Wayana Dolan Melanie Vanderhoof Heather Golden Jay Christensen Charles Lane Grey Evenson 20260629 .csv and .shp Denver, CO U.S. Geological Survey https://doi.org/10.5066/P1OBBNGG Wayana Dolan Melanie Vanderhoof Heather Golden Jay Christensen Charles Lane Grey Evenson TBD Publication (journal) TBD TBD TBD TBD TBD TBD Surface water storage, including wetlands and lakes, is not typically considered in hydrological model calibrations. We tested a multivariate calibration process (Latin Hypercube Sampling), incorporating Sentinel-1/2 surface water storage, for a Soil and Water Assessment Tool model across the 0.5 million km^2 Upper Mississippi River Basin. While 19% of the 2000 parameter sets adequately simulated discharge (Kling-Gupta efficiency greater than 0.5), only 5% also adequately simulated surface water storage (mean absolute error less than 2 m), which reduced model output uncertainty. Using the best calibrated model, we found that changes in surface water storage capacity in watersheds with many non-floodplain wetlands (NFWs) most strongly affected discharge during the first annual peak flow, when storage was filling. Increases in upstream surface water storage capacity resulted in projected decreases in peak flow and flashiness, with changes persisting downstream to the watershed outlet. Our findings demonstrate the importance of including surface water storage in multivariate model calibration processes to inform flood impact predictions. The purpose of this study was to calibrate a large-scale Soil and Water Assessment Tool (SWAT) model of the Upper Mississippi River Basin (UMRB) using USGS gage data (discharge) and Sentinel-1 and Sentinel-2 remote sensing data (surface water storage). We then used the calibrated model to investigate the impact of surface water storage capacity loss and gain scenarios on downstream riverine peak flows and flashiness. This data release contains daily SWAT-simulated river discharge outputs for subbasins in the UMRB. Specifically, it includes outputs from the baseline calibrated model and from the simulated scenarios where surface water storage capacity and area were changed in watersheds vulnerable to wetland loss (watersheds with many non-floodplain wetlands), ranging from 100% surface water loss to 100% surface water gain. 20170101 20221222 Ground Condition Complete None planned -97.0870 -85.9699 47.7642 38.6907 None hydrologic modeling surface water storage SWAT Upper Mississippi River Basin model calibration wetland loss and restoration USGS Metadata Identifier USGS:6a206d64b66b01180072ecf7 None United States Midwest Mississippi River Wisconsin Minnesota Iowa Illinois Missouri Indiana South Dakota None Unless otherwise stated, all data, metadata and related materials are considered to satisfy the quality standards relative to the purpose for which the data were collected. Although these data and associated metadata have been reviewed for accuracy and completeness and approved for release by the U.S. Geological Survey (USGS), no warranty expressed or implied is made regarding the display or utility of the data on any other system or for general or scientific purposes, nor shall the act of distribution constitute any such warranty. Wayana Dolan U.S. Geological Survey, Geoscience and Environmental Change Science Center Research Physical Scientist physical

