Assessing Impacts of Climate Change in a semi arid Watershed using statistically

downscaled IPCC Climate Output

Seshadri Rajagopal1

Dr. Hoshin V Gupta1 (Project Adviser)

Collaborators

Dr. Peter A Troch1

Dr. Francina Dominguez2

Dr. Christopher L Castro2

1Department of Hydrology & Water Resources

2Department of Atmospheric Sciences

College of Science

The University of Arizona

Abstract

This research is helping water managers at the Salt River Project (SRP), Phoenix, to assess long term climate change impacts for the Salt and Verde River basins, and make informed policy decisions. We estimate important components of water balance for the next 100 years by developing and testing a physically based distributed hydrologic model forced by statistically downscaled IPCC climate information. The variable infiltration capacity (VIC) model was calibrated and used to simulate historical observed streamflow at the outlet of Salt and Verde River basins using gridded observed precipitation and temperature data. Bias-corrected and spatially downscaled climate projections were derived from the World Climate Research Programme's (WCRP's) Coupled Model Intercomparison Project phase 3 multi-model dataset. Three of the Global Climate Model’s (GCM’s) participating in the IPCC fourth assessment (UK-HADCM3, MPI-ECHAM5, NCAR-CCSM3) were found to best simulate the climatology for the study region. Data for three emission scenarios (A1B, A2, B1) was then temporally downscaled and used to drive the hydrologic model. In addition, based on stakeholder feedback from SRP and the City of Phoenix, we included two more GCM's to our model ensemble - one that predicts wetter (than model average) conditions and one that predicts drier conditions, to better represent the full range of GCM prediction uncertainty. Analysis of the basin scale responses to predicted climate change shows that all models predict a statistically significant increase in temperature over the 21st century. The multi-model ensemble predicts that streamflow will decrease by about 25% by the end of the 21st century, caused mainly by a significant decrease in storage of snow within the basin and by decreased winter precipitation. The results clearly show that water management in central Arizona will be significantly impacted by projected changes in regional climate.

1

1. Introduction

To meet much of its water requirement, semi-arid Arizona relies upon surface water sources within the state supplemented by an allocation of approximately 2.8 million acre-ft/year from the Colorado River. In particular, central Arizona relies on big reservoir projects such as the Salt River Project (SRP) and the Central Arizona Project (CAP) to meet its ever-growing demand. Projected changes in regional climate are expected to directly impact this surface water availability. Consequently, the complex question of how to maintain sustainable water resources in Arizona has become increasingly important.

Past studies in the Salt-Verde basin (Figure 1) have examined paleo-climate records

from tree rings (Smith and Stockton, 1981) to quantify historical streamflow and drought in the basin, and Sheppard et al. (2002) have investigated the climate system of the southwest US. However, there remains a lack of understanding regarding how climate affects surface hydrological processes in central Arizona. In particular more understanding is needed about the: 1) spatial and temporal variability of precipitation, evapotranspiration (ET) and snow water equivalent (SWE) within the basin, 2) partitioning of precipitation into ET and recharge to compute the available water, and 3) the impacts of future climate change on variables such as temperature, streamflow, soil moisture, ET and SWE. Understanding the complex relationships between climate and hydrology is critical to the development of long term sustainable water management plans.

Figure 1 Salt-Verde River Basin

2

1.2 Relevance

The Colorado River drains 7 states in the western US. Several studies indicate that allotted deliveries (~2.8 MAF) from the Colorado River to Arizona may fall short due to population growth and a changing climate (e.g. Nash and Gleick, 1991, Christenssen et al., 2004, 2007, McCabe and Wolock, 2008, Rajagopalan et al., 2009, Barnett and Pierce, 2009). While Central Arizona receives part of this water through the CAP canal, the CAP has the lowest priority of Lower Colorado allocations and must curtail usage in a shortage year. Since Central Arizona also receives water from the Salt-Verde Basin through SRP (a water and power utility company for the City of Phoenix & the Salt River Valley that serves 1 Million retail customers and delivers more than 1 million acre-feet annually), it is critical that we understand the impacts of climate change on these two rivers. Our research objective is to improve understanding of the physical processes governing the hydrology of the Salt and Verde River basins by quantifying the components and dynamics of the water budget and by assessing the impacts of climate change on local hydrologic variables. This information will enable improved water resources management for the Salt-Verde River system.

