the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Simulation of spatially distributed sources, transport, and transformation of nitrogen from fertilization and septic system in an exurban watershed
Abstract. Excess export of reactive nitrogen in the form of nitrate (NO3–) export from exurban watersheds is a major source of water quality degradation and threatens the health of downstream and coastal waterbodies. Ecosystem restoration and best management practices (BMPs) can be introduced to reduce in-stream NO3– loads by promoting vegetation uptake and denitrification on uplands. However, accurately evaluating the effectiveness of these practices and setting regulations for nitrogen inputs requires an understanding of how human sources of nitrogen interact with ecohydrological systems. We evaluated how the spatial and temporal distribution of nitrogen sources, and the transport and transformation processes along hydrologic flowpaths control nitrogen cycling, export, and the development of “hot spots” of nitrogen flux in suburban ecosystems. We chose a well-monitored exurban watershed, Baisman Run in Baltimore County, Maryland, USA, to evaluate patterns of in-stream NO3– concentrations and upland nitrogen-related processes in response to three common activities: irrigation, fertilization, and on-site sanitary wastewater disposal (septic systems). We augmented a distributed ecohydrological model, RHESSys, with estimates of these additional loads to improve prediction and understanding of the factors generating both upland nitrogen cycling and stream NO3– concentrations. The augmented model predicted streamflow-weighted NO3– concentrations of 1.37 mg NO3–-N/L, compared to observed 1.44 mg NO3–-N/L, while the model predicted concentrations of 0.28 mg NO3–-N /L without the additional loads from human activities from water year 2013 to 2017. Estimated denitrification rates in grass lawns, a dominant land cover in suburban landscapes, were in the range of measured values. The highest predicted denitrification rates were downslope of lawn and septic locations in a constructed wetland, and at a sediment accumulation zone at the base of a gully receiving street drainage. These locations illustrate the development of hot spots for nitrogen cycling and export in both planned and “accidental” retention features. Appropriate siting of best BMPs and the identification of spontaneously developed nutrient hot spots should be pursued to retain nutrients and improve water quality.
- Preprint
(6899 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on hess-2023-256', Anonymous Referee #1, 20 Dec 2023
Review Comments for HESS-2023-256
Simulation of spatially distributed sources, transport, and transformation of nitrogen from fertilization and septic system in an exurban watershed
By Ruoyu Zhang et al.
Summary
The work uses a calibrated fully-distributed ecohydrological model to explore the nitrogen sources, and the transport and transformation processes within a small exurban catchment. The manuscript seems to contribute to important process understanding, but the current presentation needs substantial revision to legible contribution to existing literature. My key concerns are: 1) lack of clarification on research novelty in the context of existing literature; 2) questionable capacity of the calibrated model to represent key processes, which raise further question on whether the model is appropriate to infer nitrogen dynamics from. I therefore suggest this manuscript to be returned to the authors for substantial revision and resubmission.
General Comments
- What is the novelty of this study? A clear statement of this is needed with respect to the key knowledge gap in the existing literature: what is missing from current literature, and why are they important to consider. At present, the review of process-based modelling literature seems technically comprehensive, but it does not explain why the current study is needed as a useful addition to literature.
- The Introduction started discussion different types of models and their pros/cons from an earlier stage, lots of them are about inclusion of key processes (e.g., L55: hillslope water and nutrient mixing along hydrologic flowpaths). But for the readers’ benefit, it might be clearer by adding a separate paragraph before introducing all the models, to discuss the theory about key processes at the particular spatial/temporal scale that you are interested in? Then you can start discussing and contrasting models based on their process representation.
- You have a comprehensive review of process-based water quality models, what about the data-driven ones? The latter seem very useful to explain processes/changes at larger scales (e.g.,) – what’s their relevance to your study? I think this comment can be potentially addressed once you have resolved my Comment #2.
- The Methods section states that for model calibration ‘the parameter set yielding the highest NSE was used to simulate ecohydrological processes’ – this does not allow for structural uncertainty, is there any implication on your results? It might be a more robust practice to include multiple sets of ‘better performing’ parameters and then compare how they represent the hydrology; the current calibrated model seems to capture broad seasonality patterns, but either misses a few high-flow events or is a bit delayed compared to the observation (Figure 3), but it’s difficult to tell as the lines for observations and simulations in Figure 3 are on top of each other – it would be clearer to use dots and lines in showing the two sets of data
- I think the abovementioned issue in simulating hydrology also brings question on whether the water quality dynamics are well represented by the model. Besides a consistent lower bias (i.e., for ‘both’ scenario has an approx. -50% average bias, Figure 5), the simulated seasonality of NO3 concentrations also seem to differ from the observation too. I’m not convinced that this calibrate model is reasonable to further infer on hydrological/water quality processes. Has any model performance metric been calculated for NO3?
- There are some key information lacking in the Methods, some examples are listed below but they highlight need for a substantial improvement of the Methods section:
- Section 2.2 on calibration, was the model calibrated to only the streamflow record or with the water chemistry concentration data as well, and at which gauge? Please specify.