PO Box 25046, DFC, MS 980

Denver CO 80225 US n/a wdolan@usgs.gov Sentinel-1 and Sentinel-2 imagery was accessed through Google Earth Engine from the image collections COPERNICUS/S1_GRID and COPERNICUS/S2, respectively. The Sentinel-1 and -2-informed surface water extent and storage dataset that was used for SWAT model setup and calibration is published as a USGS data release: Dolan, W., Vanderhoof, M. K., Christensen, J. R., Golden, H. E., Lane, C. R., Rajib, A., Keenan, W., Zheng, Q., and Khare, A. (2025). Data release for Remotely Sensed Surface Water Storage Shows Distinct Patterns from SWAT-Simulated Data (ver. 2.0, February 2026): U.S. Geological Survey data release [Dataset]. https://doi.org/10.5066/P14WAHYW The original depression-integrated SWAT model is published in Rajib et al. (2020). Rajib, A., Golden, H. E., Lane, C. R., and Wu, Q. (2020). Surface Depression and Wetland Water Storage Improves Major River Basin Hydrologic Predictions. Water Resources Research, 56(7). https://doi.org/10.1029/2019WR026561. n/a The baseline SWAT model (the no–surface-water‑change scenario) was calibrated using daily discharge data from 27 USGS gages and biweekly average surface water storage from 274 subbasins reported in Dolan et al. (2025). Storage calibration excluded the snow‑covered period (as flagged in the Dolan et al. 2025 dataset), subbasins 75, 80, 220, and 275 where no surface water storage was present, and subbasin 183, which lacked observations in the Dolan et al. (2025) dataset. For discharge, accuracy was assessed using the Kling-Gupta efficiency metric at each gage, and then averaged across all gages (weighted by gage drainage area). The gage IDs used in this analysis are: 05267000, 05286000, 05365500, 05344500, 05316580, 05400760, 05317000, 05382000, 05378500, 05404000, 05407000, 05476750, 05458500, 05437050, 05421740, 05451500, 05443500, 05552500, 05484650, 05527500, 05465500, 05474000, 05568500, 05474500, 05582000, 05586100, 05587450. The KGE for the baseline model is 0.73 For surface water storage, accuracy was assessed using the mean absolute error (MAE) metric for each subbasin. The subbasin-specific MAE was normalized by each subbasin's area to account for differences in MAE driven by basin size. Lastly, these normalized MAE values were summed across the entire UMRB. The MAE for the best model is 1.65 m. For full discussion of model calibration methods, please see the manuscript associated with this data release. Dolan, W., Vanderhoof, M. K., Christensen, J. R., Golden, H. E., Lane, C. R., Rajib, A., Keenan, W., Zheng, Q., and Khare, A. (2025). Data release for Remotely Sensed Surface Water Storage Shows Distinct Patterns from SWAT-Simulated Data (ver. 2.0, February 2026): U.S. Geological Survey data release [Dataset]. https://doi.org/10.5066/P14WAHYW No formal logical accuracy tests were conducted. Data are complete. No formal positional accuracy tests were conducted. No formal positional accuracy tests were conducted. change_scenarios_discharge_20260625.csv This study used a daily SWAT2012 (Rev. 681) model of the Upper Mississippi River Basin (UMRB), which extends from the Mississippi River headwaters to its confluence with the Missouri River (~440,000 km^2), originally developed by Rajib et al. (2020). The 279 subbasin model represented each subbasin with a single hydrologic response unit (HRU) based on dominant land cover, soil type, and slope, an approach chosen to maintain computational efficiency across the large model domain. Tile drainage was incorporated at the subbasin scale, and all surface water storage was represented using SWAT’s hydrologically equivalent wetland (HEW) concept in SWAT’s wetland module, in which each subbasin contains one water body whose storage dynamics reflect aggregated depressional water features. We integrated satellite derived surface water observations (Dolan et al., 2025) into the model to set wetland module initial conditions and to support model calibration. Wetland module initial conditions included maximum/normal/initial volumetric surface water storage (WET_MXVOL, WET_NVOL, WET_VOL), maximum/normal areal surface water extent (WET_MXSA, WET_NSA), and the fraction of area draining to surface water (WET_FR). To reduce the influence of outliers, we set WET_MXVOL and WET_MXSA to the 90th percentile of the remotely sensed surface water storage and extent time series and set WET_NVOL and WET_NSA to their 50th percentiles. Following Rajib et al. (2020), initial storage WET_VOL was set equal to WET_NVOL. WET_FR was computed as 1 – (WET_MXSA/subbasin area) (areas in ha). We updated parameters for 274 of 279 subbasins (one lacked remote sensing data; four had negligible storage) and ran the updated model for 2017–2022 with a 3 year spin up (2014–2016), aligned to the remote sensing record. Multivariate calibration of the SWAT model used daily discharge and biweekly surface water storage, and utilized a Latin Hypercube Sampling approach to generate 2000 unique parameter sets. Performance of each parameter set was evaluated with the Kling Gupta efficiency (KGE) objective function across 27 USGS streamflow gages (averaged across all gages using drainage area as a weight to generate KGE_sim) and subbasin area normalized mean absolute error (MAE) for storage in each subbasin (summed across all subbasins to generate MAE_sim). To select a single “best” parameter set, we min max scaled KGE_sim and MAE_sim, averaged the scaled scores per parameter set, and chose the highest scoring set. Using this best parameter set and model, we then ran scenarios that systematically changed wetland storage and area parameters in SWAT’s wetland module (WET_MXVOL, WET_NVOL, WET_VOL; WET_MXSA, WET_NSA, WET_FR) from −100% to +100% in 10% increments within high non-floodplain wetland (NFW) subbasins to quantify how changes in surface water storage and extent affect downstream streamflow magnitude and variability. Subbasins were classified into 114 high NFW subbasins and 160 low NFW subbasins based upon NFW areal extent using data from Lane and D’Amico (2016). High NFW subbasins were primarily located in the northern portion of the UMRB. Dolan, W., Vanderhoof, M. K., Christensen, J. R., Golden, H. E., Lane, C. R., Rajib, A., Keenan, W., Zheng, Q., and Khare, A. (2025). Data release for Remotely Sensed Surface Water Storage Shows Distinct Patterns from SWAT-Simulated Data (ver. 2.0, February 2026): U.S. Geological Survey data release [Dataset]. https://doi.org/10.5066/P14WAHYW Lane, C. R., and D’Amico, E. (2016). Identification of Putative Geographically Isolated Wetlands of the Conterminous United States. JAWRA Journal of the American Water Resources Association, 52(3), 705–722. https://doi.org/10.1111/1752-1688.12421 Rajib, A., Golden, H. E., Lane, C. R., and Wu, Q. (2020). Surface Depression and Wetland Water Storage Improves Major River Basin Hydrologic Predictions. Water Resources Research, 56(7). https://doi.org/10.1029/2019WR026561 2024 Albers_Conical_Equal_Area 29.5 45.5 -96.0 23.0 0.0 0.0 coordinate pair 0.001 0.001 meters North American Datum of 1983 Geographic Reference System 80 6378137.0 298.257222101 FILE DESCRIPTIONS Data_Dictionary_UMRB_EMS.csv - Descriptions for all of the fields for the different layers used in this analysis in a machine-readable format. change_scenarios_discharge_20260625.csv - Daily simulated discharge outputs for each subbasin in the UMRB. This file includes outputs from the baseline calibrated model as well as outputs associated with scenarios where surface water storage and area were changed in watersheds vulnerable to wetland loss (watersheds with many non-floodplain wetlands, i.e., NFWs), ranging from 100% surface water loss to 100% surface water gain. Subbasin: Subbasin ID in the SWAT model. Pairs with the Subbasin column in the UMRB_Subbasins.shp file [1 to 3-digit integer, min: 1, max: 279] Date: Date string [YYYYMMDD, min: 20170101, max: 20221222] FlowOut_m3s1: Simulated average daily discharge out of subbasin's river channel [m^3s^-1, min: 0.00, max: 15320.00] Change_pct: Percent change in SWAT wetland module input parameters (WET_MXVOL - maximum volumetric surface water storage capacity, WET_NVOL - normal volumetric surface water storage, WET_VOL - initial volumetric surface water storage, WET_MXSA - maximum surface water area, WET_NSA - normal surface water area, WET_FR - fraction of watershed area draining into the wetland module) [%, min:-100, max: 100] NFW_class: Column describes whether or not a subbasin is considered 'high NFW' or 'low NFW' per the percent subbasin area underlain by NFWs from Lane and D'Amico (2016). Subbasins are classified as either high NFW or low NFW using a Gaussian finite mixture model. Subbasin's whose storage were not calibrated (75, 80, 220, 275, 183), which make up less than 1% of the land area in the UMRB, were counted as low NFW subbasins. [String] UMRB_Subbasins.shp, .dbf, .shx, .prj - Shapefile containing the spatial extent of each model subbasin (adapted from Rajib et al., 2020). Projection is NAD83 / Conus Albers FID: An autogenerated unique identifier assigned through ArcPro. [1 to 3-digit integer, min: 0, max:278] Shape: An autogenerated text field defining the geometry of features in this shape file. [text field, value = "Polygon"] Subbasin: Subbasin ID in the depression-integrated SWAT model (model adapted from Rajib et al., 2020). [1 to 3-digit integer, min: 1, max: 279] HydroID: Subbasin HydroID in the depression-integrated SWAT model (model adapted from Rajib et al., 2020). [6-digit integer, min: 300001, max: 300279] Lane, C. R., and D’Amico, E. (2016). Identification of Putative Geographically Isolated Wetlands of the Conterminous United States. JAWRA Journal of the American Water Resources Association, 52(3), 705–722. https://doi.org/10.1111/1752-1688.12421 Rajib, A., Golden, H. E., Lane, C. R., and Wu, Q. (2020). Surface Depression and Wetland Water Storage Improves Major River Basin Hydrologic Predictions. Water Resources Research, 56(7). https://doi.org/10.1029/2019WR026561 n/a U.S. Geological Survey ScienceBase U.S. Geological Survey ScienceBase mailing and physical

