Future streamflow regime changes in the United States: assessment using functional classification

Abstract

Abstract. Streamflow regimes are changing and expected to further change under the influence of climate change with potential impacts on flow variability and the seasonality of extremes. However, not all types of regimes are going to change in the same way. Climate change impact assessments can therefore benefit from identifying classes of catchments with similar streamflow regimes. Traditional catchment classification approaches have focused on specific meteorological and/or streamflow indices usually neglecting the temporal information stored in the data. The aim of this study is two-fold: (1) develop a catchment classification scheme that allows for the incorporation of such temporal information and (2) use the scheme to evaluate changes in future flow regimes. We use the developed classification scheme, which relies on a functional data representation, to cluster a large set of catchments in the conterminous United States (CONUS) according to their mean annual hydrographs. We identify five regime classes that summarize the behavior of catchments in the CONUS: (1) Intermittent regime, (2) weak winter regime, (3) strong winter regime, (4) New Year's regime, and (5) melt regime. Our results show that these spatially contiguous classes are not only similar in terms of their regimes, but also their flood and drought behavior, as well as their physiographical and meteorological characteristics. We therefore deem the functional regime classes valuable for a number of applications going beyond change assessments including model validation studies or the prediction of streamflow characteristics in ungauged basins. To assess future regime changes, we use simulated discharge time series obtained from the Variable Infiltration Capacity hydrologic model driven with meteorological time series generated by five general circulation models. A comparison of the future regime classes derived from these simulations with current classes shows that robust regime changes are expected only for currently melt-influenced regions in the Rocky Mountains. These changes in mountainous, upstream regions may require the adaptation of water management strategies to ensure sufficient water supply in dependent downstream regions.

Flood hazard and change impact assessments may profit from rethinking model calibration strategies

“…Flood events are identified for each of the five time series (one observed, four simulated) using a peak-over-threshold (POT) approach similar to the one used in Brunner et al (2019aBrunner et al ( , 2020b. This approach consists of two main steps and results in two data sets each, which are used for the local and spatial analysis, respectively: (1) POT events in individual catchments and 2event occurrences across all catchments.…”

“…a) Map of the 488 catchments in the conterminous United States belonging to the five regime classes: 1) Intermittent, 2) weak winter, 3) strong winter, 4) New Year's, and 5) melt. b) Median regime per regime class (colored lines) and variability of regimes within a class (on line per catchment, grey) (Brunner et al, 2020b). identifying and documenting model weaknesses regarding regional and future flooding will highlight avenues for future model development.…”

How Probable Is Widespread Flooding in the United States?

“…On a seasonal scale, different regimes show high and low regional flood susceptibilities in dependence of their main season of flood occurrence. In winter, susceptibility is high for regions around catchments with a New Year's regime, that is, a streamflow regime characterized by high discharge around the time of the new year, which is mainly found in the Pacific Northwest (Brunner, Newman, et al, 2020). In contrast, susceptibility is close to zero for catchments with a melt regime, where precipitation is accumulated as snow.…”

“…We compute annual and seasonal regional flood hazard for 18 large hydrological regions in the United States from a river basin perspective and from a local perspective using a data set of 671 catchments (Figure 1) with minimal human flow alteration (Catchment Attributes and MEteorology for Large‐sample Studies data set CAMELS; Newman et al, 2015). The data set comprises catchments with a wide range of streamflow regimes including intermittent regimes with a weak seasonality mainly located in the Great Plains; regimes governed by winter precipitation showing a strong seasonality, for example, in the Appalachian Mountains; and melt regimes with a spring melt peak mainly located in the Rocky Mountains (Brunner, Newman, et al, 2020). The catchments are almost exclusively small headwater catchments with a very low degree of nestedness which avoids overestimating spatial dependencies.…”

“…In addition, we identify factors influencing regional flood susceptibility among (1) the hydrological regime type (Brunner, Newman, et al, 2020), (2) flood characteristics, and (3) physiographical and climatological catchment characteristics in a correlation analysis. Flood characteristics considered are the mean and standard deviation of the date of occurrence and mean magnitude, volume, and duration.…”