Junyu Qi

Sustainable agriculture in water-limited regions requires understanding how future climate change may affect crop water use and productivity and whether locally feasible adaptations can offset adverse impacts. This study used an improved SWAT (Soil and Water Assessment Tool) model to simulate multiple future climate scenarios—five single-factor adaptation scenarios, including early or late sowing, long- or short-season cultivars, and crop substitution, and one integrated scenario—in the Palo Duro Watershed of the Texas High Plains. Under future climate, yields declined by 6.7–13.1% for irrigated corn (Zea mays L.) and 11.6–34.1% for irrigated sorghum (Sorghum bicolor L.), but increased by 8.0–32.8% and 8.4–33.7% for dryland and irrigated winter wheat (Triticum aestivum L.), respectively. Early sowing improved yields and water productivity (WP) for irrigated corn, irrigated sorghum, and dryland winter wheat. However, long-season cultivars increased corn and sorghum yields and WP with greater irrigation water use. Crop substitution was generally unfavorable under local conditions. A combined strategy of early sowing and long-season cultivars would increase yields and WP for irrigated corn, irrigated sorghum, and dryland winter wheat, whereas business-as-usual management remained appropriate for irrigated winter wheat. These findings support crop-specific adaptation planning for climate-resilient agriculture and water-resource management in semi-arid systems.

Due to ongoing global warming, the frequency and intensity of extreme weather events have increased substantially. Compared to individual extremes, compound drought and high temperature (CDHT) events represent a major climate risk in China. However, their spatiotemporal characteristics remain insufficiently understood, particularly at fine temporal scales. To address this gap, this study systematically investigated CDHT events across China from 1982 to 2023. Methodologically, CDHT events were identified at the raster level by combining an improved daily Standardized Precipitation Evapotranspiration Index (SPEI) with daily maximum temperature using a quantile relative dynamic threshold. The results show strong spatial heterogeneity: the longest event durations are primarily observed in Xizang, while higher event severity is concentrated in regions south of 30° N. Trend analysis reveals a widespread increase in the duration, frequency, and severity of CDHT events across most of China, with the most pronounced intensification detected in Xinjiang, Inner Mongolia, and Yunnan. Overall, these findings highlight a clear climate-driven intensification of CDHT events, offering new insights into their spatiotemporal dynamics. The results offer a robust scientific basis for improving risk assessment and developing targeted adaptation strategies to mitigate the impacts of compound climate extremes in China.

Accurate representation of soil moisture dynamics is essential for reliable hydrologic prediction, yet watershed-scale models are often calibrated using streamflow alone, leaving key terrestrial processes weakly constrained. Advances in satellite remote sensing provide new opportunities to incorporate soil moisture and ecohydrological variables into multi-objective calibration. Here, we assess how integrating satellite soil moisture improves hydrologic simulations in the Upper Mississippi River Basin (UMRB), a large and heterogeneous agricultural–forest watershed. Building on a previously MODIS-calibrated Soil and Water Assessment Tool (SWAT) framework that incorporated evapotranspiration (ET), leaf area index (LAI), and net primary productivity (NPP), we further constrained the model using NASA SMAP Level 4 surface (0-5 cm) and root-zone (0-100 cm) soil moisture during the growing season. We compared three model configurations: an uncalibrated baseline (BASE), a MODIS-only calibrated model, and a MODIS+SMAP calibrated model. Incorporating SMAP substantially improved surface soil moisture simulations, increasing R2 and KGE and generally reducing bias in many subbasins relative to MODIS-only across both cropland- and forest-dominated subbasins. In contrast, root-zone soil moisture improvements were driven mainly by better temporal dynamics (higher R2 and KGE), while bias responses were mixed across subbasins. These improvements propagated to streamflow, enhancing performance across 43 USGS gauges spanning drainage-area classes from HUC-4 to HUC-12, with the MODIS+SMAP configuration showing the strongest gains in NSE and KGE and improved long-term water-balance behavior. Spatially, the largest streamflow improvements occurred in areas where upstream soil moisture skill increased the most, supporting a process-level link between improved soil moisture representation and runoff generation. Importantly, adding SMAP constraints did not degrade–and in some cases improved–simulations of ET, LAI, and NPP, indicating that soil moisture information can be integrated without compromising ecosystem process fidelity. Overall, our results demonstrate that combining SMAP soil moisture with MODIS terrestrial products in a multi-objective calibration framework improves watershed-scale hydrologic realism and strengthens ecohydrological assessments across diverse land-use settings, highlighting the value of current and emerging soil moisture missions for large-basin water-resources modeling.

