About

Professor Maria L. Chu leads the Watershed-Ecosystem Research Laboratory (WERL) within the Department of Agricultural and Biological Engineering at the University of Illinois Urbana-Champaign. Her research is dedicated to unraveling the complex interactions between natural and human factors and their effects on ecosystems in a rapidly changing environment. By integrating diverse data sources such as remote sensing, ground-based measurements, and extensive databases, she develops data- and computationally-intensive models to better understand watershed-ecosystem dynamics, particularly in intensively managed landscapes. Her work focuses on several key areas including the fate and transport of contaminants like sediment, nutrients, and pesticides within these landscapes. She also emphasizes the development and integration of models, algorithms, and systems to analyze ecosystem responses to environmental changes such as climate variability and land-use alterations. Additionally, Professor Chu evaluates model performance through global uncertainty and sensitivity analyses as well as pattern identification techniques. Her research further explores soil and water interactions on large scales, approached from local perspectives, and involves hydrologic time series analysis to deepen understanding of watershed processes.

Research topics

• Environmental science
• Geography
• Ecology
• Computer Science
• Water resource management
• Environmental engineering
• Soil science
• Environmental resource management
• Engineering
• Business
• Geology
• Environmental planning

Selected publications

• Model-based estimation of the isolated impacts of urban expansion on projected streamflow values under varied climate scenarios

  Computers Environment and Urban Systems · 2025-02-16 · 1 citations

  articleOpen access

  The use of computationally intensive hydrologic models under future climate scenarios has become a common practice to project water resource concerns in the coming decades. Under this approach, hydrologic models are parameterized and run under various climate forcings. Although urban areas are expected to expand during the time frame of these simulations, potentially impacting watershed hydrology, the uncertainty of forecasted streamflow is usually estimated based on the ensemble of climate scenarios, with minimal (if any) attention given to the uncertainty introduced by land transformations. The objective of this study is to quantify the Isolated Impacts on Projected Streamflow (IIPS) caused by urban expansion as climate changes in a watershed in the midwestern United States. IIPS time series were estimated as the difference between projected streamflows under future climate scenarios with and without urban expansion and weighted by the historical (1980–2010) monthly average. Two gradual and two abrupt urbanization scenarios, having equivalent developed areas by the end of the 21st century, were implemented. Results indicate that gradual urbanization could result in both increased (up to 26 %) and decreased (up to 16 %) projected streamflows, suggesting the increase in variability of extremes, with potential impacts on human and natural systems. Yearly minimum and maximum IIPS for all scenarios were found to be more likely to occur in summer and fall months, respectively. Impacts of the abrupt urban expansion were mainly observed in the cumulative IIPS and the ensemble variability of extreme IIPS. These results provide insights into the uncertainty of future streamflow estimates attributable to urban expansion. • Isolated Impacts on Projected Streamflow (IIPS) due to urbanization were estimated. • Larger urbanization rates increased the frequency of extreme (+ and -) IIPS in time. • Maximum and minimum IIPS will likely occur in fall and summer months, respectively. • Projected IIPS surpassed the critical threshold for ecological protection (10 %). • Abrupt urbanization approach intensified IIPS and its climate ensemble variability.

  PublisherDOI

• Enhancing corn leaf fiber as phosphorus adsorbent material

  Water Environment Research · 2025-04-29 · 2 citations

  articleOpen access

  Abstract The contribution of dissolved phosphorus (P) from tile drain systems in agricultural lands is significant, leading to water impairment and promoting algae bloom development in water bodies. Hence, there is an urgent need for sustainable and efficient technology, such as absorbent material, that can effectively remove P at low concentrations in these systems. This study aimed to evaluate fiber extraction from corn leaves and its potential for reducing dissolved P. Corn fibers were extracted from corn leaves using alkali treatment by varying the concentration of sodium hydroxide (5–15%w/w), extraction temperature (85–95°C), and time (60‐120 min Results of the alkali extraction showed that the highest fiber recovery of 45.18 ± 0.39% g g −1 was achieved at 10% NaOH at 85°C for 60 min condition. To enhance the phosphorus adsorption capacity of the extracted corn fibers, kaolinite clay (0–30% w/w) and calcium carbonate (0–50% w/w) were incorporated into the leaf fiber. Adsorption tests revealed that corn leaf fiber alone reduced phosphate concentration by 8.75 ± 1.49% within 60 minutes. However, when enhanced with 30% w/w kaolinite clay and 35% w/w calcium carbonate, the phosphate concentration in the solution decreased by 79.40 ± 11.90%. Energy‐dispersive X‐ray fluorescence analysis confirmed the presence of phosphorus in the enhanced adsorbent material following treatment. This study demonstrates the potential of enhancing agricultural wastes like corn leaf fiber as a low‐cost alternative for phosphorus removal in agricultural tile drain systems that can later disposed of as fertilizer in a circular economy scheme. Practitioner Points Corn leaf fiber, extracted using alkali treatment, shows potential as a sustainable, low‐cost adsorbent material for phosphorus removal Enhancing corn leaf fiber with kaolinite clay and calcium carbonate significantly improves phosphate reduction Converting agricultural waste like corn leaf as adsorbent material can help manage phosphorus levels in agricultural tile drains

