Identifying post-reservoir construction periods for monotonic trend analysis at streamgages in the United States

The U.S. Geological Survey (USGS) Water Resources Mission Area (WMA) is working to address a need to understand where the Nation is experiencing water shortages or surpluses relative to the demand for water need by delivering routine assessments of water supply and demand and an understanding of the natural and human factors affecting the balance between supply and demand. A key part of these national assessments is identifying long-term trends in water availability, including groundwater and surface water quantity, quality, and use. An understanding of the impacts of reservoirs on water availability is essential for this assessment. This data release contains decadal time series of cumulative reservoir storage capacity upstream of U.S. Geological Survey streamgages in the National Water Information System (NWIS) conducted as part of the U.S. Geological Survey's Integrated Water Availability Assessment (IWAAs) Trends & Drivers project. These data were used to identify post-reservoir construction periods of record at these streamgages for Mann-Kendall gradual trend analyses of streamflow and stream temperature records contingent upon a subsequent screening for post-reservoir construction record completeness This data release builds on a prior data release containing cumulative upstream reservoir storage capacity data and flow-regulation metrics (Wieczorek et al., 2018) (https://doi.org/10.5066/F7765D7V ), the National Inventory of Dams (for dam construction dates and other information about individual dams), and the IWAAs Trends and Drivers database. In particular, the Wieczorek et al. (2018) data release provides decadal cumulative storage capacity upstream of streamgages from 1930 to 2010 as well as additional information for 2013.

We combine this decadal time series information with completion dates of individual dams to identify the last year during which a major change in reservoir storage took place. Abrupt changes in reservoir storage are identified using storage ratios, which we define as the ratio of cumulative upstream reservoir normal storage capacity to mean annual streamflow between 1980 and 2020. Decades during which the storage ratio increases by more than 10% of mean annual streamflow are identified. Next, to determine the final year of dam construction, we examine the years of completion for all dams constructed upstream of a streamgage in a given decade. Any individual dam which increases the storage ratio by more than 5% of mean annual streamflow, then the last year in which a dam producing such an increase is considered the end of the construction period and the post-reservoir discharge record starts the following year. Sites that did not experience an increase in storage ratio greater than 10% in a single decade but had cumulative increase above 10% were flagged but records were not truncated. Records at streamgages that already had a storage ratio above 10% in 1930, the first year for which decadal cumulative upstream storage capacity data from Wieczorek et al. (2018) are available, but did not exhibit any subsequent decadal increase in storage ratio above 0.10 were truncated to start in 1930.

Three key assumptions underpin this analysis:

1. Reservoirs are assumed to have been filled within a year after dam construction was completed and, thus, the start of the year of the post-reservoir construction trend period is one year after the year of completion. This assumption may not reflect the multi-year filling periods of many large reservoirs in regions with high interannual precipitation variability.
2. Identification of a single post-reservoir construction period implicitly assumes that reservoir operating policies remained consistent throughout this period, with no subsequent changes in storage following dam construction.
3. The flows at gages downstream of dams between 1980 and 2020 are assumed to be unaffected by any changes in streamflow due to dams or changes in climate and other environmental conditions. Given that dams tend to reduce mean annual streamflow through forced evaporation and water withdrawals, we assume dam impacts lower the naturally occurring mean annual streamflow. Consequently, such storage ratios will become higher if dam impacts lower mean annual flows, making our storage ratio estimates used to identify dam-impacted sites conservatively high.