Code for training monthly natural flow models and computing Natural Flow Anomalies (NFA)
The code repository has been produced by the CONUS Ecological Flows project in support of national-scale ecological flow analyses. This workflow has two main components. First, the workflow creates a random-forest model to predict the long-term average monthly reference natural flow conditions expected for a gage (the expected, or E). Second, the natural flow predictions are then compared to given observed monthly flows (the O), determines the departure from the long-term reference conditions (the O/E ratio), also called the Natural Flow Anomaly (NFA).
The NFA, a dynamic timeseries of monthly-scale observed conditions normalized to a fixed, long-term natural baseline, helps establish a static baseline to evaluate anomalies, and potential temporal nonstationarity, against. NFA values are centered on 1 and are direct proportional values related to natural normal flows, with values less than 1 reflecting deficit conditions (i.e. lower flows than normal) and values greater than 1 representing surplus conditions (i.e. greater flows than normal). NFA at the monthly scale can be driven by climate, landcover, water use, or other water-management factors which affect the availability of water for a given month.
Although approached and calculated somewhat differently, the NFA values created in this workflow are conceptually interchangeable with the monthly scale Surface Water Supply and Use Index (SUI) used in the National Water Availability Assessment, where the NFA = (1-SUI), but encompass both surplus and deficit anomalies, compared to only deficits in the SUI.
The NFA values are then used in subsequent causal analyses to create dose response function to evaluate changing ecological risk of poor conditions due to changes in different aspects of flow regime alteration. These analyses are contained in separate workflows, associated with any associated derivative publication(s).
Data produced by this workflow are available in the Companion Data Release available on ScienceBase.
Citation Information
| Publication Year | 2026 |
|---|---|
| Title | Code for training monthly natural flow models and computing Natural Flow Anomalies (NFA) |
| DOI | 10.5066/P14DDURW |
| Authors | Matthew J Cashman, Daren M Carlisle |
| 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 |