  PublisherOA PDFDOI

• Prediction of gully erosion susceptibility through the lens of the SHapley Additive exPlanations (SHAP) method using a stacking ensemble model

  Journal of Environmental Management · 2025-04-25 · 11 citations

  articleOpen accessSenior author

  This study develops a novel explainable stacking ensemble model that combines the stacked generalization ensemble method with SHapley Additive exPlanations (SHAP) to enhance the prediction and interpretation of gully erosion susceptibility. Applied to Jefferson County, Illinois, our approach leverages Random Forest (RF), Gradient Boosting Machine (GBM), Logistic Regression (LR), and Deep Neural Networks (DNN) as both base and meta-learners in various configurations, resulting in 44 distinct stacking models. The comparative analysis demonstrated the superior predictive performance of the stacked models when evaluated at 200 randomly gully sites selected points based on LiDAR difference observations; all but three exceeded the highest area under the curve (AUC) value of 0.86 achieved by the best-performing base model (GBM). The LR stacking model, combining RF and GBM as base models with LR as the meta-learner, emerged as the most effective, achieving an AUC of 0.916. The resulting gully erosion susceptibility map by the LR stacking model classified 33 % of the agricultural land (89,208 ha) as the "very high" class, compared to 27 %, 87 %, 27 %, and 55 % predicted by individual RF, LR, GBM, and DNN models, respectively. Crucially, SHAP analysis elucidated how changes in feature values influence model behavior, considering feature interactions within both the base models and the meta-learner. The SHAP identified the annual leaf area index (LAI) as the most influential feature in both RF and GBM base models. Additionally, it highlights the significance of the GBM model in comparison to the RF base model in the final decision-making process of the stacking model. By offering a transparent mechanism to evaluate how different features and models contribute to final decisions, this approach can be extended to broader environmental management and policy-making contexts, facilitating more informed and responsible resource allocation.

  PublisherDOI

• Dynamic land cover evapotranspiration model algorithm: DyLEMa

  Computers and Electronics in Agriculture · 2024-03-27

  articleOpen accessSenior author

  • Among decision tree-based MLs, RF Performance compared better than CART and XGB. • DyLEMa can reconstruct ET based on seasonal segregated atmospheric and land data. • DyLEMa ET PBIAS was reduced in temporal validation compared to the USGS ET data. • DyLEMa ET estimates are robust across various cloud contamination rates. • DyLEMa ET estimates are robust across Landsat sensor failure. This study presents the “Dynamic Land Cover Evapotranspiration Model Algorithm: DyLEMa” for continuous spatiotemporal evapotranspiration (ET) estimates across diverse land uses. DyLEMa employs a coupled Random Forest model with a novel dynamic recalibrating strategy to improve pre-optimized seasonal hyperparameters following satellite acquisition alongside land cover classes. An analysis of feature importance indicated the significant variability in ET processes across different land cover classes and seasons. Hence, DyLEMa was applied to 20 years of daily 30x30 m pixel resolution Landsat-derived ET data in Illinois to address spatial and temporal discontinuities due to cloud contamination and sensor failures. DyLEMa performance was evaluated on Eddy Covariance measurements to find out that DyLEMa predictions reduced the average PBIAS error from + 31 % to −7% compared to existing US Geological Survey ET products. Spatially, DyLEMa underscores the value of a land cover-aware approach in ET estimation under varied cloud cover rates and their ability to preserve landscape features. However, the performance of DyLEMa was affected by the quality of land cover classification, suggesting the need for a refined region-specific land cover classification. DyLEMa’s flexibility and performance suggest its applicability to other regions and satellite datasets, offering a promising reduction in uncertainty of ET estimates with impacts on environmental and water resources assessments on regional scales.

