Abstract. L12-13: In terms distances??? Please correct this sentence.
Introduction
You have used some abbreviations which are not appropriate in many parts of the manuscript. E.g. L38 Fig.s…., VegeTab.s (l.218).
L41: it is not only the transfer of phosphorus and nutrients but also the transfer of heavy metals. Please add a sentence there with a proper reference.
L70: “was” ? better to put in present. In L72: You can say that the objective of this study is addressed by 1)…………
L89-91: Your reference is always the Fig.1. Please put (B, Fig.1).
L96: The results were analysed spatially (you do not need this sentence). I tis obvious.
For Equations 2 and 3 you refer to the relevant publication. However, please be more specific by providing the reference to the LANDUM model which estimates them.
Section 2.2. what is the difference between your high resolution LS factor (2m) and the European one (at 25m)?
In table 1, please add a column with the Spatial resolution of each dataset.
Somewhere in section 2, please provide a map with the Agricultural land of Finland, with a zoom also in the location of the seven monitoring sites, etc. Maybe can you include also the boarders of the 14 selected basins?
In the paragraph 175-185: there is a fourth reason which does not allow to compare RUSLE with sediment data. Sediments are the results of many processes: gully, wind erosion, harvest erosion, landslides (not only sheet and rill erosion as RUSLE predicts).
L214: attention 2,34 should be 2.34
In section 2.6: You shoud be more specific about Emax, Emin, Ei. What are they? How they are calculated?
In figure 4, important to see also the C-factor
Figure 6a: legend. The Field area is misleading. Please use the 6a) description in the legend. The same applies for 6d. not simly EMI but what is Erosion Management Index.
The last paragraph of the conclusion is very generic. I would expect something relevant to your findings.
Finally, it will be excellent to know the most effective practices to reduce erosion in Finland.
Section 4.1 can be renamed (earlier is not an appropriate term)
Citation: https://doi.org/10.5194/hess-2021-457-RC1 -
AC2: 'Reply on RC1', Timo Räsänen, 10 Dec 2021
Dear Referee #1,
Thank you for your thoughtful and constructive comments (RC) – they helped to improve the manuscript. Altogether, the comments from the editor and the two referees prompted major revisions, and we have revised the manuscript. The major revisions are:
- The country scale results on actual erosion and erosion management index were removed from the revised manuscript due to limitations in the used C-factor data, which was pointed out by Pedro Batista (Referee #2). Correction of the C factor data requires considerable research work and decision was made to leave it as future work to be published in another publication.
- The LS factor and consequent erosion data was recalculated to account for field borders.
- The results on RUSLE evaluation, erosion susceptibility and susceptibility near water bodies were slightly restructured to accommodate the changes caused by removal of the two parts in mentioned in the previous point.
- New sensitivity analyses were added that provide estimates on
- the propagation of uncertainties from RUSLE factors to erosion estimates
- effects of location specific cropping and management practices and temporal rainfall erosivity distribution on C- factor values and the consequent erosion estimates
- The terms “potential erosion risk” and “actual erosion risk” were replaced by terms, “erosion susceptibility” and “actual erosion” to avoid the misuse of the term “risk”.
Thus, the new findings of the revised manuscript are:
- New high-resolution (two-meter) country scale erosion susceptibility data for Finland
- New evaluation of RUSLE and its performance in boreal conditions, which considers also different spatial scales and issues relate to upscaling from field parcel to larger spatial scales
- Improved scientific understanding of agricultural erosion and its spatial distribution
These findings provide new opportunities for research and erosion management. In the following we provide our comments and responses (AC) to the referee comments (RC) point by point.
Comments:
RC1: This is a very interesting paper addressing soil erosion in a Nordic country. The interesting aspects are the modified approaches of a well-known model and the calibration of different factors. This manuscript may help in better spatial planning and better decision making in agricultural sector. There are some issues that can be improved. I would suggest a moderate revision.
AC1: Thank you for your positive comments and encouragement.RC2: Abstract. L12-13: In terms distances??? Please correct this sentence.
AC2: We have revised the sentence as follows: ” The developed data revealed spatially varying erosion patterns, which has implications on erosion management. For example, high erosion rates were found in intensive agricultural areas, and in several areas high erosion rates were concentrated near water bodies, where the eroded soil is more likely to cause negative off-site impacts.”RC3: You have used some abbreviations which are not appropriate in many parts of the manuscript. E.g. L38 Fig.s…., VegeTab.s (l.218).
AC3: We will perform a proof reading of the revised manuscript. Thank you for noting these.RC4: L41: it is not only the transfer of phosphorus and nutrients but also the transfer of heavy metals. Please add a sentence there with a proper reference.
AC4: We have mentioned heavy metals in the introduction together with appropriate reference (e.g. Shi et al., 2018) in the revised manuscript.RC5: L70: “was” ? better to put in present. In L72: You can say that the objective of this study is addressed by 1)…………
AC5: Corrected as suggested in the revised manuscript.RC6: L89-91: Your reference is always the Fig.1. Please put (B, Fig.1).
AC6: The whole section was revised, and this problem does not occur in the revised manuscript.RC7: L96: The results were analysed spatially (you do not need this sentence). I tis obvious.
AC7: Agreed. The sentence was removed from the revised manuscript.RC8: For Equations 2 and 3 you refer to the relevant publication. However, please be more specific by providing the reference to the LANDUM model which estimates them.
RC8: The manuscript has been revised regarding the C factor and we use now the original definition from Renard et al. (1997).RC9: Section 2.2. what is the difference between your high resolution LS factor (2m) and the European one (at 25m)?