- In Table 1, what does the column ‘sensitivity parameter’ refers to? Also, for completeness, the table should also present the original parameter values estimated from SSURGO soils dataset besides the calibrated multipliers
- How are rainfall routing and runoff handled by the model? Are there any parameter to calibrated related to the rainfall-runoff processes?
- Figure 2: why is rainfall not considered as a key process? How possible is lawn only irrigated by groundwater but not rain water?
- Equation 3: PET and ET – how are there estimates?
- The Results section presents a lot of information but there is no direct link of them to the modelling outputs. I think the Methods section misses a sub-section at the end on which model outputs are analysed and how, to answer which research question (which links to the Introduction). This would be very helpful for readers to link the Results section with the rest of the paper.
Specific Comments
- Line 21 – the statement seems too long and might be confusing, can you break this into two sentences, or use labels e.g., i), ii) if a single sentence is used?
- Figure 1 can be improved by including more information on the study area, including: locations of the two monitoring sites mentioned (01583580, 01583570) and the boundary of the sub-catchment, Pond Branch. The base map would be more informative presented as a map of key land uses (e.g., forest, urban, exurban) instead of a satellite image – it is a bit hard to visualize the land use components from the latter.
- Relating to my Comment #5, I’m also confused by your statement in L403 ‘our model underestimated the mean in-stream NO3 – concentration by 0.1 mg NO3 – -N/L (-7%) with stronger variability (Fig. 5)’. In Fig. 5, I see an approx. -50% bias comparing the simulated concentration for the ‘both’ scenario compared with the observation.
Citation: https://doi.org/10.5194/hess-2023-256-RC1 - AC1: 'Reply on RC1', Ruoyu Zhang, 08 Mar 2024
-
RC2: 'Comment on hess-2023-256', Anonymous Referee #2, 18 Jan 2024
General Comments
The manuscript entitled “Simulation of spatially distributed sources, transport, and transformation of nitrogen from fertilization and septic system in an exurban watershed” presents and uses an augmented version of the RHESSys Model to evaluate the hydrologic and biogeochemical N cycling, and transport in a mixed land use watershed characterized by anthropogenic N inputs from irrigation, fertilization, and on-site sanitary wastewater disposal in form of septic systems.
The study is motivated by enhancing the understanding of transport, cycling and subsequent export to streams of N in exurban watersheds. It declares the need to defer appropriate siting for effective best management practices (BMP) to reduce N export to downstream water bodies. To my perception, it fits within the thematic scope of HESS. It addresses a highly relevant research topic and could become a substantial contribution to scientific progress; however, the novelty of the presented approach remains unclear.
The manuscript promises to address certainly interesting aspects e.g., to evaluate how the spatial and temporal distribution of human nitrogen sources in exurban water shed controls N export to downstream water bodies and nitrification rates. However, the conclusions reached in the manuscript are rather general. At the current state, the concept of the study is limited to compare the simulated N loads and concentrations and nitrification rates between simulations without and with one or two human N input types and concludes that including fertilization and septic systems improves the simulation results when comparing to observations in the case study watershed. It remains unclear whether the augmented model can be transferred to other watersheds and how it can support to situate BMPs effectively.
The presentation of the results in the figures and tables does not keep up with the high quality in other articles published in HESS. Furthermore, to my understanding the methodology lacks significant steps: i) the model validation and ii) statistics that substantiate the results. Therefore, I suggest to reject the manuscript for publication at its current state and encourage the authors to improve it.
Specific comments
#1 The motivation behind the study and the relevance of the research are well elaborated. However, the current state of the knowledge in regarding to the research questions is not elaborated. Are there no prior studies that have addressed similar research questions?
#2 The methodology should be written clearer and more structured. Given the fact that the study uses a rich base of data, for the reader it would be beneficial to have an overview of the data used for setting up the model e.g., in the form of a Table providing specifications on each dataset and how it was employed in the study.
#3 The model validation needs to be provided as well as the statistical methods for evaluating the simulation results.
#4 The results need to include estimates of errors and indication of deviation between the analyzed years.
#5 Some figures (maps) are obsolete as to my opinion the difference between them can not be spotted. Some figures have confusing axis labeling, lack titles for the legends and some have unclear captions (see technical corrections file for more details).
#6 The manuscript needs to be reviewed i) to comply with the HESS requirements for manuscript composition, ii) for unit formatting as required by the submission guidelines, iii) for correct referencing of used data sets and software
Technical corrections
Please find all technical corrections and suggestions for improvement of the figures as comments in the attached manuscript.
- AC2: 'Reply on RC2', Ruoyu Zhang, 08 Mar 2024
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
344 | 89 | 20 | 453 | 15 | 6 |
- HTML: 344
- PDF: 89
- XML: 20
- Total: 453
- BibTeX: 15
- EndNote: 6
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1