the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Nov 2021
Long-term trends in agricultural droughts over Netherlands and Germany: how extreme was the year 2018?
Abstract. Droughts can have important impacts on environment and economy like in the year 2018 in parts of Europe. Droughts can be analyzed in terms of meteorological drought, agricultural drought, hydrological drought and social-economic drought. In this paper, we focus on meteorological and agricultural drought and analyzed drought trends for the period 1965–2019 and assessed how extreme the drought year 2018 was in Germany and the Netherlands. The analysis was made on the basis of the following drought indices: standardized precipitation index (SPI), standardized soil moisture index (SSI), potential precipitation deficit (PPD) and ET deficit. SPI and SSI were computed at two time scales, the period April-September and a 12-months period. In order to analyze drought trends and the ranking of the year 2018, HYDRUS 1-D simulations were carried out for 31 sites with long-term meteorological observations and soil moisture, potential evapotranspiration (ET) and actual ET were determined for five soil types (clay, silt, loam, sandy loam and loamy sand). The results show that the year 2018 was severely dry, which was especially related to the highest potential ET in the time series 1965–2019, for most of the sites. For around half of the 31 sites the year 2018 had the lowest SSI, and largest PPD and ET-deficit in the 1965–2019 time series, followed by 1976 and 2003. The trend analysis reveals that meteorological drought (SPI) hardly shows significant trends over 1965–2019 over the studied domain, but agricultural droughts (SSI) are increasing, at several sites significantly, and at even more sites PPD and ET deficit show significant trends. The increasing droughts over Germany and Netherlands are mainly driven by increasing potential ET and increasing vegetation water demand.
- Preprint
(3102 KB) - Metadata XML
-
Supplement
(1936 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on hess-2021-569', Anonymous Referee #1, 06 Dec 2021
This study aims to assess how extreme the drought year 2018 was in Germany and the Netherlands, based on standard drought indices (SPI, SSI), also potential precipitation deficit (PPD), and ET deficit. The study used HYDRUS 1D simulations for 31 stations with long-term meto observations, and calculated soil moisture, potential ET, and actual ET for five soil types. Their results show that the increasing droughts over Germany and the Netherlands are mainly driven by increasing potential ET and increasing vegetation water demand. While the topic is relevant and analysis is interesting, this reviewer found the manuscript cannot be accepted with the current form.
Major concern:
This reviewer found that the study/experiment design could be flawed, mainly due to the lack of description on how representative the five soil types used for representing the different domains, and the lack of description on why the pasture is assumed for all these stations. To this reviewer, the current study is merely a synthetic study. Thus, it is far from understanding the drought year 2018 over the Netherlands and Germany.
Citation: https://doi.org/10.5194/hess-2021-569-RC1 - AC2: 'Reply on RC1', Yafei Huang, 08 Jan 2022
-
RC2: 'Comment on hess-2021-569', Anonymous Referee #2, 12 Dec 2021
The authors use Hydrus-1D to simulate water balance to test the impact of meteorological drought on the agricultural drought over 31 (over a 55-long period in 1965-2019) meteorological stations Germany and Netherlands by focusing on the exceptional drought in the year 2018.
The evaluation of this manuscript is based on the following questions:
- Is it a novel work based on a reliable scientific technique?
- Is it clearly structured and well-written?
- Are the experimental design and analysis of data adequate and appropriate to the investigation?
The manuscript is well-written and potentially interesting for HESS. It presents novel work on the extreme drought recorded in 2018 in continental Europe. Nonetheless, this manuscript requires substantial improvement before publication. The manuscript is not well-presented, model set up is oversimplified and data analysis is fair.
The main scientific question to the authors: is the 2018 drought an episodic event (as 1976 and 2003) or a consequence of a significant drying trend? From the abstract, the authors state that meteorological drought is episodic and SPI and other rainfall indeces indicate no significant declining trend while temperature-based ET is characterized by an increasing trend. The authors should quantify the probability related to this extreme episodic event from the SPI distribution. Same for the SSI as a consequence.
Lines 384-385: “In summary, the increase of droughts is mainly related to increasing soil moisture deficits, and reduction in actual ET. The main driver is not a precipitation decrease, but an increase of potential ET.”