AC9: We assume that you refer to the 25 meter resolution EU DEM, and to the European LS data calculated from the EU DEM by Panagos et al. (2015). We used a national a two-meter resolution LiDAR-based DEM (National Land Survey of Finland, 2020) and used the same calculation tool (Conrad, 2013) and method (Desmet and Govers, 1996) as Panagos et al. (2015). It is known that the resolution of the DEM influences the calculation of the LS and the erosion estimates (e.g. Chen et al., 2018; Beeson et al., 2014). For example, coarser resolution DEMs can result in larger estimates of L values and finer DEM’s on larger estimates of S values (Fu et al., 2015). To our knowledge there are no guidelines on how to account for the effect of DEM resolution, which consequently adds uncertainty in RUSLE estimates. We are considering to add the following sentence to the discussion section of the revised manuscript: ” According to an analysis conducted in Finland (Lilja et al. 2017b), the use of two-meter resolution DEM with a modified LS calculation method of Desmet and Govers (1996) resulted in 37-43% larger erosion estimates compared to the use of 25 m resolution DEM, but it is not clear how the modification of the LS calculation method affected these estimates compared to the original approach (Desmet and Govers, 1996) and whether the field parcels were considered hydrologically isolated in the calculation of the LS factor. This comparison is, however, available only in Finnish language, and it is not published in a peer-reviewed publication.RC10: In table 1, please add a column with the Spatial resolution of each dataset.
AC10: The Tab. 1 was removed from the revised manuscript due to major revisions resulting from Pedro Batista’s (Referee #2) comments. In the revised manuscript the summary of C and P factor data are not needed, and therefore, the need for data summary table is also reduced.RC11: Somewhere in section 2, please provide a map with the Agricultural land of Finland, with a zoom also in the location of the seven monitoring sites, etc. Maybe can you include also the boarders of the 14 selected basins?
AC11: A new map is provided in the revised manuscript that shows agricultural areas, the seven field sites, the small catchments, and large river basins.References
Beeson, P.C., Sadeghi, A.M., Lang, M.W., Tomer, M.D., Daughtry, C.S.T., 2014. Sediment Delivery Estimates in Water Quality Models Altered by Resolution and Source of Topographic Data. Journal of Environmental Quality 43, 26–36. https://doi.org/10.2134/jeq2012.0148
Chen, W., Li, D.-H., Yang, K.-J., Tsai, F., Seeboonruang, U., 2018. Identifying and comparing relatively high soil erosion sites with four DEMs. Ecological Engineering 120, 449–463. https://doi.org/10.1016/j.ecoleng.2018.06.025
Conrad, O., 2013. Module LS-Factor, Field Based [WWW Document]. SAGA-GIS Module Library Documentation (v2.1.4). URL http://www.saga-gis.org/saga_tool_doc/2.1.4/ta_hydrology_25.html (accessed 5.29.20).
Desmet, P.J.J., Govers, G., 1996. A GIS procedure for automatically calculating the USLE LS factor on topographically complex landscape units. Journal of Soil and Water Conservation 51, 427–433.
Fu, S., Cao, L., Liu, B., Wu, Z., Savabi, M.R., 2015. Effects of DEM grid size on predicting soil loss from small watersheds in China. Environ Earth Sci 73, 2141–2151. https://doi.org/10.1007/s12665-014-3564-3
Lilja, H., Puustinen, M., Turtola, E., Hyväluoma, J., 2017. Suomen peltojen karttapohjainen eroosioluokitus (Map-based classificication of erosion in agricultural lands of Finland ). Natural Resources Institute Finland (Luke) 36.
National Land Survey of Finland, 2020. Elevation model 2 m [WWW Document]. URL https://www.maanmittauslaitos.fi/en/maps-and-spatial-data/expert-users/product-descriptions/elevation-model-2-m (accessed 5.29.20).
Panagos, P., Borrelli, P., Meusburger, K., 2015. A New European Slope Length and Steepness Factor (LS-Factor) for Modeling Soil Erosion by Water. Geosciences 5, 117–126. https://doi.org/10.3390/geosciences5020117
Renard, K.G., Foster, G.R., Weesies, G.A., McCool, D.K., Yoder, D.C., 1997. Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE). Agricultural Handbook 703. US Department of Agriculture, Washington, DC, pp. 404.
Shi, T., Ma, J., Wu, X., Ju, T., Lin, X., Zhang, Y., Li, X., Gong, Y., Hou, H., Zhao, L., Wu, F., 2018. Inventories of heavy metal inputs and outputs to and from agricultural soils: A review. Ecotoxicology and Environmental Safety 164, 118–124. https://doi.org/10.1016/j.ecoenv.2018.08.016
Citation: https://doi.org/10.5194/hess-2021-457-AC2
-
AC2: 'Reply on RC1', Timo Räsänen, 10 Dec 2021
-
RC2: 'Comment on hess-2021-457', Pedro Batista, 01 Nov 2021
Dear authors,
I enjoyed reading your manuscript “Improving the agricultural erosion management in Finland through high-resolution data”. I appreciated the use of the high-resolution DEM and the field-parcel data for model parameterisation. However, there are some issues, which, in my opinion, need to be addressed before the manuscript can be considered for publication.
First, you state that the goal of your study is to produce ‘erosion risk data for agricultural lands in Finland’. However, what you produced are soil loss maps, which do not translate into erosion risk assessments. I understand this is a common misconception in erosion modelling research, but the manuscript should not add to the confusion. For instance, you assume risk is the modelled erosion value for a given location, without stating the assets at risk, what negative consequences erosion could bring to these assets, and what are the probabilities of these consequences occurring.
Second, there are serious problems with methodology used for calibrating the C factor for the RUSLE. From what I understood, your approach considerably deviates from the original USLE or RUSLE methodology, neglects the influence of rainfall erosivity and crop stages, and relies solely on a deterministic parameter optimisation procedure, without considering the uncertainty in the input data and the calibration methodology. These issues are described in detail below, and please correct me if I am wrong.
Third, there is no uncertainty analysis. Although you dedicate a large amount of text to pointing out the uncertainties in the model, you did not attempt to quantify them. In my opinion, if you wish public policy to be guided by your results, you should at least provide a forward error assessment to quantify the uncertainty associated to the model parameterisation. For instance, you state that your study provides a generalisation of the effects of management practices on erosion. Do you believe it is sound to provide such a generalisation, based on a limited number of observations, without a measure of uncertainty?
As pointed out by Christian Stamm, I also have some concerns regarding how tile drainage and field borders were incorporated (or not) into the modelling.
I believe you can address these issues in a new manuscript, but that would require a different submission, in my opinion. I hope these comments are at all useful and I wish the best of luck with your research.
All the best,
Pedro Batista
Detailed comments
L26: I did not understand what you meant with “the key process causing erosion is hydrological”.
L34: Do you mean erosion is affected by the short growing period?
L36-37: I think superscripts are missing here.
