Unveiling the role of groundwater in hydrology and nitrate-N transport over agricultural watersheds in the U.S. Corn Belt: insights from model structure intercomparison

Wang, Yaji; Peng, Bin; Jia, Mengqi; Ma, Zewei; Zhao, Qianyu; Bailey, Ryan; Abrams, Daniel; Yang, Jie; Guan, Kaiyu; Pan, Ming; Song, Yuanxin

doi:10.5194/egusphere-2026-4671

Preprints

https://doi.org/10.5194/egusphere-2026-4671

Preprints

07 Aug 2026

| 07 Aug 2026

Status: this preprint is open for discussion and under review for Hydrology and Earth System Sciences (HESS).

Unveiling the role of groundwater in hydrology and nitrate-N transport over agricultural watersheds in the U.S. Corn Belt: insights from model structure intercomparison

Yaji Wang, Bin Peng, Mengqi Jia, Zewei Ma, Qianyu Zhao, Ryan Bailey, Daniel Abrams, Jie Yang, Kaiyu Guan, Ming Pan, and Yuanxin Song

Abstract. Groundwater is an important component in water cycling across agricultural landscapes, but its role in regulating hydrology and nitrogen transport within agricultural watersheds remains highly uncertain, due to limited observations and the reliance on modeling approaches with different model structural representations. Here, we investigated groundwater’s role at the watershed scale in the Upper Sangamon River Basin over 2001–2020 by evaluating its annual and seasonal contributions to both hydrology and nitrogen transport using three validated models with different representations of groundwater-related processes, including (1) default SWAT+, (2) SWAT+gwflow, and (3) SWAT+MODFLOW+RT3D. These models represent increasing levels of groundwater-process complexity within the same model family, enabling us to isolate the effects of groundwater representation on hydrology and nitrogen transport while minimizing differences arising from other model components. We find that groundwater’s annual average net contribution to streamflow ranges from 24.0 % to 31.0 % across models, but its partitioning among contributing pathways varies across models, being dominated by groundwater discharge in the default SWAT+, and by tile flow in SWAT+gwflow and SWAT+MODFLOW+RT3D. The multi-year mean net groundwater contribution to nitrate-N export ranges from 16.0 % to 35.0 %. Seasonal analysis shows that groundwater is a year-round source of nitrate–N to the river system, and its contribution peaks after fertilization at more than 10 times that of the non-growing season. Model comparison shows that physically based groundwater models (SWAT+gwflow and SWAT+MODFLOW+RT3D) better capture surface-subsurface exchange processes, while the empirical groundwater model of default SWAT+ provides efficient flux estimates. Our results highlight the importance of groundwater representation in shaping watershed hydrologic and nitrogen dynamics and underscore the need to balance process realism with computational efficiency when selecting models for agricultural watershed management.

Received: 03 Aug 2026 – Discussion started: 07 Aug 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 5773 KB)

Supplement (1505 KB)

Download & links

Yaji Wang, Bin Peng, Mengqi Jia, Zewei Ma, Qianyu Zhao, Ryan Bailey, Daniel Abrams, Jie Yang, Kaiyu Guan, Ming Pan, and Yuanxin Song

Status: open (until 18 Oct 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-4671', Anonymous Referee #1, 18 Sep 2026 reply
This manuscript compares three groundwater representations within the SWAT+ model family to examine groundwater contributions to streamflow and nitrate-N export. I think the comparison within a common modeling framework is valuable, and the integration of streamflow, groundwater-head, stream nitrogen, and groundwater nitrate observations represents a substantial modeling effort. The manuscript is well organized, and the results provide useful insights into how different groundwater representations affect simulated water and nitrogen pathways. Overall, I found this to be a useful and potentially valuable comparison for the hydrologic modeling community. I have several comments that I hope will help clarify the comparison and strengthen the interpretation of the results.
I think the components included in the reported groundwater contribution need to be clarified. Default SWAT+ primarily includes conceptual groundwater discharge, whereas SWAT+gwflow and SWAT+MODFLOW+RT3D also include tile drainage and, in SWAT+gwflow, saturation-excess flow. Because tile drainage makes up a large proportion of groundwater outflow in the coupled models, the comparison may partly reflect differences in how groundwater pathways are represented and grouped. In my view, the authors should explain how they ensured comparability across models and, if possible, report direct aquifer–stream discharge separately from tile drainage and saturation-excess flow.

The nitrogen terminology should be checked. TN is defined as nitrate-N plus nitrite-N. This quantity is generally described as nitrate plus nitrite, or NOx-N, rather than total nitrogen because it excludes ammonium and organic nitrogen. Please identify the relevant USGS parameter codes, clarify the constituent used for calibration, and ensure that the observed and simulated quantities are directly comparable.

