Adaptive Strategies for Collaborative Watershed Management: A Regression Discontinuity Approach

• In Event: Adaptation Strategies in Collaborative Environmental Management

Abstract

Theoretical Contribution: In this research, we investigate whether and how collaborative watershed management adapts to its water quality improvement outcomes by adjusting the leadership and funding structures of restoration projects over time. This topic is critical for understanding collaborative governance as a policy implementation tool for managing common-pool natural resources. The existing literature on collaborative environmental management primarily focuses on two streams: one examining the dynamic changes in the network structures of collaborative organizations, and the other exploring the policy outcomes influenced by these networks. To address our research question, we integrate these two perspectives and offer policy actors practical adaptation strategies that respond to evolving environmental outcomes, thereby ensuring sustained long-term success.

Hypotheses: We focus on examining different network and funding structures as adaptation strategies in collaborative watershed management. We propose two hypotheses concerning adaptive leadership structures.

·      H1a: When water quality is “very poor” in a watershed, we hypothesize that the state government is more likely to lead directly and centralize the collaborative networks in restoration projects.

·      H1b: When water quality exceeds the state-required “acceptable” threshold in a watershed, we hypothesize that nonprofit watershed councils are more likely to lead, resulting in a decentralized network structure.

We further propose two hypotheses regarding adaptive funding distribution.

·      H2a: When water quality is “very poor” in a watershed, we hypothesize that the state government is more likely to provide direct funding to restoration projects.

·      H2b: When water quality exceeds the state-required “acceptable” threshold in a watershed, we hypothesize that funding sources will be more diverse and allocated among different policy actors.

Data: To test the above hypotheses, we integrate multiple data sources from 1997 to 2023 in Oregon following the enactment of the 1997 Oregon Plan for Salmon and Watersheds. Our dataset comprises 29,797 monthly water quality monitoring records from 160 sites, the Oregon Watershed Restoration Inventory, which documents over 19,000 collaborative projects, and a range of climate-related variables. We merged these datasets using watersheds’ GIS boundaries, which are based on the Hydrologic Unit Codes.

Method: We then employ a regression discontinuity design (RDD) to estimate the causal effects of water quality thresholds on changes in network and funding structures. In our RDD, the water quality index (ranging from 0 to 100) serves as the running variable, with two cutoff points: scores below 59, defined as “very poor,” and scores above 80, considered “acceptable.” Our analysis focuses on a narrow window of values immediately above and below these thresholds to isolate the causal impact of water quality on the network structure and funding variables.

Authors

• Yixin Liu, Baruch College (City University of New York)
  Presenting Author

• Xue Gao, Florida State University
  Non-Presenting Co-Author