FORECASTING STREAM COMEBACKS: AI MODELS OF POST-RESTORATION WATER QUALITY IN THE NORTHEASTERN UNITED STATES
Keywords:
machine learning; turbidity; stream restoration; dam removal; culvert replacement; Random Forest; Long Short-Term Memory (LSTM); sediment stabilization; northeastern United StatesAbstract
Stream restoration projects such as dam removals and culvert replacements are accelerating across the northeastern United States to improve aquatic habitat, enhance climate resilience, and restore water quality. Yet, practitioners still lack tools to reliably forecast the outcomes of these interventions—especially in predicting changes in sediment and particulate dynamics reflected by stream turbidity. This study develops and evaluates machine learning (ML) models to forecast post-restoration turbidity recovery trajectories across a regional cohort of restored stream sites. Using publicly available data from USGS sensors, the EPA Water Quality Portal, and North Atlantic Aquatic Connectivity Collaborative (NAACC) inventories, we train two model types: a Random Forest regressor and a Long Short-Term Memory (LSTM) neural network. These models predict daily turbidity over 6 to 24 months following restoration activities. Results show that both model types significantly outperform seasonal climatology baselines, with the Random Forest achieving lower RMSE and the LSTM better capturing temporal dynamics such as storm-driven turbidity spikes and post-disturbance recovery. Feature importance and uncertainty analyses highlight the roles of precipitation, streamflow, and restoration attributes (e.g., dam removal vs. culvert replacement) in shaping water-quality improvements. The findings demonstrate that AI-based forecasting can complement traditional monitoring and support data-driven decision-making in restoration planning, particularly under variable hydrologic conditions. This modeling framework provides a scalable, transferable approach for anticipating sediment stabilization and stream recovery following restoration across diverse watersheds in the Northeast.
References
Abbott, K. M., Roy, A. H., & Nislow, K. (2025). Restoring aquatic habitats through dam removal. In Cooperator Science Series (No. CSS-148-2022). U.S. Fish and Wildlife Service. https://doi.org/10.3996/css92498424
Ali, B., . A., & Mishra, A. (2022). Effects of dissolved oxygen concentration on freshwater fish: A review. International Journal of Fisheries and Aquatic Studies, 10(4), 113–127. https://doi.org/10.22271/fish.2022.v10.i4b.2693
Barbalat, G., Hough, I., Dorman, M., Lepeule, J., & Kloog, I. (2024). A multi-resolution ensemble model of three decision-tree-based algorithms to predict daily NO2 concentration in France 2005–2022. Environmental Research, 257, 119241. https://doi.org/10.1016/j.envres.2024.119241
Bari, L. F., Ahmed, I., Ahamed, R., Zihan, T. A., Sharmin, S., Pranto, A. H., & Islam, Md. R. (2023). Potential Use of Artificial Intelligence (AI) in Disaster Risk and Emergency Health Management: A Critical Appraisal on Environmental Health. Environmental Health Insights, 17, 11786302231217808. https://doi.org/10.1177/11786302231217808
Buchanan, B. P., Sethi, S. A., Cuppett, S., Lung, M., Jackman, G., Zarri, L., Duvall, E., Dietrich, J., Sullivan, P., Dominitz, A., Archibald, J. A., Flecker, A., & Rahm, B. G. (2022). A machine learning approach to identify barriers in stream networks demonstrates high prevalence of unmapped riverine dams. Journal of Environmental Management, 302, 113952. https://doi.org/10.1016/j.jenvman.2021.113952
Feigl, M., Lebiedzinski, K., Herrnegger, M., & Schulz, K. (2021). Machine-learning methods for stream water temperature prediction. Hydrology and Earth System Sciences, 25(5), 2951–2977. https://doi.org/10.5194/hess-25-2951-2021
Guilfoos, T., & Walsh, J. (2023). A hedonic study of New England dam removals. Ecological Economics, 203, 107624. https://doi.org/10.1016/j.ecolecon.2022.107624
Hansen, H. H., Andersen, K. H., & Bergman, E. (2023). Projecting fish community responses to dam removal – Data-limited modeling. Ecological Indicators, 154, 110805. https://doi.org/10.1016/j.ecolind.2023.110805