  PublisherDOI

• Understanding the spatial dynamics of values and disvalues in the Kaskaskia River Watershed, USA through a social-ecological lens

  Ecosystems and People · 2024-02-15 · 4 citations

  articleOpen access

  Maintaining a resilient and sustainable agro-production system is replete with challenges, because management practices designed to enhance productivity can overlook the range of values derived from ecosystems and concerns about their future. Further complicating the decisions being made about agricultural settings is the variation in social-ecological stressors that shape how diverse community members interpret landscape change. We engaged residents of the Kaskaskia River Watershed in Illinois, USA in discussions about agroecosystems through participatory mapping exercises and focus groups. The spatially explicit data that were derived from this process were then modeled in relation to changes in watershed hydrology simulated using a Soil and Water Assessment Tool (SWAT) model. We observed that the four most salient ‘values’ associated with the watershed were recreation, erosion protection, crop production, and flood control. Erosion, siltation and sedimentation, increased flooding, and invasive species were considered the most relevant ‘disvalues’ because they represented the social-ecological stressors of greatest concern. Respondents believed that the values from ecosystems were more spatially clustered than disvalues, and the main river corridor was at greater risk of degradation than the associated tributaries despite these second order streams being more biologically diverse. The use of participatory mapping data coupled with SWAT to simulate changes in systemic responses of the watershed provided a social-ecological basis for identifying high and low-priority locations at a regional scale. Our results therefore aim to spatial prioritize and guide evidence-based decisions anchored in the social and ecological complexities of a Midwestern watershed.

  PublisherOA PDFDOI

• Modeling the impact of measured and projected climate and management systems on agricultural fields: Surface runoff, soil moisture, and soil erosion

  Journal of Environmental Quality · 2024-05-11 · 7 citations

  articleOpen access

  As global climate change poses a challenge to crop production, it is imperative to prioritize effective adaptation of agricultural systems based on a scientific understanding of likely impacts. In this study, we applied an integrated watershed modeling framework to examine the impacts of projected climate on runoff, soil moisture, and soil erosion under different management systems in Central Oklahoma. The proposed model uses measured climate data and three downscaled ensembles from the Coupled Model Intercomparison Project Phase 6 (CMIP6) at the water resources and erosion watershed to understand the impact of climate change and various climate conditions under three management systems: (1) continuous winter wheat (Triticum aestivum) under conventional tillage (WW-CT; baseline system), (2) continuous winter wheat under no-till (WW-NT), and (3) cool and warm season forage cover crop mixes under no-till (CC-NT). The study indicates that the occurrence of agricultural drought is projected to increase while erosion rates will remain unchanged under the WW-CT. In contrast, climate simulations imposed on the WW-NT and CC-NT systems significantly reduce runoff and sediment while preserving soil moisture levels. Especially, implementing the CC-NT system can bolster food security and foster sustainable farming practices in Central Oklahoma in the face of a changing climate.

  PublisherOA PDFDOI

• Prediction of Gully Erosion Susceptibility Through the Lens of the Shapley Additive Explanations (Shap) Method Using a Stacking Ensemble Model

  SSRN Electronic Journal · 2024-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Transport of nitrogen in grassed watersheds accounting for the combined influence of grazing and climate

  Ecological Modelling · 2024-08-08 · 2 citations

  articleOpen accessCorresponding

  • Grazing's impact on nitrogen loss varies with the prevailing weather conditions. • Grazing, aligned with climate, minimizes nitrogen loss, boosts livestock production. • A distributed overland flow model better represents grazing's nutrient loss impact. A comprehensive modeling framework utilizing MIKE-SHE to investigate the feasibility of enhancing livestock production while concurrently mitigating the impact of grassland ecosystems on surface water quality is presented. In this study, a modeling framework that simulates nitrogen transport was developed using a physically based distributed hydrologic model, MIKE SHE. The model was calibrated and validated using observed overland flow and water quality data from El Reno, Oklahoma's Water Resources and Erosion Unit watersheds. Different scenarios involving variations in timing, duration, and frequency of grazing activities, stocking rates, and climate conditions were simulated using the calibrated MIKE-SHE model. The goal was to explore the extent of the influence of these aspects of grazing on water quality through these scenario simulations. The results provided valuable insights into the key factors for developing an intelligent grazing decision tool to delineate targeted grazing windows that optimally balance heightened livestock production with environmental sustainability. The study's findings showed that the observed variability in existing literature originates mainly from climatic differences, with precipitation being the primary driver of nutrient loss, while evapotranspiration and soil moisture conditions are secondary factors. The simulations revealed that the impact of grazing on nitrogen loss in the pasture is evident only when grazing activities, irrespective of the stocking rate, duration, and frequency, occur under weather conditions conducive to nutrient loss in the pasture. These intertwined processes suggest that the impact of grazing on nitrogen loss can only be generalized within the context of the prevailing weather conditions in the pasture. Hence, strategically matching grazing activities with prevailing weather patterns can increase livestock production while promoting environmental sustainability in pasture management.

