Assessing Spatially Distributed Temperatures for Water Suitability and Habitat

Project ID: 7872
Principal Investigator: Yong Lai
Research Topic: Water Quality
Funded Fiscal Years: 2008, 2009 and 2010
Keywords: None

Research Question

* How does the spatial distribution of temperature in rivers impact the ability to meet biological and habitat constraints on water operations and power generation?

* How can the impact of different project scenarios on thermal processes be assessed and evaluated?

Features with significant spatial variability include tributaries and agricultural returns, gravel pits, ground water upwelling, side channel activation, streamside vegetation, point bars and pools. Flows and temperature within these features behave differently than those in the main channel and are not captured in low-order models. Improved representation of temperature processes will reduce uncertainty in predicting impacts and help improve Reclamation's ability to comply with thermal criteria.

Need and Benefit

Most rivers have thermal criteria such that the temperature cannot exceed a specific limit for habitat and biological reasons. These temperature criteria can limit water deliveries and impose constraints on water deliveries and power generation. With growing emphasis on maintaining biological function, the impact of thermal criteria in constraining water operations will increase.

Existing tools based on low-order modeling include only the main channel and poorly represent the physics and spatial distribution of inputs and results. The simplified picture of interactions requires inferring or estimating real processes through conversion to lower-order dimensions with abstract calibration coefficients. These coefficients increase the difficulty of model usage by requiring large calibration data sets and increase the uncertainty by masking processes. Further, coefficients not tied to physical processes do not apply across a broad range of conditions. Predictive capability is limited to circumstances within measured ranges and cannot estimate the impact from changes to system operations. The lack of strong ties to physical processes requires spending more water to account for uncertainty. Limited spatial extent restricts the usefulness of low-order modeling with features such as tributary inflows, agricultural returns, gravel pits, ground water upwelling, side channel activation, variable flow rates, and streamside vegetation.

This research adds a temperature module into an existing two-dimensional (2D) hydraulic model. The result will incorporate data with geographic extents (both lateral and longitudinal) in a 2D manner rather than lumping results into a point-to-point or one-dimensional (1D) representation. 2D models account for real features with more direct process modeling resulting in greater accuracy, predictive ability for future outcomes under different conditions, and identification of the mechanisms required to develop potential solutions for reducing the impact of temperature requirements on operations. An application to a field site will compare how incorporation of spatially distributed processes changes results and conclusions versus point-to-point or 1D solutions.

A 2D representation provides a more direct and quantitative estimate of temperature impacts using the same dimensions that are significant for biology. Accurate flow hydraulics eliminate the need for abstract travel time coefficients and flow routing parameters. In lateral splits, such as gravel pits or side channels, flow moves at different speeds along the different paths and, as a result, heats or cools at different rates. The difference in flow velocities between the river centerline and banks is also captured, eliminating calibration for cross section averaging. Spatially distributed sources of heat and cooling (ground water, solar, wind, vegetation) are directly transferable to the model and do not require grouping by reach. Spatially explicit results may show acceptable areas within generally unacceptable conditions and permit compromises where simplified models might indicate temperature violations.

A process model can indicate the most critical component of critical areas to focus solutions. Grouping and lumping with coefficients obscures the physical steps preventing temperature compliance. Better representation of processes leads to simpler model formulation, increased accuracy, higher confidence, and less additional water required to account for model uncertainty.

Contributing Partners


Research Products

Bureau of Reclamation Review

The following documents were reviewed by experts in fields relating to this project's study and findings. The results were determined to be achieved using valid means.

On a Two-Dimensional Temperature Model: Development and Verification (final, PDF, 149KB)
By Yong Lai and David Mooney
Publication completed on June 26, 2012

Government regulators on many rivers have specified acceptable temperatures based upon habitat and biological criteria. These temperature thresholds impose constraints on reservoir operations and can limit water deliveries and power generation. Existing tools based on low-order modeling simplify a river to a simple line with limited spatial distribution of inputs and poorly represent physics of the river processes. The limited spatial extents restrict the usefulness of low-order modeling for suc
Keywords: temperature model; 2d-depth averaged; numerical model

Last Updated: June 29, 2015