L64: In my opinion, acquiring spatial data for parameterisation and calibration is more of a challenge than computational power.
L68: Could you also state some of the limitations of the USLE-family models here? For instance, you cite our paper to corroborate the ability of the USLE to simulate annual loads – I imagine you mean at the erosion plot scale. However, our review also shows how spatially distributed erosion rates compare poorly to independent measurements.
L81: Wouldn’t the RUSLE require longer time series for estimating the R factor?
L82-87: How are you defining risk? Can risk be expressed in mass area-1 time-1? It seems to me you are calculating erosion rates, which of course can be a threat to multiple assets (e.g. the soil itself, downstream infrastructure, etc). However, threats, assets, and potential consequences need to be identified in order to produce an actual risk assessment. This is a common misconception in model-based erosion risk assessments, in my opinion.
L96-97: With a 2 m resolution, couldn’t you assess risk at field-block scale?
L112: These are not the sub-factors defined in the RUSLE (see Renard et al. 1997), correct? If so, please make it clearer you are using an adaptation.
L128-130: What is the ICECREAM model? In general I could not understand how the K factor was calculated. Are you taking single K factor values for mapping units in a soil map? This can introduce large errors to model outputs (see Van Rompaey and Govers, 2002).
L134-135: This is an interesting point about the sink filling. Have you made any tests with and without it?
L150: I have some questions about this calibration. Usually, we would calculate soil loss ratios for different crop management systems/crop rotations by use of erosion plot data and/or plant, soil, and residue measurements. These ratios would then be weighted with rainfall erosivity to calculate the C factors for specific locations. However, here you are using an optimisation approach – could you explain why? Moreover, did you perform any kind of split-off test, in which part of the data is used for calibration and another for testing? Would you agree that parameter calibration is necessarily conditional and that different parameter values can produce acceptable model responses? If not, why? If so, shouldn’t you use a range of behavioural parameter values to estimate the uncertainty in your model outputs? Moreover, it seems like you calibrated the sub-factor Ccrop, not the C factor. Did I understand this correctly?
L175-176: I agree model evaluation is difficult. However, particularly when models are being used to influence public policy or to guide decision-making, model testing is a necessary step. Our point in the paper you are citing was not to say it is okay not to evaluate models because it is difficult and rarely done. Instead, we wanted to incentivise the erosion modelling community to improve how we perform model testing and uncertainty analysis.
L180-187: I agree, and I appreciate how you are open about the limitations of your testing data.
L210-221: Why was rainfall erosivity not considered in any of the C factor calculations? This is crucial in USLE-type models, since erosion rates will vary largely for a same crop if the time in which the soil is exposed coincides with the time in which there is greater rainfall erosivitiy.
L235-242: Have you considered using a sediment routing model, which would allow you to deal with these issues?
L267: “Reasonable, with limitations” seems subjective. I suggest sticking to the numbers at this point of the results.
L270: The mean error shows you if the model is biased or not, but this metric is affected by cancelation. Could you also provide the RMSE, NSE, or such? Moreover, are you comparing the average soil loss from the entire period or individual annual losses?
L287-290: Would you not expect the (mean?) estimated erosion rates per catchment to deviate from the TSS measurements anyhow? These are two different things. I agree that if there is a correlation between them, there is an indication that the model might be consistently identifying the catchments where there is greater sediment production. However, since the RUSLE does not quantify sediment delivery to water courses and only represents rill and interrill erosion, do you believe there is any chance this correlation does not amount to causation?
L290-298: Do you believe mean values per catchment are good descriptors of your data here?
L313-315: Do you mean highly erodible soils?
Fig.4: By looking at your R factor map, I imagine the spatial variability of rainfall erosivity to have a large influence on the C factors for croplands.
L324-325: There seems to be a concern here regarding connectivity and sediment delivery to water bodies. Do you think the RUSLE is an appropriate model for looking at these issues?
L390-395: If you are using the same data for calibration and testing, considering average annual soil losses, and with a +- 50% limit of acceptability, how rigorous do you believe your model evaluation procedure to be? Moreover, this methodology for incorporating the effects of sub-surface drainage into the C factor needs more explanation, in my opinion. Is (tile?) drainage affecting rill and interrill erosion or the sediment export from agricultural plots?
L400-420: Considering all these uncertainties, do you believe it is acceptable to provide a single, deterministic, model output?
L403-404: Yes, quantifying uncertainty is challenging, but shouldn’t we step up to the challenge? We have published a simple, open access, code to propagate the errors in the RUSLE (Batista et al., 2021b). Similar code using high-resolution data is available in Batista et al. (2021a). If these approaches are too computationally intensive considering the scale of the model application, I suggest using the sub-catchments where you have measured TSS data as case studies (see Tetzlaff et al., 2013).
L426-435: With a 2 m resolution, isn’t there any other way you could incorporate the influence of the buffer zones in your model?
L436: How do you define reasonable quality?
L436-441: I am sorry, but I disagree that the calibration of the C factor provided a good basis for modelling. The main reasons being:
- The inclusion of sub-surface drainage into the soil loss estimates used for calibration is not well justified;
- The methodology for calculating the C factor is not in accordance with the RUSLE handbook (Renard et al., 1997);
- Rainfall erosivity is not considered in the C factor calculations;
- The parameter optimisation procedure does not consider the uncertainty in the data, and a single parameter value per crop type is used.
References:
Batista, P., Fiener, P., Scheper, S. and Alewell, C.: A conceptual model-based sediment connectivity assessment for patchy agricultural catchments, Hydrol. Earth Syst. Sci. Discuss., (May), 1–32, doi:10.5194/hess-2021-231, 2021a.
Batista, P. V. G., Laceby, J. P., Davies, J., Carvalho, T. S., Tassinari, D., Silva, M. L. N., Curi, N. and Quinton, J. N.: A framework for testing large-scale distributed soil erosion and sediment delivery modelsâ¯: Dealing with uncertainty in models and the observational data, Environ. Model. Softw., 137, doi:10.1016/j.envsoft.2021.104961, 2021b.
Renard, K. ., Foster, G. R., Weesies, G. A., McCool, D. K. and Yoder, D. C.: Predicting Soil Erosion by Water: A Guide to Conservation Planning With the Revised Universal Soil Loss Equation (RUSLE), 1997.