Abstract
Abstract is generally OK. In order to get any feedback or relationship between climate and hydrological response through the use of indeces referred to six months or to the growing season makes sense. I recommend to remove indeces referred to 12 month duration since it ignores the fundamental impact of rainfall seasonality.
- Introduction
The state-of-the-art is well written. I list other interesting references on meteorological and agricultural droughts. Please see and comment about soil moisture index (Hunt et al., 2009)
- Data and Methodology
The authors should specify that the grass-reference potential evapotranspiration, ET0 is converted into crop-specific potential evapotranspiration, ETc by using a time-variant crop coefficient, Kc. Then, ETp is partitioned into potential transpiration, Tp and potential evaporation, Ep by using a time-variant leaf area index, LAI. Moreover, root depth is time-variant is the crop is annual and root distribution across the root zone needs to be specified. Actual evaporation, Ea and transpiration, Ta are calculated in Hydrus-1D depending on soil surface and root zone pressure head values, respectively. It is therefore clear that the simulation of ETa depends on time-variant crop characteristics and local soil hydraulic properties. In lines 174-177 I understand that Kc is ignored, LAI and root depth are considered as 2.0 (or 2.88) and 50 cm, respectively. The soil hydraulic properties should be ideally measured. If direct measurements are not available, it is highly recommended to use PTFs based on silt, clay, sand contents, bulk density and organic matter (Weihermüller et al., 2021; Nasta et al., 2021). This study is basically a sensitivity analysis by considering that the spatial variability is quantified only in terms of soil texture classes (the van Genuchten’s soil hydraulic parameters are crudely derived from tabulated values in Carsel and Parrish, 1988 in Table 3) and vegetation (assumed pastureland over the 31 stations) characteristics are constant in time and uniform in space.
In Eq. 2 remove P and ET from the Richards equation. P and Ep are the climate forcings on the upper boundary. Tp is reduced to Ta through the sink term, S in Eq. 2
- Results
Line 261: The strongest decrease in precipitation is in southern Germany (-2.2 mm/year) sounds really insignificant if compared to its mean annual value. From Fig. 2d, I see that the trend is from 800 mm/year (or so) to almost 700 mm/year (or so). Please explain.
Same problem for ET trends (Fig. 3)
To tell the truth, I don’t understand Fig. 7 and description of Fig. 7. Please, improve the presentation
- Discussion
Please, be more critical and evidence if there is room for future improvements
References
Hein, A., Condon, L., Maxwell, R. 2019. Evaluating the relative importance of precipitation, temperature and land-cover change in the hydrologic response to extreme meteorological drought conditions over the North American High PlainsHydrol. Earth Syst. Sci., 23, 1931–1950, 2019
Hunt, E.D., K.G. Hubbard, D.A. Wilhite, T.J. Arkebauer, and A.L. Dutcher. 2009. The development and evaluation of a soil moisture index. Int. J. Climatol. 29:747–759. doi:10.1002/joc.1749
Martínez-Fernández, J., González-Zamora, A., Gamuzzio, A. 2015. A soil water based index as a suitable agricultural drought indicator. Journal of Hydrology 522, 265–273
Nasta P., B. Szabó, N. Romano. 2021. Evaluation of Pedotransfer Functions for predicting soil hydraulic properties: A voyage from regional to field scales across Europe. Journal of Hydrology: Regional Studies 37, https://doi.org/10.1016/j.ejrh.2021.100903
Sánchez, N., Á. González-Zamora, M. Piles and J. Martínez-Fernández. 2016. A New Soil Moisture Agricultural Drought Index (SMADI) Integrating MODIS and SMOS Products: A Case of Study over the Iberian Peninsula. Remote Sensing. 8, 287; doi:10.3390/rs8040287
Van Loon, A.F. 2015. Hydrological drought explained. WIREs Water 2015, 2:359–392. doi: 10.1002/wat2.1085
von Gunten, D., T. Wöhling, C. P. Haslauer, D. Merchán, J. Causapé, and O. A. Cirpka. 2016. Using an integrated hydrological model to estimate the usefulness of meteorological drought indices in a changing climate. Hydrol. Earth Syst. Sci., 20, 4159–4175
Weihermüller, L., Lehmann, P., Herbst, M., Rahmati, M., Verhoef, A., Or, D., et al. (2021). Choice of pedotransfer functions matters when simulating soil water balance fluxes. Journal of Advances in Modeling Earth Systems, 13, e2020MS002404. https://doi.org/10.1029/2020MS002404
Citation: https://doi.org/10.5194/hess-2021-569-RC2 - AC3: 'Reply on RC2', Yafei Huang, 08 Jan 2022
-
RC3: 'Comment on hess-2021-569', Anonymous Referee #3, 16 Dec 2021
The main idea of this study is to assess how extreme the drought was in the year 2018 in Netherlands and Germany. They used Hydrus-1D model to simulate soil moisture, actual ET. Based on this and together with the meteorological station data, they calculated SPI, SSI, PPD and ET deficit to analyze the drought.
