• A new reservoir module is integrated into an open source distributed hydrologic model..

• The newly integrated model is able to capture the hydrologic characteristics of the reservoir

systems at a sub-monthly time step.

• Precipitation has a larger effect on reservoir storage and release than temperature does.

• Positive trends are observed in precipitation elasticity and temperature sensitivity.

• Precipitation elasticity is always positive and temperature sensitivity is always negative.

Sensitivity of Reservoir Storage and Outflow to Climate Change in a Water-limited River Basin

Gang Zhao1, Huilin Gao1, Bibi S. Naz2, Shih-Chieh Kao2, Nathalie Voisin3

1Zachry Department of Civil Engineering, Texas A&M University, College Station, TX 778432Environmental Sciences Division and Climate Change Science Institute, Oak Ridge National Laboratory, Oak Ridge, TN 37831

3Hydrology Group, Pacific Northwest National Laboratory, Richland, WA 99352

ReferencesChao BF, Wu Y, Li Y. 2008. Impact of artificial reservoir water impoundment on global sea level. Science, 320: 212-214.Hamlet AF, Lettenmaier DP. 1999. Effects of climate change on hydrology and water resources in the Columbia river basin1. JAWRA Journal of the American Water Resources Association, 35: 1597-1623. DOI: 10.1111/j.1752-1688.1999.tb04240.x.Moy W-S, Cohon JL, ReVelle CS. 1986. A Programming Model for Analysis of the Reliability, Resilience, and Vulnerability of a Water Supply Reservoir. Water Resources Research, 22: 489-498. DOI: 10.1029/WR022i004p00489.Wigmosta MS, Vail LW, Lettenmaier DP. 1994. A distributed hydrology-vegetation model for complex terrain. Water Resources Research, 30: 1665-1679. DOI: 10.1029/94WR00436.

Motivation: During the past several decades, numerous reservoirs have been constructed for

various water management purposes throughout the world. According to Chao et al. (2008),

the global accumulative reservoir impounded water volume reached 11,000 km3 in 2007. With

global environmental and anthropogenic change, great challenges have posed on the

reservoir systems. However, our understanding of the intricacies about how regulated flows

respond to these changes is critically limited due to following reasons:

• For those few hydrological models which have a explicit reservoir component, the

operation rules are typically optimized and rather simple;

• Water management models intend to use historical inflows for decision making, while the

potential alterations of future hydrological processes are often ignored;

• Commercial models are not suitable for public scientific researches due to unavailability.

Objective: Develop a reservoir module using conditional rules and integrate it with an open

source physically based distributed hydrologic model; and use this powerful modeling tool to

assist decision making related to reservoirs under changing environment.

Introduction

Modeling Approach

Study Area

Summary

Results and Discussions

• Fully distributed hydrological model

• Open source

• Physically based with simulation of:

Topography

Soil

Vegetation

Snow

Sediment

• Full water-energy balance simulation

• High spatial-temporal resolution

• Detailed urban component

• Reservoir module (from this study)

REPLACE THIS BOX WITH YOUR ORGANIZATION’

SHIGH

RESOLUTION LOGO

H33D-1644Dec 16 2015

2015

Figure 1 (left). (a) and (b) Topographically based

basin discretization and water movement in DHSVM

(Wigmosta et al., 1994); (c) the newly integrated

multi-purpose reservoir with each pool.

• Evaporation SchemePenman’s Equation for estimating open water evaporation (𝐸𝑡

𝑜) from the reservoir.

• Release Scheme Water released at time 𝑡, 𝑄𝑡

𝑜𝑢𝑡, is a function of several factors (i.e. upstream flow 𝑄𝑡𝑖𝑛,

current storage 𝑆𝑡, downstream channel capacity 𝐷0, and water demand 𝑈𝑤).

• Reservoir Storage Reservoir storage (𝑆𝑡) is calculated based on mass balance equation:

𝑆𝑡 = 𝑆𝑡−1 + ∆𝑡 ∙ (𝑄𝑡𝑖𝑛 − 𝐸𝑡

𝑜 ∙ 𝐴𝑡 − 𝑄𝑡𝑜𝑢𝑡 − ∆𝑠𝑒𝑑)

Figure 3. Schematic of the reservoir module and its integration to DHSVM.