Van Rompaey, A. J. J. and Govers, G.: Data quality and model complexity for regional scale soil erosion prediction, Int. J. Geogr. Inf. Sci., 16(7), 663–680, doi:10.1080/13658810210148561, 2002.
Tetzlaff, B., Friedrich, K., Vorderbrügge, T., Vereecken, H. and Wendland, F.: Distributed modelling of mean annual soil erosion and sediment delivery rates to surface waters, Catena, 102, 13–20, doi:10.1016/j.catena.2011.08.001, 2013.
Citation: https://doi.org/10.5194/hess-2021-457-RC2 -
AC3: 'Reply on RC2', Timo Räsänen, 10 Dec 2021
Dear Pedro Batista,
Thank you for your thoughtful and constructive comments (EC). They made us to rethink our analyses and they provided a great learning opportunity. Most of all, they help to improve the manuscript. Altogether, the comments from the editor and the two referees prompted major revisions, and we have revised the manuscript. The major revisions are:
- The country scale results on actual erosion and erosion management index were removed from the revised manuscript due to limitations in the used C-factor data, which was pointed out by Pedro Batista (Referee #2). Correction of the C factor data requires considerable research work and decision was made to leave it as future work to be published in another publication.
- The LS factor and consequent erosion data were recalculated to account for field borders.
- The results on RUSLE evaluation, erosion susceptibility and susceptibility near water bodies were slightly restructured to accommodate the changes caused by removal of the two parts in mentioned in the previous point.
- New sensitivity analyses were added that provide estimates on
- the propagation of uncertainties from RUSLE factors to erosion estimates
- effects of location specific cropping and management practices and temporal rainfall erosivity distribution on C- factor values and the consequent erosion estimates
- The terms “potential erosion risk” and “actual erosion risk” of the original manuscript were replaced by new terms, “erosion susceptibility” and “actual erosion” as proposed by the referee to avoid the misuse of the term “risk”.
Thus, the new findings of the revised manuscript are:
- New high-resolution (two-meter) country scale erosion susceptibility data for all agricultural land in Finland
- New evaluation of RUSLE and its performance in boreal conditions, which considers also different spatial scales and issues relate to upscaling from field parcel to larger spatial scales
- Improved scientific understanding of agricultural erosion and its spatial distribution
These findings provide new opportunities for research and erosion management. In the following we provide our comments and responses (AC) to the referee comments (RC) point by point.
Comments:
RC1: I enjoyed reading your manuscript “Improving the agricultural erosion management in Finland through high-resolution data”. I appreciated the use of the high-resolution DEM and the field-parcel data for model parameterisation. However, there are some issues, which, in my opinion, need to be addressed before the manuscript can be considered for publication.
AC1: Happy to hear that you found our manuscript interesting. Thank you.RC2: First, you state that the goal of your study is to produce ‘erosion risk data for agricultural lands in Finland’. However, what you produced are soil loss maps, which do not translate into erosion risk assessments. I understand this is a common misconception in erosion modelling research, but the manuscript should not add to the confusion. For instance, you assume risk is the modelled erosion value for a given location, without stating the assets at risk, what negative consequences erosion could bring to these assets, and what are the probabilities of these consequences occurring.
AC2: We agree. This is an unfortunate blunder in terminology from our side. We have removed the word “risk” from the revised manuscript and used terms “erosion susceptibility” and “actual erosion”.RC3: Second, there are serious problems with methodology used for calibrating the C factor for the RUSLE. From what I understood, your approach considerably deviates from the original USLE or RUSLE methodology, neglects the influence of rainfall erosivity and crop stages, and relies solely on a deterministic parameter optimisation procedure, without considering the uncertainty in the input data and the calibration methodology. These issues are described in detail below, and please correct me if I am wrong. Third, there is no uncertainty analysis. Although you dedicate a large amount of text to pointing out the uncertainties in the model, you did not attempt to quantify them. In my opinion, if you wish public policy to be guided by your results, you should at least provide a forward error assessment to quantify the uncertainty associated to the model parameterisation. For instance, you state that your study provides a generalisation of the effects of management practices on erosion. Do you believe it is sound to provide such a generalisation, based on a limited number of observations, without a measure of uncertainty?
AC3: We agree with the comments here, and we have revised to the manuscript accordingly. In the revised manuscript we use the original definition of C from Renard et al. (1997), perform sensitivity analysis of forward propagation of uncertainties, and make a preliminary analysis on the effects of location specific cropping and management practices and temporal distribution of rainfall erosivity on C factor values and erosion estimates. We exclude the country scale estimation of actual erosion (A=RKLSCP) from the revised manuscript given the spatial variability C and aim to publish it later in the future, as explained in the beginning of our comment. These revisions are explained also in more detail in the responses to your specific comments.RC4: As pointed out by Christian Stamm, I also have some concerns regarding how tile drainage and field borders were incorporated (or not) into the modelling.