This topic is interesting and the manuscript is well-written. However, this manuscript needs some revision before publication. There are some questions to the authors:
1) Data and Methodology
Q1: Do you have insitu measurements to do validation about the HYDRUS 1-D model simulation? If not, how could you be sure the simulation results (actual ET, soil moisture) meet the evaluation requirements in statistics, for example, RMSE, pearson correlation coefficient. In lines 90-96, you mentioned in situ measured soil moisture data and remotely sensed soil moisture are not available for such long time series and are in general strongly affected by measurement uncertainties. But at least you should have some in situ measured soil moisture and actual ET to do the validation of the HYDRUS 1-D model simulation.
Q2: In lines 174-177, please give the equation and explain how did you derive the parameters of Penman-Monteith equation.
Q3: Why the pasture is assumed for all these stations? For these 31 stations, do you have the vegetation types information? You should use them in the simulation.
Q4: In lines 85-87, you used five different soil types out of 12 textural soil classes. How did you determine these five soil types? Do you have the soil types information of these 31 stations?
2)Results
Q1: The trend analysis of each variable should be more in-depth. The summary of part 3.1 is one sentence. The in-depth analysis and summary should be done based on the trends of multiple variables, combined with physical processes. For different sites, you should analyze the potential causes that may cause severe drought based on the specific local geographic environment. Probably you can put it in the discussion part, but at least you need to analyze it.
Q2: For the trend figure, I did not see the significance test result although you mentioned MK can do it in lines 250-252.
Q3: Logically, I did not understand the connection between the section 3.1 and section 3.2. You need to strengthen the logical connection.
Q4: To be honest, I do not understand what do you want to say in figure7. The description needs to be improved.
3)Discussion
You need to analyze more in this section.
Firstly, for the model set up, you need to point out the potential for further improvement.
Secondly, for the analysis part, you used precipitation, potential ET, actual ET, soil moisture, and four drought indices for drought trend analysis. There are other variables could be considered as well from the physical process point of view, please describe more about the future improvements.
Citation: https://doi.org/10.5194/hess-2021-569-RC3 - AC4: 'Reply on RC3', Yafei Huang, 08 Jan 2022
Status: closed
-
RC1: 'Comment on hess-2021-569', Anonymous Referee #1, 06 Dec 2021
This study aims to assess how extreme the drought year 2018 was in Germany and the Netherlands, based on standard drought indices (SPI, SSI), also potential precipitation deficit (PPD), and ET deficit. The study used HYDRUS 1D simulations for 31 stations with long-term meto observations, and calculated soil moisture, potential ET, and actual ET for five soil types. Their results show that the increasing droughts over Germany and the Netherlands are mainly driven by increasing potential ET and increasing vegetation water demand. While the topic is relevant and analysis is interesting, this reviewer found the manuscript cannot be accepted with the current form.
Major concern:
This reviewer found that the study/experiment design could be flawed, mainly due to the lack of description on how representative the five soil types used for representing the different domains, and the lack of description on why the pasture is assumed for all these stations. To this reviewer, the current study is merely a synthetic study. Thus, it is far from understanding the drought year 2018 over the Netherlands and Germany.
Citation: https://doi.org/10.5194/hess-2021-569-RC1 - AC2: 'Reply on RC1', Yafei Huang, 08 Jan 2022
-
RC2: 'Comment on hess-2021-569', Anonymous Referee #2, 12 Dec 2021
The authors use Hydrus-1D to simulate water balance to test the impact of meteorological drought on the agricultural drought over 31 (over a 55-long period in 1965-2019) meteorological stations Germany and Netherlands by focusing on the exceptional drought in the year 2018.