1.3 Research Outline

To accomplish our goal a physically based distributed hydrologic model of the system has been setup, based on the macro-scale Variable Infiltration Capacity model (VIC; Liang et al., 1994). VIC is ideally suited for water/energy balance studies and has been extensively used to assess impacts of climate change elsewhere (e.g., Wood et al., 2004; Christenssen et al., 2007). We first calibrated the model to observed streamflow using historical forcing data from the Maurer et al. (2002) dataset. This calibrated model is being used to investigate the spatio-temporal variability of precipitation-runoff-ET, partitioning of precipitation into ET and recharge under current and potential future climates.

Advances in understanding of the coupled ocean-land-atmosphere system have helped scientists to develop sophisticated models that forecast both weather and climate around the globe. To evaluate future scenarios of available surface water we use the outputs from these climate models to drive our distributed hydrologic model and thereby to assess the expected impacts of climate change on local and regional water resources.

Part One – Water Budget of the Salt-Verde Basin, Arizona To investigate the water budget of the Salt-Verde Basin for the period 1949-2005, we merged observed meteorological data and forest cover, land use and land cover information with the VIC hydrologic model, to answer the following questions. Research Questions:

What are the contributions of precipitation to ET, streamflow, SWE and soil moisture and will these contributions change under potential future climates?

3

What are the trends in the historical time series of hydrologic variables in the Salt-Verde basin and how will these contributions change under potential future climates?

Part Two – Statistical Downscaling of Climate model and Impact Assessment

To assess the potential water balance impacts of anticipated climate over the next 100 years, we use spatial and temporal statistical downscaling to develop the data sets needed to drive the hydrological model. Since most climate models are run at relatively coarse scales (~30-200 km) it is impractical to directly assimilate the output from these models (precipitation, temperature) into our hydrologic models that must be run at relatively fine scales (~1-4 km) to provide meaningful information. In this research statistical downscaling is used to bridge the gap. Five climate models carefully selected from those participating in the IPCC fourth assessment are being used. These models have been selected based on how well they represent the historical summer monsoon and winter precipitation over the southwestern US, and to encompass the range of model uncertainty inherent in the IPCC projections. Research Questions:

What are suitable criteria to select GCM models used for downscaling? How do hydrologic variables such as streamflow, ET, SWE change in the future?

2. Methods-Model-Data

This section provides a brief overview of the methods, hydrologic model, and data used in this study.

2.1 Methods To investigate the potential impacts of climate change on the Salt-Verde watershed we combined three methods. The physically based macro scale VIC model (Liang et al., 1994), calibrated to match the Maurer et al (2002) dataset, was used to simulate the spatio-temporal dynamics of the hydrology of the basin. Anticipated hydrological changes were simulated by droving VIC with a multi-model ensemble of datasets (Pierce et al., 2009) derived from IPCC GCM runs. To ensure relevance to our specific study region, we did not use all available (25) GCMs but instead applied a conditional approach to selecting which GCMs should be included, based on metrics that evaluate historical similarity to historical observed climatology and atmospheric circulation patterns for the region (Dominguez et al., 2010). Since the GCM data is available at monthly timescale, it must first be spatially and temporally downscaled to correspond to the resolution of the hydrologic model. Temporal downscaling was performed by randomly resampling observed daily time-series sequences from historical data, and rescaling these values to match the monthly totals projected by the GCM. Sensitivity analysis showed that this resampling approach provides reasonable results. Finally, to evaluate and provide context for the model-based projections of declining snowpack, we analyze historical observed temperature records at SNOTEL sites within the basin.

4

2.2 Hydrologic Model

VIC is a distributed, macro scale hydrological model that simulates the water and energy budget at the land-surface (Liang et al., 1994). The model accounts for horizontal sub-grid variability in land use and elevation based on fractional areal coverage, and splits the vertical soil column into three layers (the soil moisture of the lowest layer controls long-term baseflow). For each grid cell the model produces both surface runoff and baseflow, which are then routed to the outlet of the grid via a unit hydrograph approach, and thence into the river network. This approach assumes that all runoff exits a grid cell in a single flow direction. Once in the river, linearized Saint-Venant equations are used to rout flows to the watershed outlet. Detailed descriptions of the routing model appear in Lohmann et al (1996, 1998). Figure 2 presents a conceptual diagram of the VIC model.