AC4: We have justified the incorporation of subsurface drainage in a following way in the revised manuscript. According to Renard et al. (1997) subsurface drainage is considered in the P factor. Bengtson and Sabbagh (1990) suggested an average P factor value of 0.6, which was also recognised by Renard et al. (1997). However, the research on how the subsurface drainage should be considered in the RUSLE is limited, and to our understanding there is no commonly accepted approach for this. Except, RUSLE2 (USDA, 2013) incorporates a method for this, which considers the changes in K factor due to the subsurface drainage, but in our case the data was too limited to consider this. Therefore, we chose to follow the original suggestions by Renard et al. (1997) and Bengtson and Sabbagh (1990). According to our literature review the subsurface drainage reduces erosion 8-90% (on average 38%) (Bengtson et al., 1988, 1984; Bengtson and Sabbagh, 1990; Formanek et al., 1987; Gilliam et al., 1999; Grazhdani et al., 1996; Istok et al., 1985; Maalim and Melesse, 2013; Skaggs et al., 1982), which results to average P value of 0.62. We used the P value of P 0.6 as suggested and used earlier in Finland by Lilja et al. (2017a). Also, the research in Finland showed that substituting of old drainage pipes and drain trenches with new ones reduced erosion up to 15% on a clay soil (Turtola and Paajanen, 1995). The use of sum of surface and subsurface sediment is justified also by research. In Finland, it is observed that up to 50-90 % of the erosion loading at clay soils occurs via subsurface drainage (Finnish Environment Institute, 2019; Turtola et al., 2007; Turunen et al., 2017; Warsta et al., 2014, 2013) and that in clay soils erosion material in subsurface drainage flow seems to originate mainly from the surface soil (Uusitalo et al., 2001). In this Finnish research the origin of the erosion material was determined by an analysis of Cesium-137 contents of soil layers and eroded soil material in the subsurface flow. The modelling studies in Finland also support this finding (Turunen et al., 2017). A study from Norway also reports that soil material in the drain flow originated most likely from the plough layer and, and the soil material was transported to subsurface drains via cracks and macropores in the soil (Øygarden et al., 1997). However, we do acknowledge that the consideration of subsurface drainage in the P factor lumps a complex and poorly understood process into single value and has therefore limitations and is a considerable source of uncertainty. In the revised manuscript we perform a sensitivity analysis which also includes the uncertainty in the P value. This justification was added to the revised manuscript. Note also that our modelling approach tests the above assumptions against empirical data and the resulting performance can be considered reasonable.Regarding the field borders, the LS factor was recalculated in the revised manuscript to account for the field parcel borders. We explain this in revised manuscript in a following way: “The LS factor was calculated from a two-meter resolution LiDAR-based digital elevation model (DEM) of Finland (National Land Survey of Finland, 2020), and by using the SAGA-GIS Module LS Factor (Conrad, 2013) and the method of Desmet and Govers (1996 ) with default settings. The LS calculation was performed in two-meter resolution for agricultural lands in 301 hydrological units that consisted of river basins, sub-basin groups (Finnish Environment Institute, 2010) and groups of islands (Fig. S1). The agricultural lands were defined according to the field parcel data from Finnish Food Authority, which contains over one million vectorized field parcels and it accounts almost all agricultural land in Finland. The use of vectorized field parcels treated each field parcel as an isolated hydrological unit (in terms of overland flow) in the LS calculation to account for the effects of varying landcover on surface runoff, as recommended by Desmet and Govers (1996). The approach is considered adequate as in Finland the fields are well drained and commonly surrounded by ditches, which advocates for the hydrological isolation of the field parcels. However, adjacent field parcels that shared the same parcel border were treated in the calculation as a single field parcel, since it is common that uniform field areas are divided into separate field parcels, as reported in the data of Finnish Food Authority, for annual cropping and management purposes.” The evaluation of RUSLE at the seven field sites was not affected by the recalculation of the LS. Their LS factors were originally calculated using field borders.
RC5: L26: I did not understand what you meant with “the key process causing erosion is hydrological”.
AC5: This is just poor language from our side. The purpose was to say that soil erosion by water is a hydrologically driven process. We have revised this in the revised manuscript.RC6: L34: Do you mean erosion is affected by the short growing period?
AC6: The word “process” was removed. Thank you for noting.RC7: L36-37: I think superscripts are missing here.
AC7: Corrected in the revised manuscript.RC8: L64: In my opinion, acquiring spatial data for parameterisation and calibration is more of a challenge than computational power.
AC8: We have added the issue of parameterisation into the discussion.RC9: L68: Could you also state some of the limitations of the USLE-family models here? For instance, you cite our paper to corroborate the ability of the USLE to simulate annual loads – I imagine you mean at the erosion plot scale. However, our review also shows how spatially distributed erosion rates compare poorly to independent measurements.
AC9: We have clarified that the sentence refers to the plot scale and we added the following sentences to the revised manuscript: ”The USLE type models are also commonly used over large spatial scales with varying spatial resolutions, but the success of these applications varies more than on plot scale. It is however, observed that the spatially distributed approaches are often able to rank erosion-prone fields if high quality data are available for parameterization, but the actual erosion rates compare poorly to independent measurements (Batista et al., 2019)”RC10: L81: Wouldn’t the RUSLE require longer time series for estimating the R factor?
AC10: The short data period can indeed cause uncertainties in the R factor, since in short timeseries for example exceptional events can have a greater effect on the average R estimates. This may lead to uncertainties in the magnitude and the spatial distribution of the R. Following sentence was added to the discussion section of the revised manuscript: “The R factor was based on European data, which in the case of Finland was based on measurement data from 60 stations from the period 2007-2013. This data period is relatively short and may therefore cause inaccuracies in the average level and spatial patterns of rainfall erosivity.” However, the used data is currently the best available, and improvements can be made by estimating new R factor data, but this is beyond the scope of the manuscript.RC11: L82-87: How are you defining risk? Can risk be expressed in mass area-1 time-1? It seems to me you are calculating erosion rates, which of course can be a threat to multiple assets (e.g. the soil itself, downstream infrastructure, etc). However, threats, assets, and potential consequences need to be identified in order to produce an actual risk assessment. This is a common misconception in model-based erosion risk assessments, in my opinion.
AC11: We have removed the word “risk” in the revised manuscript and as explained earlier. We will use terms “erosion susceptibility” and “actual erosion”.RC12: L96-97: With a 2 m resolution, couldn’t you assess risk at field-block scale?
AC12: Yes. The developed two-meter resolution erosion data, which will be publicly available, allows field parcel and within-field parcel scale analyses. In the manuscript we however, considered that it is more useful to introduce the developed data and the spatial patterns of erosion primarily on a broader scale, since the development of country scale data is the key novelty of the research.RC13: L112: These are not the sub-factors defined in the RUSLE (see Renard et al. 1997), correct? If so, please make it clearer you are using an adaptation.
AC13: In the revised manuscript we use the original definition of C by Renard et al. (1997).RC14: L128-130: What is the ICECREAM model? In general I could not understand how the K factor was calculated. Are you taking single K factor values for mapping units in a soil map? This can introduce large errors to model outputs (see Van Rompaey and Govers, 2002).