  PublisherDOI

• Pride and guilt as place-based affective antecedents to pro-environmental behavior

  Frontiers in Psychology · 2023-01-19 · 18 citations

  articleOpen access

  The interrelated concepts of place attachment and place meaning are antecedents to pro-environmental behavior and essential for supporting decisions that foster relationships between people and places. Previous research has argued that affect is instrumental in conceptualizing place-related phenomena but has not yet been considered in terms of discrete emotions. We disentangled the empirical relationships between concepts of place and the emotions of pride and guilt to understand how they collectively contributed to individuals' decisions about environmental sustainability. Specifically, we conducted an online survey of residents living in the Midwestern US and asked questions about their attachments to places and their place-related behavior. We then tested a latent variable path model with first- and second-order factors that shaped the behavioral intentions of survey respondents, as well as evaluated the psychometric properties of a place meaning scale, to uncover the range of reasons why human-nature relationships were formed. Our findings show that multiple place meanings predicted place attachment, which in turn predicted the discrete emotions of pride and guilt. Place attachment, pride, and guilt positively correlated with pro-environmental behavior. We also observed that the relationships between multi-dimensional conceptualizations of place attachment and behavioral intentions were partially mediated by pride but not guilt, as hypothesized in response to the broaden and build theory of positive emotions. This study develops theoretical insights to clarify how cognitive-emotional bonding can lead people to behave in more environmentally friendly ways.

  PublisherOA PDFDOI

• Gully erosion susceptibility mapping using the stacking ensemble machine learning method

  2023-01-01

  articleSenior author

  <b><sc>Abstract.</sc></b> Soil erosion poses a worldwide challenge, significantly impacting water and soil resources. Gully erosion, in particular, creates large channels that are difficult to remediate using conventional agricultural equipment, leading to substantial socio-economic losses. Identifying areas vulnerable to gully erosion is crucial for devising preventive action plans that reduce gully formation and associated damages, ultimately contributing to the achievement of sustainable development goals. However, the environmental factors linked to gully development, such as weather, vegetation, land management, and slope, are complex and vary across time and space. This research explores the potential of using a stacking ensemble learning technique for predicting gully erosion susceptibility. A dataset comprising 1000 gully erosion locations and 38 environmental factors in agricultural land in Jefferson County, Illinois, U.S., was employed to develop the stacking model for gully prediction. Collinearity detection involved calculating the correlation coefficient and Cramer's V, followed by feature selection based on mutual information. The chosen features were divided into training, validation, and test sets to establish a robust model and facilitate a more generalized performance evaluation. Random Forest (RF), K-Nearest Neighbors (KNN), Gradient Boosting Machines (GBM), and Deep Neural Networks (DNN) were employed as base models. The optimization of each base model was achieved through a combination of K-fold cross-validation and Bayesian optimization. Although RF, KNN, GBM, and DNN models demonstrated comparable performance, the DNN emerged as the best meta-model, exhibiting the highest accuracy. Furthermore, a comparison between base models and meta-models revealed that the stacking-based ensemble model outperformed single algorithms in terms of predictive accuracy. The findings of this study indicate that the stacking ensemble machine learning method holds promise for predicting gully erosion susceptibility in the study area. This information can aid policymakers and land managers in concentrating their prevention or mitigation efforts and allocating resources where they are most needed.

  PublisherDOI

Frequent coauthors

• Jorge A. Guzmán

  University of Illinois Urbana-Champaign

  39 shared

• Matteo Convertino

  23 shared

• Igor Linkov

  U.S. Army Engineer Research and Development Center

  21 shared

• Glenn V. Wilson

  19 shared

• Rafael Muñoz‐Carpena

  University of Florida

  18 shared

• Garey A. Fox

  North Carolina State University

  17 shared

• Gregory A. Kiker

  University of Florida

  17 shared

• Richard A. Fischer

  15 shared