Independent estimates of groundwater recharge and groundwater nitrate-N contributions are used as soft constraints during calibration and are later compared with the model results. I think this is a reasonable way to constrain processes that are difficult to observe, but the resulting agreement is not fully independent validation. It would be helpful to report the values and uncertainties of these constraints, explain how they were incorporated into the calibration, and make clear that the resulting groundwater contributions are model estimates informed by those constraints.

Differences among the simulations may arise not only from model structure but also from separate calibration, parameterization, boundary conditions, and the processes included in each model. The authors should clarify which parameters were shared, fixed, or independently calibrated. I also think the interpretation should acknowledge that the reported differences probably reflect a combination of these factors rather than model structure alone.

The discussion of groundwater model performance could be more balanced. Default SWAT+ has the lowest groundwater-head MAE at Fisher and the lowest overall groundwater nitrate-N MAE, whereas the physically based models appear to represent temporal or spatial variability better. To me, these findings suggest that the models have different strengths rather than one group consistently performing better. It would help to distinguish more clearly between reproducing mean conditions and capturing temporal dynamics or spatial patterns.

I think more explanation is needed for the selection of 2001–2007 as the calibration period and 2008–2020 as the validation period. The unequal lengths are not necessarily a concern, but the two periods may differ in hydroclimatic conditions, nitrogen inputs, extreme events, and observation coverage. A brief comparison of the periods would help readers evaluate the calibration–validation design. Please also clarify whether the same periods were used for streamflow, nitrogen loads, groundwater heads, and groundwater nitrate concentrations.

The seasonal nitrate-N pattern is interesting, but I think it may partly reflect the prescribed fertilizer schedule and seasonal changes in groundwater discharge. Showing groundwater discharge and nitrate concentration separately would help identify whether the seasonal peak in nitrate flux is driven mainly by water flux, concentration, or both. In the absence of independent observations, this finding would be better described as a model-based seasonal pattern associated with the prescribed management schedule.

The broader implications of this model comparison could be discussed more clearly. The Upper Sangamon River Basin is characterized by shallow groundwater, extensive tile drainage, and relatively flat agricultural land, and these features likely influence the relative importance of the groundwater pathways represented by the three models. I think the authors should distinguish between findings that are specific to this watershed and those that may be transferable to other agricultural watersheds. It would also be useful to discuss the watershed or hydrogeologic conditions under which each groundwater representation may be most appropriate.

The title refers to agricultural “watersheds” in the U.S. Corn Belt, but the study examines one watershed. The title could be revised to reflect the case-study scale.

The management discussion is useful, but the recommendations concerning fertilizer timing, subsurface residence time, and conservation practices should be presented as implications of the simulations rather than as strategies directly evaluated in this study.

Although the source code for the three modeling frameworks is available, reproducing this comparison would also require the calibrated parameter sets, model configuration files, input-processing procedures, and analysis scripts. Providing these materials in a public repository or the Supplement would improve the reproducibility and broader usefulness of the study.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-4671-RC1

Yaji Wang, Bin Peng, Mengqi Jia, Zewei Ma, Qianyu Zhao, Ryan Bailey, Daniel Abrams, Jie Yang, Kaiyu Guan, Ming Pan, and Yuanxin Song

Supplement

https://doi.org/10.5194/egusphere-2026-4671-supplement

Yaji Wang, Bin Peng, Mengqi Jia, Zewei Ma, Qianyu Zhao, Ryan Bailey, Daniel Abrams, Jie Yang, Kaiyu Guan, Ming Pan, and Yuanxin Song

Viewed

Total article views: 223 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
155	60	8	223	27	8	7

HTML: 155
PDF: 60
XML: 8
Total: 223
Supplement: 27
BibTeX: 8
EndNote: 7

Views and downloads (calculated since 07 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	107	32	3	142
Sep 2026	48	28	5	81

Cumulative views and downloads (calculated since 07 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	107	32	3	142
Sep 2026	48	28	5	81

Viewed (geographical distribution)

Total article views: 187 (including HTML, PDF, and XML) Thereof 187 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 18 Sep 2026

Short summary

Groundwater is critical for supplying rivers and transporting nitrogen, but watershed models describe these processes in different ways. We compared three watershed models with different groundwater detail levels. More detailed simulations better captured underground water movement and nitrogen transport, while simpler approaches provided reliable estimates of groundwater contributions to rivers. The results help researchers and water managers choose models that best fit different applications.


Total:	0
HTML:	0
PDF:	0
XML:	0