AC14: ICECREAM is a version of CREAMS model that has been adapted to boreal winter conditions (Rekolainen and Posch, 1993). The spatial K factor data was not calculated in this study, and it was developed earlier by Lilja et al. (2017a, 2017b) , by using the K values that have been estimated and evaluated in national research (e.g., Rekolainen and Posch, 1993; Bärlund and Tattari, 2001; Bärlund et al., 2009; Huttunen et al., 2016; Rankinen et al., 2001). The soil class specific K factor values are shown in the supplement. The Van Rompaey and Govers (2002) was very interesting reading, and we have taken a note on this and also included it in the discussion section of the manuscript. However, we did not follow their idea of simplifying K factor data, given that their findings are application and location specific and therefore they don’t recommend direct extrapolation of their findings. However, as a sensitivity analysis such simplification would be an interesting, but we decided to leave it for future work. Also, the simplification of K factor data would have resulted in loss of known soil regions in Finland that affect the regional distribution of erosion. These soil regions include, for example clay soils areas in Southern Finland, and the highly erodible soils in the river valleys in the western coast. We have clarified the nature of the K factor data in the revised manuscript.RC15: L134-135: This is an interesting point about the sink filling. Have you made any tests with and without it?
AC15: Unfortunately, we have not made tests on this, and it is a potential future work.RC16: L150: I have some questions about this calibration. Usually, we would calculate soil loss ratios for different crop management systems/crop rotations by use of erosion plot data and/or plant, soil, and residue measurements. These ratios would then be weighted with rainfall erosivity to calculate the C factors for specific locations. However, here you are using an optimisation approach – could you explain why? Moreover, did you perform any kind of split-off test, in which part of the data is used for calibration and another for testing? Would you agree that parameter calibration is necessarily conditional and that different parameter values can produce acceptable model responses? If not, why? If so, shouldn’t you use a range of behavioural parameter values to estimate the uncertainty in your model outputs? Moreover, it seems like you calibrated the sub-factor Ccrop, not the C factor. Did I understand this correctly?
AC16: The optimisation approach was used simply because of the lack of measurement data that would allow calculation of soil loss ratios. The optimisation was intended to provide rough estimates of the general magnitude of the C factors, which can be improved later by future measurements. Note that the approach has been applied previously by Lilja et al. (2017a). The erosion measurement data was also limited, and it did not allow splitting of the data to calibration and validation sets. We are also very aware that different parameter combinations can provide similar outcomes, which is common feature in modelling. However, we did our best to parameterise the RUSLE with realistic factor values, given the data limitations. In the revised manuscript the C factor is defined according to the Renard et al. (1997), and the optimisation was done for the C factor of each cropping and management practice separately. Subfactor thinking of Panagos et al. (2015) is not included in the manuscript anymore.RC17: L175-176: I agree model evaluation is difficult. However, particularly when models are being used to influence public policy or to guide decision-making, model testing is a necessary step. Our point in the paper you are citing was not to say it is okay not to evaluate models because it is difficult and rarely done. Instead, we wanted to incentivise the erosion modelling community to improve how we perform model testing and uncertainty analysis.
AC17: We have removed this sentence to avoid the misinterpretation.RC18: L180-187: I agree, and I appreciate how you are open about the limitations of your testing data.
AC18: In this case, we considered it is important to be very specific on the nature of the analysis to avoid misinterpretations.RC19: L210-221: Why was rainfall erosivity not considered in any of the C factor calculations? This is crucial in USLE-type models, since erosion rates will vary largely for a same crop if the time in which the soil is exposed coincides with the time in which there is greater rainfall erosivitiy.
AC19: In the revised manuscript we have made an analysis of the effects of location specific cropping and management practices and temporal distribution on rainfall erosivity on the C factor values and erosion estimates using the seven field sites as analysis locations. This was done by estimating the average annual C values at the seven field sites on monthly basis according to Renard (1997). The monthly R data was taken from Ballabio et al. (2017) and two sets of crop sequence specific soil loss ratio values were taken from Wischmeier and Smith (1978) as there were no measurement based soil loss ratio data available from Finland. The selected soil loss ratios were intended to parameterise cereals with normal autumn ploughing and spring cereals with reduced autumn tillage, and the soil loss ratio values were adapted to Finnish location specific cropping schedules. The main difference between the Southern, Northern and Eastern field sites is that in the two latter the spring field preparation and sowing occurs on average two weeks later. The aim of this preliminary analysis was to provide indication of the variability C values in Finland and its effect on erosion estimates, and it was not aimed to provide accurate estimates of the C factors in Finland, as described in more detail in the revised manuscript. The outcomes of the analysis provided interesting insights into the variation of C and erosion estimate by location, and to our knowledge the findings are novel in Finland.RC20: L235-242: Have you considered using a sediment routing model, which would allow you to deal with these issues?
AC20: At the current development stage of RUSLE in Finland, the inclusion of routing model is too early, but we consider it as an important future research direction.RC21: L267: “Reasonable, with limitations” seems subjective. I suggest sticking to the numbers at this point of the results.
AC21: Agreed and these remarks were removed.RC22: L270: The mean error shows you if the model is biased or not, but this metric is affected by cancelation. Could you also provide the RMSE, NSE, or such? Moreover, are you comparing the average soil loss from the entire period or individual annual losses?
AC22: We have added RMSE value to the revised manuscript.RC23: L287-290: Would you not expect the (mean?) estimated erosion rates per catchment to deviate from the TSS measurements anyhow? These are two different things. I agree that if there is a correlation between them, there is an indication that the model might be consistently identifying the catchments where there is greater sediment production. However, since the RUSLE does not quantify sediment delivery to water courses and only represents rill and interrill erosion, do you believe there is any chance this correlation does not amount to causation?
AC23: Yes. Erosion estimates of RUSLE and measured TSS from streams and rivers are different aspects of the erosion, transport, and deposition process, and represent different things. We have tried to be very open on the nature of this analysis and its limitations in the manuscript, and we do acknowledge that the relationship of correlation and causation may contain uncertainties in this case. Our thinking was that it still better to include this analysis to the manuscript than to exclude it, since the information provides some indication of the performance of spatially distributed RUSLE, given their known uncertainties (Batista et al., 2019).RC24: L290-298: Do you believe mean values per catchment are good descriptors of your data here?
AC24: We assume that you are referring to the slope values. The presented slope ranges describe how the average slope of field parcel varies by sub-catchment. We are not performing any quantitative analyses based on these values, and they are only for illustrating the differences of the two catchments. Therefore, we think average slopes are adequate for this purpose. We clarified in the revised manuscript that these values refer to the average slopes of field parcels by subcatchments.RC25: L313-315: Do you mean highly erodible soils?
