A simplified method to classify streams and improve California’s water management

by Belize Lane, Sam Sandoval, and Sarah Yarnell

Alterations to the natural flow regime for human water management activities have degraded river ecosystems worldwide. Such alterations are particularly destructive in regions with highly variable climates like California, where native riverine species are highly adapted to natural flooding and drought disturbances. In California, less than 2% of the total streamflow remains unaltered, while over 80% of the native fish species are now imperiled or extinct .

Determining the natural flow regime for altered stream reaches is difficult as unimpaired streamflow records are unavailable for many locations of interest. Where data is available, previous methods distinguished such specific stream types that their application was limited and unhelpful for regional management. To improve California’s water management, particularly around determining environmental flows for our diverse ecosystems, we needed a better method that addressed the diversity and scale of California’s streams.

Hydrologic Classification

Hydrologic classification is a strategy for distinguishing groups of stream reaches with similar streamflow characteristics for regional water management efforts. UC Davis researchers recently developed a hydrologic classification for California that is specific enough to make critical distinctions between natural streamflow patterns (also called natural flow regimes), but general enough to support the development of environmental flow targets in altered stream reaches across the state.

The California hydrologic classification is based on available hydrologic and geospatial data. First, key hydrologic metrics (e.g., measures of streamflow magnitude, duration, timing, frequency, and rate of change) pertinent to river ecosystems were calculated for all available reference streamflow gauge stations with long-term (>20 years) unimpaired or naturalized discharge data. These metrics were input to an initial streamflow gage classification model that distinguished statistically distinct natural stream classes; each reference streamflow gage was classified into a stream class based on its specific hydrologic metric values.

Then, a predictive linear regression model was developed based on relationships between the initial streamflow gauge classification and upstream catchment attributes (e.g., climate, topography, soils, and geology). The model was highly accurate at predicting the stream classes of reference gauges and performed well compared to regional hydrologic studies. This second model (Figure 1) was then used to predict the stream classes of all the reaches in California.

Predictive model of stream classes based on upstream catchment attributes, from Lane et al. (2017a).

The resulting hydrologic classification (Lane et al. 2017b) identified nine natural stream classes (Figure 2) with distinct streamflow patterns that are the result of several characteristics of rainfall-runoff response, including: dominant water source (snowmelt, rain, groundwater), hydrologic attributes (mean annual flow, extreme low flow duration and timing, etc), climate setting (mean annual precipitation, mean August and January precipitation, etc.) and topographic and geologic setting (slope, catchment area, dominant rock type, soils compositions, etc.).

Hydrologic classification of California, combing results of Lane et al. (2017a) and Pyne et al. (2017).

California’s Natural Stream Classes

Snowmelt (SM): SM streams exhibit highly seasonal flow regimes with spring snowmelt peak flows, predictable recession curves, very low summer flows, and minimal winter rain influence. These sites exist along the crest of the Sierra Nevada with most sites in the southern, higher elevation portion of the mountain range.

High elevation, Low Precipitation (HLP): HLP streams are distinguished from SM streams by their higher base flow (due to porous geology) and lower peak flows (due to less snow) but exhibit a similar seasonal signature and predictability.

High- and Low-volume Snowmelt and Rain (HSR and LSR): The transition from a SM to a LSR to a HSR regime closely tracks the elevation gradient from the peaks of the Sierra Nevada to the floor of the Central Valley. LSR and HSR streams exhibit similar bimodal snow-rain patterns but illustrate a transition toward earlier snowmelt peak and increasing winter rain contributions along the elevation gradient.

Rain and seasonal Groundwater (RGW): Generally at lower elevations, RGW streams exhibit higher minimum flows and earlier summer peak flows than LSR streams, as well as the distinctive influence of winter rain storms in high and unpredictable winter flows.

Winter Storms (WS): WS streams, driven by winter rain storms, exhibit distinct duration, timing and magnitude of high flows during the rainy season. They are characterized by high interannual flow variance, due to the variability of winter storm patterns, and very low base flows during summer. WS streams generally follow the spatial distribution of strong orographic precipitation in the north coast region.

Groundwater (GW): GW streams are distinguished by significantly higher and more stable flows year-round, mostly located along volcanic geologic settings.

Perennial Groundwater and Rain (PGR): PGR streams combine the stable, base flow-driven conditions of GW streams during summer with the high magnitude winter peak flows of WS streams in catchments with low annual streamflow.

Flashy Ephemeral Rain (FER): Prevalent in arid southeastern California, FER streams are characterized by the highest interannual flow variance, extended extreme low flows and large floods, and the lowest average daily flow of any class.

The following figure (Fig. 3) illustrates the extreme seasonal and interannual hydrologic variability between stream classes. For example, the SM flow regime exhibits a highly predictable spring snowmelt pattern with low interannual variability (<6) while the WS flow regime exhibits highly variable winter storm flows (<18) and very low summer flows.

Reference dimensionless hydrographs illustrate seasonal and interannual flow regime variability between natural stream classes for four of the nine classes (Lane et al. 2017b). Reference streamflow gauge time-series were aggregated for each stream class and daily streamflow was non-dimensionalized based on average annual streamflow to highlight pattern variability.

Environmental Water Management Implications

This hydrologic classification provides the fundamentals for understanding the diversity of natural streamflow patterns and their spatial arrangement across the state. It also supports the need for broad-scale environmental management of California’s many impaired rivers. The spatial extent and reach scale of the classification are expected to substantially improve the overlap of biological and hydrologic datasets statewide. The hydrologic classification provides a footprint of the locations of distinct natural stream classes which, combined with ecological and geomorphic information, can be used to design environmental flow targets. Future comparisons of ecological patterns between natural and hydrologically altered streams within each stream class are expected to yield flow-ecology relationships that can provide the basis for rapid statewide environmental flow standards.

Belize Lane recently received her PhD in Hydrologic Sciences from UC Davis and is now an Assistant Professor in Civil and Environmental Engineering at Utah State University. Samuel Sandoval is an Associate Professor in the Dept. of Land, Air and Water Resources and UC Agricultural and Natural Resources Cooperative Extension Specialist. Sarah Yarnell is a senior researcher at the Center for Watershed Sciences.

Further reading

Lane BA, Dahlke HE, Pasternack GB and Sandoval-Solis S (2017a). Revealing the Diversity of Natural Hydrologic Regimes in California with Relevance for Environmental Flows Applications.

Lane BA, Sandoval-Solis S, Yarnell SM, Stein ED (2017b) Characterizing diverse river landscapes using hydrologic classification and dimensionless hydrographs. In Preparation.

Magilligan FJ and Nislow KH (2005). Changes in hydrologic regime by dams. Geomorphology.

Pyne MI, Carlisle, DM, Konrad CP and Stein ED (2017). Classification of California streams using combined deductive and inductive approaches: Setting the foundation for analysis of hydrologic alteration.

Quiñones RM, Moyle PB (2015) California’s freshwater fishes: status and management. FISHMED

This entry was posted in California Water, Planning and Management, Tools and tagged . Bookmark the permalink.

2 Responses to A simplified method to classify streams and improve California’s water management

  1. Pingback: California Water News for July 18, 2017

  2. Description of a project to downscale this approach to Southern CA for ecological flows/outcomes, along with the data (well documented) is available here, fyi: https://waterboards.maps.arcgis.com/apps/MapJournal/index.html?appid=e29ee72a7964412bb2704eec25520c17


Leave a Comment (moderated)

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s