Non Technical Summary
This project assesses how wetland regulation affects US rural economic development. The project's first component builds a machine learning model to predict which wetlands the Clean Water Act regulates under the Supreme Court's 2023 Sackett decision, which created a new landscape in wetland law. The project's second component estimates the causal impact of wetland regulation on rural land markets. It analyzes impacts of wetland regulation on land values, development, and transactions. It focuses on vacant parcels in primarily rural areas, though uses residential, commercial, and industrial land uses as benchmarks. The project's third component provides a proof-of-concept investigation of how machine learning models predicting wetland regulation can support and improve that regulation. The third component investigates the potential of decision support tools to support enforcement and detection of violations of wetland regulation, and determinations of which wetlands are jurisdictional. We consider applications to state and federal wetland law. The three components have important implications for environmental and resource economics and for rural economic development. National media assert that recent Clean Water Act changes have removed regulation for over half of US wetlands. Wetlands can mitigate floods, improve water quality, and provide other important ecosystem services, and thus recent changes in regulation potentially have large consequences for the well-being of rural communities. Many ranching, farming, and other agricultural organizations have argued in published reports, media stories, amicus curiae briefs at the Supreme Court, and other domains that wetland regulation plays a central role in rural economic well-being.

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
605	0210	3010	100%

Knowledge Area
605 - Natural Resource and Environmental Economics;

Subject Of Investigation
0210 - Water resources;

Field Of Science
3010 - Economics;

Goals / Objectives
The long-term goal of this project is to assess how wetland regulation affects US rural economic development. The project has three supporting objectives:Quantify the scope of wetland regulation in the USQuantify how wetland regulation in the US affects land marketsProvide a proof-of-concept investigation of how machine learning can support wetland regulation

Project Methods
The first component, predicting Clean Water Act jurisdiction,will use a deep learning model, specifically a convolutional neural network (ResNet). It will use a held-out test set to measure accuracy, and permutation methods to assess which inputs most contribute to predictions.The second component, assessing how the Clean Water Act affects land markets, will use multivariate panel data regressions.The third component, a proof of concept of machine learning fordecision support, will explore a range of methods, including a pilot with decision support, a randomized controlled trial, and an industrial organizationmodel of enforcement impacts.