Hart, D. D., Johnson, T. E., Bushaw-Newton, K. L., Horwitz, R. J., Bednarek, A. T., Charles, D. F., Kreeger, D. A., & Velinsky, D. J. (2002). Dam Removal: Challenges and Opportunities for Ecological Research and River Restoration: We develop a risk assessment framework for understanding how potential responses to dam removal vary with dam and watershed characteristics, which can lead to more effective use of this restoration method. BioScience, 52(8), 669–682. https://doi.org/10.1641/0006-3568(2002)052%255B0669:DRCAOF%255D2.0.CO;2
Igwaran, A., Kayode, A. J., Moloantoa, K. M., Khetsha, Z. P., & Unuofin, J. O. (2024). Cyanobacteria Harmful Algae Blooms: Causes, Impacts, and Risk Management. Water, Air, & Soil Pollution, 235(1), 71. https://doi.org/10.1007/s11270-023-06782-y
Isaak, D. J., & Young, M. K. (2023). Cold-water habitats, climate refugia, and their utility for conserving salmonid fishes. Canadian Journal of Fisheries and Aquatic Sciences, 80(7), 1187–1206. https://doi.org/10.1139/cjfas-2022-0302
Jiang, Y., Guo, C., Su, F., Xu, W., Ma, L., Cui, L., & Mi, C. (2024). Climate warming effects on temperature structure in lentic waters: A bibliometric analysis from the recent 20 years. Ecological Indicators, 167, 112740. https://doi.org/10.1016/j.ecolind.2024.112740
Kim, J., Kwak, J., Ahn, J. M., Kim, H., Jeon, J., Kim, K., Kim, J., Kwak, J., Ahn, J. M., Kim, H., Jeon, J., & Kim, K. (2022). Oscillation Flow Dam Operation Method for Algal Bloom Mitigation. Water, 14(8). https://doi.org/10.3390/w14081315
Lokman, A., Ismail, W. Z. W., Aziz, N. A. A., Lokman, A., Ismail, W. Z. W., & Aziz, N. A. A. (2025). A Review of Water Quality Forecasting and Classification Using Machine Learning Models and Statistical Analysis. Water, 17(15). https://doi.org/10.3390/w17152243
Mejia, F. H., Ouellet, V., Briggs, M. A., Carlson, S. M., Casas-Mulet, R., Chapman, M., Collins, M. J., Dugdale, S. J., Ebersole, J. L., Frechette, D. M., Fullerton, A. H., Gillis, C.-A., Johnson, Z. C., Kelleher, C., Kurylyk, B. L., Lave, R., Letcher, B. H., Myrvold, K. M., Nadeau, T.-L., … Torgersen, C. E. (2023). Closing the gap between science and management of cold-water refuges in rivers and streams. Global Change Biology, 29(19), 5482–5508. https://doi.org/10.1111/gcb.16844
Null, S. E., Mouzon, N. R., & Elmore, L. R. (2017). Dissolved oxygen, stream temperature, and fish habitat response to environmental water purchases. Journal of Environmental Management, 197, 559–570. https://doi.org/10.1016/j.jenvman.2017.04.016
Qiu, R., Wang, Y., Rhoads, B., Wang, D., Qiu, W., Tao, Y., & Wu, J. (2021). River water temperature forecasting using a deep learning method. Journal of Hydrology, 595, 126016. https://doi.org/10.1016/j.jhydrol.2021.126016
Rubin, Z., Kondolf, G. M., Rios-Touma, B., Rubin, Z., Kondolf, G. M., & Rios-Touma, B. (2017). Evaluating Stream Restoration Projects: What Do We Learn from Monitoring? Water, 9(3). https://doi.org/10.3390/w9030174
Salvatierra, D., González, M. P., Blasco, J., Krull, M., & Araújo, C. V. M. (2025). Habitat loss and discontinuity as drivers of habitat fragmentation: The role of contamination and connectivity of habitats. Environmental Research, 266, 120609. https://doi.org/10.1016/j.envres.2024.120609
Schonlau, M., & Zou, R. Y. (2020). The random forest algorithm for statistical learning. The Stata Journal, 20(1), 3–29. https://doi.org/10.1177/1536867X20909688
Spinti, R. A., Condon, L. E., & Zhang, J. (2023). The evolution of dam induced river fragmentation in the United States. Nature Communications, 14(1), 3820. https://doi.org/10.1038/s41467-023-39194-x
Trabelsi, A., Chaabane, M., & Ben-Hur, A. (2019). Comprehensive evaluation of deep learning architectures for prediction of DNA/RNA sequence binding specificities. Bioinformatics, 35(14), i269–i277. https://doi.org/10.1093/bioinformatics/btz339
Yavuz, V. S. (2025). Impact of temperature and flow rate on oxygen dynamics and water quality in major Turkish rivers. Scientific Reports, 15, 22830. https://doi.org/10.1038/s41598-025-06433-8