Figure 4. Lake Whitney watershed in the Brazos

River Basin, Texas. Confluence of Brazos River with

Bosque River was chosen as the outlet and USGS

gauge 08091000 was chosen as the inlet.

Figure 5. Calibration and validation (weekly) of Lake Whitney (a) inflow, (b) elevation, and (c) release.

Panel (d) shows the monthly hydropower generation comparison between simulated results and US

EIA net generation.

Calibration and validation

Figure 2. Fitted storage-area and storage-elevation

relationships according to observations at Lake

Whitney using the following exponential functions.

𝐴 = 𝛼𝐴 ∙ 𝑆𝛽𝐴 + 𝛾𝐴

𝐻 = 𝛼𝐻 ∙ 𝑆𝛽𝐻 + 𝛾𝐻where 𝐴,𝐻, 𝑆 surface area, elevation, and

storage; 𝛼, 𝛽, 𝛾 are corresponding coefficients.

Reservoir Lake Whitney Aquilla Lake

Main purpose

Flood control,

hydropower,

municipal

Flood control

Impoundment

DateDecember 10, 1951 April 29, 1983

Sedimentation

rate (m3/year)1.91×106 1.91×105

Storage

(m3)

Elevation

(m)

Storage

(m3)

Elevation

(m)

Inactive pool 5.27×106 136.79 1.15×106 155.45

Conservation

pool7.73×108 162.46 6.46×107 163.83

Flood control

pool2.47×109 174.04 1.80×108 169.47

Acknowledgement: This research was supported by U.S. National Science Foundation grant CBET-1454297 and Mills Scholar Program provided by Texas Water Resources Institute. This paper was coauthored by employees of Oak Ridge National Laboratory under contract DE-AC05-00OR22725 with the U.S. Department of Energy. Contact information: Huilin Gao (hgao@civil.tamu.edu) or Gang Zhao (zeternity@tamu.edu)

As the largest of all 28 reservoirs (by storage capacity) in the Brazos River Basin, Lake

Whitney plays an essential role in the water resources management of Central Texas. After

construction, it significantly reduced flood damage in the Brazos River Basin.

Distributed Hydrology Soil Vegetation

Model (DHSVM)

R2=0.82NSE=0.81

R2=0.80NSE=0.68

Reservoir Module

A point reservoir module was developed and coupled to DHSVM. At each time step, the

simulation involves three components:

• Reservoir Rating Curves

(continued)

R2=0.87NSE=0.73

R2=0.92NSE=0.90

Climate change impacts

Temperature effects

• The trend in temperature

sensitivity is similar to

precipitation elasticity except

that the values are all negative.

• Higher temperature will make

the evapotranspiration larger,

which leads less runoff from the

same amount of precipitation.

Figure 6. Precipitation elasticity (-30% ~ +30%) on storage and release.

Figure 7. Temperature sensitivity (-2 oC~+2 oC) on storage and release.

Precipitation effects

• Precipitation has larger effect

(Fig 6a) on reservoir storage

and release than temperature

does (Fig 7c).

• Elasticity values are higher with

increase in precipitation for

both reservoir storage and

release. However, precipitation

elasticity is not always linearly

correlated with the percentage

change of precipitation.

𝑄𝑡𝑖𝑛

𝐸𝑡𝑜

𝑆𝑡

∆𝑠𝑒𝑑

𝑄𝑡𝑜𝑢𝑡

𝑆0

𝐴𝑡

𝑈𝑤

𝜀𝑃(𝑃, 𝑄) = ∆𝑄 𝑄

∆𝑃 𝑃

𝜀𝑃(𝑃, 𝑆) = ∆𝑆 𝑆

∆𝑃 𝑃

a) b)

c)

𝑆𝑇(𝑃, 𝑄) = ∆𝑄 𝑄

∆𝑇 𝑇

𝑆𝑇(𝑃, 𝑆) = ∆𝑆 𝑆

∆𝑇 𝑇

a)

b)

c)

Table 1. Reservoir configuration of Lake Whitney

and Aquilla Lake