AC25: Yes, thank you for noting this. Corrected in the revised manuscript.RC26: Fig.4: By looking at your R factor map, I imagine the spatial variability of rainfall erosivity to have a large influence on the C factors for croplands.
AC26: This was analysed in the revised manuscript. See author comment (AC19)RC27: L324-325: There seems to be a concern here regarding connectivity and sediment delivery to water bodies. Do you think the RUSLE is an appropriate model for looking at these issues?
AC27: RUSLE does not incorporate sediment transport, nor did we account for connectivity, but we do think that the RUSLE provides useful information on the variability of erosion near water bodies, both from the scientific and management point of view. However, RUSLE is not able to indicate which of the identified high erosion areas near water bodies will lead to sediment delivery to rivers and streams, and at this point such evaluation from the developed data requires judgement from the data user.RC28: L390-395: If you are using the same data for calibration and testing, considering average annual soil losses, and with a +- 50% limit of acceptability, how rigorous do you believe your model evaluation procedure to be?
AC28: This model evaluation was not designed in our manuscript, but it was used by Lilja et al. (2017a). Our intention was to make our results comparable within the evaluation framework of Lilja et al. (2017a). We have clarified this in the revised manuscript. Note that typically the error was lower, as explained in Section 3.2. of the manuscript.RC29: Moreover, this methodology for incorporating the effects of sub-surface drainage into the C factor needs more explanation, in my opinion. Is (tile?) drainage affecting rill and interrill erosion or the sediment export from agricultural plots?
AC29: We have justified the incorporation of subsurface drainage in the P factor in the revised manuscript in a following way. According to Renard et al. (1997) subsurface drainage is considered in the P factor. Bengtson and Sabbagh (1990) suggested an average P factor value of 0.6, which was also recognised by Renard et al. (1997). However, the research on how the subsurface drainage should be considered in the RUSLE is limited, and to our understanding there is no commonly accepted approach for this. Except, RUSLE2 (USDA, 2013) incorporates a method for this, which considers the changes in K factor due to the subsurface drainage, but in our case the data was too limited to consider this. Therefore, we chose to follow the original suggestions by Renard et al. (1997) and Bengtson and Sabbagh (1990). According to our literature review the subsurface drainage reduce erosion 8-90% and on average 38% (Bengtson et al., 1988, 1984; Bengtson and Sabbagh, 1990; Formanek et al., 1987; Gilliam et al., 1999; Grazhdani et al., 1996; Istok et al., 1985; Maalim and Melesse, 2013; Skaggs et al., 1982), which results to average P value of 0.62. We used the P value of P 0.6 as suggested and used earlier in Finland by Lilja et al. (2017a). Also, the research in Finland showed that substituting of old drainage pipes with new ones reduced erosion up to 15% on a clay soil (Turtola and Paajanen, 1995). According to the research cited above, the main mechanisms how subsurface drainage reduces rill and inter-rill erosion is via reduced surface runoff, increased soil permeability and increased crop yield.The use of sum of surface and subsurface sediment is justified also by research. In Finland, it is observed that up to 50-90 % of the erosion loading occurs via subsurface drainage (Finnish Environment Institute, 2019; Turtola et al., 2007; Turunen et al., 2017; Warsta et al., 2014, 2013) and that erosion material in subsurface drainage flow originates from the surface soil (Uusitalo et al., 2001). The origin of the erosion material was determined by an analysis of Cesium-137 contents of soil layers and eroded soil material, and the findings are in agreement with modelling results (Turunen et al., 2017). A study from Norway also reports that soil material in the drain flow originated most likely from the plough layer and, and the soil material was transported to subsurface drains via cracks and macropores in the soil (Øygarden et al., 1997). However, we do acknowledge that the consideration of subsurface drainage in the P factor lumps a complex and poorly understood process into single value and has therefore limitations and is a considerable source of uncertainty. In the revised manuscript we perform a sensitivity analysis which includes also the uncertainty in the P value.
References
Ballabio, C., Borrelli, P., Spinoni, J., Meusburger, K., Michaelides, S., Beguería, S., Klik, A., Petan, S., Janeček, M., Olsen, P., Aalto, J., Lakatos, M., Rymszewicz, A., Dumitrescu, A., Tadić, M.P., Diodato, N., Kostalova, J., Rousseva, S., Banasik, K., Alewell, C., Panagos, P., 2017. Mapping monthly rainfall erosivity in Europe. Science of The Total Environment 579, 1298–1315. https://doi.org/10.1016/j.scitotenv.2016.11.123
Bärlund, I., Tattari, S., 2001. Ranking of parameters on the basis of their contribution to model uncertainty. Ecological Modelling 142, 11–23. https://doi.org/10.1016/S0304-3800(01)00246-0
Bärlund, I., Tattari, S., Puustinen, M., 2009. Soil parameter variability affecting simulated fieldscale water balance, erosion and phosphorus losses. Agricultural and Food Science 18, 402–416. https://doi.org/10.23986/afsci.5949
Batista, P.V.G., Davies, J., Silva, M.L.N., Quinton, J.N., 2019. On the evaluation of soil erosion models: Are we doing enough? Earth-Science Reviews 197, 102898. https://doi.org/10.1016/j.earscirev.2019.102898
Bengtson, R.L., Carter, C.E., Morris, H.F., Bartkiewicz, S.A., 1988. The Influence of Subsurface Drainage Practiceson Nitrogen and Phosphorus Losses in a Warm, Humid Climate. Transactions of the ASAE 31, 0729–0733. https://doi.org/10.13031/2013.30775
Bengtson, R.L., Carter, C.E., Morris, H.F., Kowalczuk, J.G., 1984. Reducing Water Pollution with Subsurface Drainage. Transactions of the ASAE 27, 0080–0083. https://doi.org/10.13031/2013.32739
Bengtson, R.L., Sabbagh, G., 1990. USLE P factors for subsurface drainage on low slopes in a hot, humid climate. Journal of Soil and Water Conservation 45, 480–482.
Finnish Environment Institute, 2019. Sediment and nutrient loading to surface waters in 3 different scales [WWW Document]. Finnish Environment Institute (SYKE). URL https://metasiirto.ymparisto.fi:8443/geoportal/catalog/search/resource/details.page?uuid=%7B15893DD0-0193-40AD-9E21-452D271DB791%7D (accessed 1.25.21).
