Estimating ecological risk responses of freshwater biota to changing hydrologic volatility in the conterminous United States
This code repository has been produced by the USGS conterminous United States (CONUS) Ecological Water Availability project in support of national-scale ecological flow analyses. This workflow estimates causal dose-response relationships between Antecedent Flow Volatility (AFV) and the probability of poor ecological condition in freshwater fish and macroinvertebrate communities across the CONUS.
AFV quantifies the degree of repeated monthly flow departures from long-term natural reference conditions within the prior 12 months, measured as the log-transformed standard deviation of the monthly Natural Flow Anomalies (NFA). NFA quantifies whether observed months had greater flows (i.e., surplus) or lower flows (i.e., deficit) than expected under the predicted natural flow baseline conditions for that month. Higher AFV values indicate greater hydrologic instability and departure from typical seasonal patterns — conditions under which biological communities may face increasing risk of degradation. Please see Key Variables and Glossary below.
The workflow uses a causal inference framework to fit Average Dose Response Functions (ADRFs) that characterize how increasing levels of AFV affect the marginal probability of biological communities being in poor condition. To isolate the causal effect of AFV from confounding landscape, climate, and land-use variables (i.e., drainage area, agricultural and developed land covers, dam influence, wastewater discharge density, and ratio of prior wet to dry anomalies), the workflow calculates balancing weights (including propensity-score-style and direct optimization methods) to create a balanced quasi-experimental dataset analagous to those achieved by a randomized controlled trial. A weighted covariate-adjusted outcome model is fit, using a generalized linear models with ecoregion interaction terms, to estimate region-specific ADRFs, with significance assessed in the ADRFs via Omnibus Curve Tests using the Imhof approximation.
Citation Information
| Publication Year | 2026 |
|---|---|
| Title | Estimating ecological risk responses of freshwater biota to changing hydrologic volatility in the conterminous United States |
| DOI | 10.5066/P13WE4U7 |
| Authors | Matthew J Cashman, Taylor E. Woods |
| Product Type | Software Release |
| Record Source | USGS Asset Identifier Service (AIS) |
| USGS Organization | Water Resources Mission Area - Headquarters |
| Rights | This work is marked with CC0 1.0 Universal |