The evaluation of this manuscript is based on the following questions:
- Is it a novel work based on a reliable scientific technique?
- Is it clearly structured and well-written?
- Are the experimental design and analysis of data adequate and appropriate to the investigation?
The manuscript is well-written and potentially interesting for HESS. It presents novel work on the extreme drought recorded in 2018 in continental Europe. Nonetheless, this manuscript requires substantial improvement before publication. The manuscript is not well-presented, model set up is oversimplified and data analysis is fair.
The main scientific question to the authors: is the 2018 drought an episodic event (as 1976 and 2003) or a consequence of a significant drying trend? From the abstract, the authors state that meteorological drought is episodic and SPI and other rainfall indeces indicate no significant declining trend while temperature-based ET is characterized by an increasing trend. The authors should quantify the probability related to this extreme episodic event from the SPI distribution. Same for the SSI as a consequence.
Lines 384-385: “In summary, the increase of droughts is mainly related to increasing soil moisture deficits, and reduction in actual ET. The main driver is not a precipitation decrease, but an increase of potential ET.”
Abstract
Abstract is generally OK. In order to get any feedback or relationship between climate and hydrological response through the use of indeces referred to six months or to the growing season makes sense. I recommend to remove indeces referred to 12 month duration since it ignores the fundamental impact of rainfall seasonality.
- Introduction
The state-of-the-art is well written. I list other interesting references on meteorological and agricultural droughts. Please see and comment about soil moisture index (Hunt et al., 2009)
- Data and Methodology
The authors should specify that the grass-reference potential evapotranspiration, ET0 is converted into crop-specific potential evapotranspiration, ETc by using a time-variant crop coefficient, Kc. Then, ETp is partitioned into potential transpiration, Tp and potential evaporation, Ep by using a time-variant leaf area index, LAI. Moreover, root depth is time-variant is the crop is annual and root distribution across the root zone needs to be specified. Actual evaporation, Ea and transpiration, Ta are calculated in Hydrus-1D depending on soil surface and root zone pressure head values, respectively. It is therefore clear that the simulation of ETa depends on time-variant crop characteristics and local soil hydraulic properties. In lines 174-177 I understand that Kc is ignored, LAI and root depth are considered as 2.0 (or 2.88) and 50 cm, respectively. The soil hydraulic properties should be ideally measured. If direct measurements are not available, it is highly recommended to use PTFs based on silt, clay, sand contents, bulk density and organic matter (Weihermüller et al., 2021; Nasta et al., 2021). This study is basically a sensitivity analysis by considering that the spatial variability is quantified only in terms of soil texture classes (the van Genuchten’s soil hydraulic parameters are crudely derived from tabulated values in Carsel and Parrish, 1988 in Table 3) and vegetation (assumed pastureland over the 31 stations) characteristics are constant in time and uniform in space.
In Eq. 2 remove P and ET from the Richards equation. P and Ep are the climate forcings on the upper boundary. Tp is reduced to Ta through the sink term, S in Eq. 2
- Results
Line 261: The strongest decrease in precipitation is in southern Germany (-2.2 mm/year) sounds really insignificant if compared to its mean annual value. From Fig. 2d, I see that the trend is from 800 mm/year (or so) to almost 700 mm/year (or so). Please explain.