Formanek, G.E., ROSS, E., Istok, J., 1987. Subsurface drainage for erosion reduction on croplands in northwestern Oregon. In: Irrigation Systems for the 21st Century, in: Proceedings of the Irrigation and Drainage Division Special Conference. American Society of Civil Engineers, New York, New York, pp. 25–31.
Gilliam, J. w., Baker, J. l., Reddy, K. r., 1999. Water Quality Effects of Drainage in Humid Regions, in: Agricultural Drainage. John Wiley & Sons, Ltd, pp. 801–830. https://doi.org/10.2134/agronmonogr38.c24
Grazhdani, S., Jacquin, F., Sulçe, S., 1996. Effect of subsurface drainage on nutrient pollution of surface waters in south eastern Albania. Science of The Total Environment 191, 15–21. https://doi.org/10.1016/0048-9697(96)05168-6
Huttunen, I., Huttunen, M., Piirainen, V., Korppoo, M., Lepistö, A., Räike, A., Tattari, S., Vehviläinen, B., 2016. A National-Scale Nutrient Loading Model for Finnish Watersheds—VEMALA. Environ Model Assess 21, 83–109. https://doi.org/10.1007/s10666-015-9470-6
Istok, J.D., Boersma, L., Kling, G.F., 1985. Subsurface drainage: An erosion control practice for Western Oregon (No. 729), Special report. Agricultural Experiment Station, Oregon State University, Cornvallis.
Lilja, H., Hyväluoma, J., Puustinen, M., Uusi-Kämppä, J., Turtola, E., 2017a. Evaluation of RUSLE2015 erosion model for boreal conditions. Geoderma Regional 10, 77–84. https://doi.org/10.1016/j.geodrs.2017.05.003
Lilja, H., Puustinen, M., Turtola, E., Hyväluoma, J., 2017b. Suomen peltojen karttapohjainen eroosioluokitus (Map-based classificication of erosion in agricultural lands of Finland ). Natural Resources Institute Finland (Luke) 36.
Maalim, F.K., Melesse, A.M., 2013. Modelling the impacts of subsurface drainage on surface runoff and sediment yield in the Le Sueur Watershed, Minnesota, USA. Hydrological Sciences Journal 58, 570–586. https://doi.org/10.1080/02626667.2013.774088
Øygarden, L., Kværner, J., Jenssen, P.D., 1997. Soil erosion via preferential flow to drainage systems in clay soils. Geoderma 76, 65–86. https://doi.org/10.1016/S0016-7061(96)00099-7
Panagos, P., Borrelli, P., Meusburger, K., Alewell, C., Lugato, E., Montanarella, L., 2015. Estimating the soil erosion cover-management factor at the European scale. Land Use Policy 48, 38–50. https://doi.org/10.1016/j.landusepol.2015.05.021
Rankinen, K., Tattari, S., Rekolainen, S., 2001. Modelling of vegetative filter strips in catchment scale erosion control. Agricultural and Food Science 10, 99–112. https://doi.org/10.23986/afsci.5684
Rekolainen, S., Posch, M., 1993. Adapting the CREAMS Model for Finnish Conditions. Hydrology Research 24, 309–322. https://doi.org/10.2166/nh.1993.10
Renard, K.G., Foster, G.R., Weesies, G.A., McCool, D.K., Yoder, D.C., 1997. Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE). Agricultural Handbook 703. US Department of Agriculture, Washington, DC, pp. 404.
Rompaey, A.J.J.V., Govers, G., 2002. Data quality and model complexity for regional scale soil erosion prediction. International Journal of Geographical Information Science 16, 663–680. https://doi.org/10.1080/13658810210148561
Skaggs, R.W., Nassehzadeh-Tabrizi, A., Foster, G.R., 1982. Subsurface drainage effects on erosion. Journal of Soil and Water Conservation 37, 167–172.
Turtola, E., Alakukku, L., Uusitalo, R., 2007. Surface runoff, subsurface drainflow and soil erosion as affected by tillage in a clayey Finnish soil. AFSci 16, 332–351. https://doi.org/10.2137/145960607784125429
Turtola, E., Paajanen, A., 1995. Influence of improved subsurface drainage on phosphorus losses and nitrogen leaching from a heavy clay soil. Agricultural Water Management 28, 295–310. https://doi.org/10.1016/0378-3774(95)01180-3
Turunen, M., Warsta, L., Paasonen-Kivekäs, M., Koivusalo, H., 2017. Computational assessment of sediment balance and suspended sediment transport pathways in subsurface drained clayey soils. Soil and Tillage Research 174, 58–69. https://doi.org/10.1016/j.still.2017.06.002
USDA, 2013. Revised Universal Soil Loss Equation Version 2 (RUSLE2), Science documentation. USDA-Agricultural Research Service, Washington, D.C.
Uusitalo, R., Turtola, E., Kauppila, T., Lilja, T., 2001. Particulate Phosphorus and Sediment in Surface Runoff and Drainflow from Clayey Soils. Journal of Environmental Quality 30, 589–595. https://doi.org/10.2134/jeq2001.302589x
Warsta, L., Taskinen, A., Koivusalo, H., Paasonen-Kivekäs, M., Karvonen, T., 2013. Modelling soil erosion in a clayey, subsurface-drained agricultural field with a three-dimensional FLUSH model. Journal of Hydrology 498, 132–143. https://doi.org/10.1016/j.jhydrol.2013.06.020
Warsta, L., Taskinen, A., Paasonen-Kivekäs, M., Karvonen, T., Koivusalo, H., 2014. Spatially distributed simulation of water balance and sediment transport in an agricultural field. Soil and Tillage Research 143, 26–37. https://doi.org/10.1016/j.still.2014.05.008
Wischmeier, W., Smith, D., 1978. Predicting rainfall erosion losses: a guide to conservation planning. Agricultural Handbook No. 537. Washington DC, USA: U.S. Department of Agriculture.
Citation: https://doi.org/10.5194/hess-2021-457-AC3
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
1,030 | 505 | 49 | 1,584 | 102 | 39 | 39 |
- HTML: 1,030
- PDF: 505
- XML: 49
- Total: 1,584
- Supplement: 102
- BibTeX: 39
- EndNote: 39
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1