The VIC model is setup at 1/8th degree (~12.5 km) spatial resolution over the Salt and Verde watersheds (meaning the smallest grid cell in the hydrologic model is at the above specified resolution). Initial estimates of soil parameters used in the model were derived from the National Land Data Assimilation System (NLDAS, Mitchell et al., 2004) and Reynolds et al. (2000), and vegetation parameters were set to default values for semi-arid regions, based on land cover classification provided by Hansen et al. (2000). The model was set up to provide streamflow simulations at USGS stream gage 9498500 near Roosevelt Dam on the Salt River, at USGS 9499000 above Gun Creek near Roosevelt Dam on Tonto Creek, and at USGS 9508500 below Tangle Creek above Horseshoe Reservoir on the Verde River. These gages were chosen because they are upstream of the reservoir and represent natural flows without significant human influence.

Figure 2 Conceptual diagram of the VIC hydrologic model. Source:

http://www.hydro.washington.edu/Lettenmaier/Models/VIC/Overview/ModelOverview.shtml

5

2.3 Data

Two sets of atmospheric forcing data were used (Table 1), one for historical simulations and the other for future impact studies.

a) Historical Simulations: The Maurer et al. (2002) dataset, available at 1/8th degree spatial resolution for the continental USA, was used to force a land-surface hydrology model to generate simulations of historical streamflow. The forcing data includes gridded precipitation and temperature fields derived from National Oceanic and Atmospheric Administration Cooperative Observer (NOAA Co-op) stations using algorithms developed by Widmann and Bretherton (2000). The gridded daily precipitation data are scaled to match the precipitation and temperature climatology given by the parameter-elevation regression on independent slopes model (PRISM) (Daly et al., 1994). Wind speed was derived from the NCEP-NCAR reanalysis (Kalnay et al., 1996). Downward radiation was calculated based on daily temperature range and dewpoint temperature using a method described by Thornton and Running (1999).

b) Future Impact Studies: To evaluate the impacts of potential future climate change on local hydrologic variables, a second set of atmospheric forcing data was generated based on simulations provided by five GCMs that participated in the 4th IPCC assessment. The data was derived from phase 3 of the World Climate Research Programme (WCRP) Coupled Model Intercomparison Project multi-model dataset (hereafter called the CMIP3 dataset). The five GCMs selected are the ECHAM5 (Max-Planck Institute for Meteorology), HADCM3 (UK Hadley Center for Climate Prediction and Research), CCSM3 and PCM1.1 (National Center for Atmospheric Research), and MIROC3.2 (University of Tokyo National Institute for Environmental Studies, and Frontier Research Center for Global Change). The method for selecting these GCMs is described in section 2.3 (Methods). For each GCM, three IPCC emission scenarios (A2, A1B and B1) were selected to reflect the interests of water managers in Phoenix: A2 assumes a continuously increasing global population and regionally oriented economic growth, A1B assumes growth in population and economy until the mid 21st century followed by balanced use of energy, and B1 assumes a convergent world with same population as A1B but with introduction of clean and resource efficient technologies.

The choice of the five models used in this study (see Table 1) was guided by research performed by Dominguez et al., 2009. The criterion used for GCM model selection for the Southwestern US, is based on 1) the GCM performance for the historical 20th century climate simulations in comparison to the observed climate for the 1950-2000 period and 2) the GCMs simulation of known climate phenomena such as El Nino Southern Oscillation (ENSO) which affects the precipitation over the Southwest. Three of the five models used in this study (viz the MPI-ECHAM5, UK-HADCM3 and NCAR-CCSM3) met the criteria specified above, and two more included on the request of the City of Phoenix to represent possible futures that are wetter and drier than normal. The GCMs that showed this tendency in the precipitation signal were the NCAR-PCM1 and the MIROC3.2. Note however that the NCAR-PCM1 and the MIROC3.2 did not perform well in comparison to the other three models for the historical 20th century simulations; nonetheless they were included since they provide information regarding the possible ranges of change that one might expect.

6

Table 1Atmospheric data used in this study

Dataset Modeling group

IPCC Model

ID

Scenario Abbreviation Period

Primary

reference

Observations - - - Obs 1949-2000 Maurer et al.,

2002

GCM Max Planck Institute for

Meteorology, Germany ECHAM5 A2,A1B,B1 Mpi

Reference Future Jungclaus et al.,

2006 1960-

2000

2000-

2098

GCM

Hadley Center for Climate

Prediction and Research,

UK

UKMO-

HadCM3 A2,A1B,B1 Hadcm3

1960-

2000

2000-

2098

Gordon et al.,

2000

GCM

National Center for

Atmospheric Research,

USA

CCSM3 A2,A1B,B1 Ccsm3 1960-

2000

2000-

2098

Collins et al.,

2006

GCM

National Center for

Atmospheric Research,

USA

PCM1.1 A2 Pcm 1960-

2000

2000-

2098

Washington et

al., 2000

GCM

Center for Climate System

Research (The University of

Tokyo), Japan

MIROC3.2 A2 Miroc 1960-

2000

2000-

2098

K-1 model

developers 2004

7

3. Results

The results from our analysis of historical behaviors and potential future climate

scenarios are presented below.