Same problem for ET trends (Fig. 3)
To tell the truth, I don’t understand Fig. 7 and description of Fig. 7. Please, improve the presentation
- Discussion
Please, be more critical and evidence if there is room for future improvements
References
Hein, A., Condon, L., Maxwell, R. 2019. Evaluating the relative importance of precipitation, temperature and land-cover change in the hydrologic response to extreme meteorological drought conditions over the North American High PlainsHydrol. Earth Syst. Sci., 23, 1931–1950, 2019
Hunt, E.D., K.G. Hubbard, D.A. Wilhite, T.J. Arkebauer, and A.L. Dutcher. 2009. The development and evaluation of a soil moisture index. Int. J. Climatol. 29:747–759. doi:10.1002/joc.1749
Martínez-Fernández, J., González-Zamora, A., Gamuzzio, A. 2015. A soil water based index as a suitable agricultural drought indicator. Journal of Hydrology 522, 265–273
Nasta P., B. Szabó, N. Romano. 2021. Evaluation of Pedotransfer Functions for predicting soil hydraulic properties: A voyage from regional to field scales across Europe. Journal of Hydrology: Regional Studies 37, https://doi.org/10.1016/j.ejrh.2021.100903
Sánchez, N., Á. González-Zamora, M. Piles and J. Martínez-Fernández. 2016. A New Soil Moisture Agricultural Drought Index (SMADI) Integrating MODIS and SMOS Products: A Case of Study over the Iberian Peninsula. Remote Sensing. 8, 287; doi:10.3390/rs8040287
Van Loon, A.F. 2015. Hydrological drought explained. WIREs Water 2015, 2:359–392. doi: 10.1002/wat2.1085
von Gunten, D., T. Wöhling, C. P. Haslauer, D. Merchán, J. Causapé, and O. A. Cirpka. 2016. Using an integrated hydrological model to estimate the usefulness of meteorological drought indices in a changing climate. Hydrol. Earth Syst. Sci., 20, 4159–4175
Weihermüller, L., Lehmann, P., Herbst, M., Rahmati, M., Verhoef, A., Or, D., et al. (2021). Choice of pedotransfer functions matters when simulating soil water balance fluxes. Journal of Advances in Modeling Earth Systems, 13, e2020MS002404. https://doi.org/10.1029/2020MS002404
Citation: https://doi.org/10.5194/hess-2021-569-RC2 - AC3: 'Reply on RC2', Yafei Huang, 08 Jan 2022
-
RC3: 'Comment on hess-2021-569', Anonymous Referee #3, 16 Dec 2021
The main idea of this study is to assess how extreme the drought was in the year 2018 in Netherlands and Germany. They used Hydrus-1D model to simulate soil moisture, actual ET. Based on this and together with the meteorological station data, they calculated SPI, SSI, PPD and ET deficit to analyze the drought.
This topic is interesting and the manuscript is well-written. However, this manuscript needs some revision before publication. There are some questions to the authors:
1) Data and Methodology
Q1: Do you have insitu measurements to do validation about the HYDRUS 1-D model simulation? If not, how could you be sure the simulation results (actual ET, soil moisture) meet the evaluation requirements in statistics, for example, RMSE, pearson correlation coefficient. In lines 90-96, you mentioned in situ measured soil moisture data and remotely sensed soil moisture are not available for such long time series and are in general strongly affected by measurement uncertainties. But at least you should have some in situ measured soil moisture and actual ET to do the validation of the HYDRUS 1-D model simulation.
Q2: In lines 174-177, please give the equation and explain how did you derive the parameters of Penman-Monteith equation.
Q3: Why the pasture is assumed for all these stations? For these 31 stations, do you have the vegetation types information? You should use them in the simulation.
Q4: In lines 85-87, you used five different soil types out of 12 textural soil classes. How did you determine these five soil types? Do you have the soil types information of these 31 stations?
2)Results
Q1: The trend analysis of each variable should be more in-depth. The summary of part 3.1 is one sentence. The in-depth analysis and summary should be done based on the trends of multiple variables, combined with physical processes. For different sites, you should analyze the potential causes that may cause severe drought based on the specific local geographic environment. Probably you can put it in the discussion part, but at least you need to analyze it.
Q2: For the trend figure, I did not see the significance test result although you mentioned MK can do it in lines 250-252.
Q3: Logically, I did not understand the connection between the section 3.1 and section 3.2. You need to strengthen the logical connection.
Q4: To be honest, I do not understand what do you want to say in figure7. The description needs to be improved.
3)Discussion
You need to analyze more in this section.
Firstly, for the model set up, you need to point out the potential for further improvement.
Secondly, for the analysis part, you used precipitation, potential ET, actual ET, soil moisture, and four drought indices for drought trend analysis. There are other variables could be considered as well from the physical process point of view, please describe more about the future improvements.
Citation: https://doi.org/10.5194/hess-2021-569-RC3 - AC4: 'Reply on RC3', Yafei Huang, 08 Jan 2022
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,087 | 418 | 52 | 1,557 | 109 | 39 | 42 |
- HTML: 1,087
- PDF: 418
- XML: 52
- Total: 1,557
- Supplement: 109
- BibTeX: 39
- EndNote: 42
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1