Ecohydrological models are critical for understanding coupled hydrologic–biogeochemical processes in tile-drained watersheds and for assessing management options. Despite recent advances in SWAT’s hydrological and biogeochemical processes, there has been limited evaluation of both the original and new tile drainage and nitrogen (N) modules. We therefore applied a comparative modeling approach in a typical Midwestern tile-drained watershed, evaluating eight configurations that vary tile-drainage module (original/new), tile parameter treatment (calibrated/default), and N module (original/new) to assess performance for N-loss simulation. Using daily streamflow and nitrate (NO ₃ ⁻ ) load records from three monitoring sites, we conducted calibration, validation, sensitivity analysis, and uncertainty assessment. Each configuration effectively reproduced daily and monthly dynamics, although high-flow and associated NO ₃ ⁻ load peaks were underestimated. We found that the new tile module generally improved streamflow simulations, particularly under tile parameter calibration conditions, while the new N module consistently enhanced NO ₃ ⁻ load simulations compared to the original module. Despite improvements in streamflow and NO ₃ ⁻ loads with the new tile and N modules, the additional processes in the new N module can magnify uncertainty in N-gas-flux estimates when calibration observations are scarce. We recommend applying the new N module in conjunction with additional measurements—such as soil moisture and nitrous oxide (N ₂ O) fluxes—to constrain better N gas flux estimates beyond outlet NO ₃ ⁻ load data. If such observations are unavailable, careful calibration with reasonable estimates may still help constrain soil N cycling and improve overall N budget accuracy.

Tile drainage, a common agricultural practice in the Midwest USA, improves soil aeration and crop yields but also contributes to environmental challenges, such as nitrate (NO₃) loss and nitrous oxide (N₂O) emissions. To address these issues, ecohydrological models are essential for understanding the intricate hydrological and biogeochemical processes in tile-drained watersheds and for evaluating management strategies. Recent enhancements to the Soil and Water Assessment Tool (SWAT) have incorporated Century-based soil carbon and nitrogen cycling processes, along with N₂O emission algorithms, improving its ability to simulate nitrogen cycling and greenhouse gas emissions at the watershed scale. In this study, the Century-based soil carbon and nitrogen cycling module was integrated with two tile drainage modules within SWAT to enhance its simulation capabilities. The enhanced model was first evaluated at the Kelley experimental site, using observed data on drainage discharge, NO₃ loss, and N₂O emissions influenced by cover crops, corn-soybean rotation, and fertilization. Subsequently, the model was applied to simulate NO₃ loss and N₂O emissions in Iowa’s South Fork Watershed (SFW) using an ensemble modeling approach. This approach tested eight scenarios combining the two nitrogen modules, the two tile drainage modules, and calibration variations. Results showed that all eight scenarios effectively simulated daily stream flow but underestimated daily NO₃ load due to the underrepresentation of peak flows. Most models performed well at the monthly scale for both stream flow and NO₃ load. The ensemble modeling results aligned with prior studies, suggesting that ensemble approaches can reduce prediction uncertainties and address equifinality issues. However, the study emphasizes the need for additional data collection to improve the accuracy of denitrification and N₂O emissions simulations, especially in tile-drained systems. This research advances ecohydrological modeling for tile-drained watersheds, offering insights into improving water quality and reducing greenhouse gas emissions under agricultural management practices.