Denver Federal Center, Building 810, Mail Stop 302

Denver CO 80225 USA 1-888-275-8747 sciencebase@usgs.gov Any use of trade, product or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Geological Survey. Although this information product, for the most part, is in the public domain, it also contains copyrighted materials as noted in the text. Permission to reproduce copyrighted items for other than personal use must be secured from the copyright owner. This database has been approved for release and publication by the Director of the USGS. Although this database has been subjected to rigorous review and is substantially complete, the USGS reserves the right to revise the data pursuant to further analysis and review. Furthermore, it is released on condition that neither the USGS nor the United States Government may be held liable for any damages resulting from its authorized or unauthorized use. Although these data have been processed successfully on a computer system at the U.S. Geological Survey, no warranty expressed or implied is made regarding the display or utility of the data on any other system, or for general or scientific purposes, nor shall the act of distribution constitute any such warranty. The U.S. Geological Survey shall not be held liable for improper or incorrect use of the data described and/or contained herein. Comma Separated Value (CSV), .shp, .zip https://doi.org/10.5066/P1OBBNGG None. No fees are applicable for obtaining the data set. 20260629 U.S. Geological Survey, Geoscience and Environmental Change Science Center Wayana Dolan Research Physical Scientist Mailing

PO Box 25046, DFC, MS 980

Denver CO 80225 US n/a n/a wdolan@usgs.gov FGDC Content Standard for Digital Geospatial Metadata FGDC-STD-001-1998