3.1 Historical Hydro-Climatology of the Salt-Verde Basin

Figure 3 summarizes the historical hydro-climatology of the Salt and the Verde River Basin. Observed data has been averaged over 1960-2000 period for the Salt and Verde Rivers respectively. The error bars on streamflow show that the winter streamflow is more variable than summer streamflow. The observed record (Maurer et al., 2002; USGS Streamflow) indicates that the watersheds together receive ~160 mm of precipitation during the winter (Dec-Mar) and ~172 mm during summer (Jul-Sep). Corresponding seasonal streamflow averages are 820 million m3 (Dec-May) and 148 million m3 (Jul-Sep). The major contributor to streamflow in the Salt-Verde system is clearly winter precipitation - in fact 39% of annual precipitation at SNOTEL stations within the basin is recorded to be snowfall (Serreze et al., 1999).

Figure 3 Hydro-climatology of the Salt-Verde River Basins

8

Historical observed precipitation and temperature data was used to drive and calibrate the hydrologic model. Calibration of the model was performed for the ten year period 1981-1990; this is long enough to provide robust calibration given that it includes different climate conditions encountered in the basin (Yapo et al 1996). Figure 4 shows that the hydrologic model captures the dynamics of the system well and simulates streamflow with a fairly high efficiency.

Figure 4 Calibrated simulated historical streamflow in the Salt-Verd Basin

3.2 Future Climate Change impact

3.2.1 Projected change in Hydro-Climate variables Figure 5 shows time series of historical (1960-2000) observed as well as projected changes in temperature, precipitation and streamflow for the period 2000-2098. The projected timeseries are from the five GCMs selected for this study as explained in section 2.3 for the various emission scenarios. The multi-model mean projects that temperatures will rise by ~3.5 deg C and precipitation will decrease by about ~5%, with consequent reductions in streamflow of about 25% by the end of the 21st century (in comparison with historic averages for 1960-2000).

9

Figure 5 Historical and Future change to temperature, precipitation and streamflow in the Salt River Basin

3.2.2 Causes for the streamflow change While it might be tempting to conclude that the anticipated reductions in streamflow will be caused primarily by increases in evapotranspiration (driven by the significant rises in temperature), this conclusion would not be correct. Figure 6 shows how precipitation is transformed to streamflow at multiple (yearly, winter, summer) timescales, and reveals that decreases in ensemble mean annual streamflow are caused primarily by decreases in both winter precipitation and by reduced winter snow contributions to streamflow. This demonstrates the significancewinter processes to water supply in this basin. At present, about 70% of annual streamflow is realized in the winter and spring seasons, and winter average temperatures are on average about 2 deg C below freezing. Consequently, the

10

projected 3.5 deg C increase in temperature will cause most of the snow to vanish by the end of the 21st century. This is highlighted in the next section.

Figure 6 Monthly timeseries of various water balance components at annual, cool (Dec-May) and warm (June-Nov) timescales

11

3.2.3 Impacts to Snow Figure 6 illustrates the projected impacts of rising temperatures on snowpack within the basin as we progress into the second half of the 21st century. To examine how this impacts the spatial distribution of snow, we plot how the spatial distribution of winter average temperatures will change. We assume that all grid points within the watershed for which daily average temperatures fall below the zero degree line will be covered in snow during the winter. Figure 7 shows two panels, one for the historic period (1970-2000) and one for the future projections (2068-2098) of winter average temperature versus elevation for all grid cells within the basin. We infer that historically about 26% of the watershed has been covered by snow during the winter (Figure 7a). The model projections indicate that this area will reduce to only 5% by the end of the century (Figure 7b). Another inference is that the snow line (the elevation above which precipitation fell as snow) shifts from 2000m above mean sea level (historical) to about 2500m above mean sea level (future). Clearly, this result is significant from both a water management and a forest management perspective, since many of these high elevation areas are either national or state forests.

Figure 7 Plot of winter average temperature vs elevation to bring out the spatial change in snow

3.2.4 Impacts to Water Balance Figure 8 compares the historical and projected future water balances for the Salt River basin (results are averaged over all models and scenarios). The water balance in a simple form may be written out as

REPt

S

The change in storage (S) is due to soil moisture and snow, which are the two states of the system. P is the precipitation input and E is the evapotranspiration. The runoff (R) can be

12

split into fast runoff RO and a slow base flow (BF) component as shown in Figure 8. The historical period (1970-2010) is characterized by a significant winter snow contribution whereas the spring streamflow is driven primarily by melt of winter snow accumulations. For the future period (2051-2090), the winter snow contribution to spring melt is significantly reduced. The reason for this decrease is explained by Figures 5-6 as due to an increase in temperatures and a decrease in winter precipitation.

Figure 8 Monthly water balance change in the Salt River basin for historic and two future periods

4. Discussion and Conclusions

This work aims to provide information that will help water managers assess the impacts of climate change on the hydrological responses of the Salt and Verde watersheds in central Arizona. In particular we examine the change in various hydrologic components related to water availability. The bulk of Salt-Verde streamflow occurs in the winter season and we therefore studied this response closely. Climate projections for the basin indicate that temperature can be expected to increase by about 3.5 deg C, and precipitation can be expected to decrease slightly, with a consequent reduction in streamflow of about 25% by the end of the 21st century. The analysis indicates that this reduction in streamflow will be caused by a reduction in winter precipitation combined with reductions in snow contribution to spring runoff. Of major significance is that increased winter temperatures will cause the snow line to move from 2000m to about 2500m, causing a reduction in the snow covered area from 26% to only 5% by the end of the century. These findings are clearly significant from the standpoint of water management in central Arizona. With continued growth in population and likely reductions in available surface water supply, tough policy and management decisions will be needed to help sustain water supply. Our analysis provides important information useful for SRP and City of Phoenix assessments of the potential impacts of climate change on future operations.

13

Acknowledgement Support by the Water Sustainability Program, through the University of Arizona, Technology and Research Initiative Fund 2010/2011 Student Fellowship Program is greatly acknowledged. The author also received partial support from NASA (grant CAN NN-H-04-Z-YO-010-C), and from the Salt River Project, Phoenix. Several modeling groups, including the Program for Climate Model Diagnosis and Intercomparison (PCMDI) and the WCRP's Working Group on Coupled Modeling (WGCM) made available the datasets used in this work. Computer support by James Broermann at the Department of Hydrology and Water Resources is greatly appreciated.

References

Barnett, T. P., and D. W. Pierce (2009), Sustainable water deliveries from the Colorado River in

a changing climate, Proceedings of the National Academy of Sciences of the United States of

America, 106(18), 7334-7338.

Christensen, N. S., A. W. Wood, N. Voisin, D. P. Lettenmaier, and R. N. Palmer (2004), The

effects of climate change on the hydrology and water resources of the Colorado River basin,

Climatic Change, 62(1-3), 337-363.

Christensen, N. S., and D. P. Lettenmaier (2007), A multimodel ensemble approach to

assessment of climate change impacts on the hydrology and water resources of the Colorado

River Basin, Hydrology and Earth System Sciences, 11(4), 1417-1434.

Collins, W. D., et al. (2006), The Community Climate System Model version 3 (CCSM3),

Journal of Climate, 19(11), 2122-2143.

Daly, C., R. P. Neilson, and D. L. Phillips (1994), A STATISTICAL TOPOGRAPHIC MODEL

FOR MAPPING CLIMATOLOGICAL PRECIPITATION OVER MOUNTAINOUS

TERRAIN, Journal of Applied Meteorology, 33(2), 140-158.

Dominguez, F., J. Canon, and J. Valdes (2010), IPCC-AR4 climate simulations for the

Southwestern US: the importance of future ENSO projections, Climatic Change, 99(3-4), 499-

514.

Gordon, C., C. Cooper, C. A. Senior, H. Banks, J. M. Gregory, T. C. Johns, J. F. B. Mitchell, and

R. A. Wood (2000), The simulation of SST, sea ice extents and ocean heat transports in a version

of the Hadley Centre coupled model without flux adjustments, Climate Dynamics, 16(2-3), 147-

168.

14

Hansen, M. C., R. S. DeFries, J. R. G. Townshend, R. Sohlberg (2000), Global land cover

classification at 1km spatial resolution using a classification tree approach, Int. J. Remote Sens.,

21,1331–1364.

Jungclaus, J. H., N. Keenlyside, M. Botzet, H. Haak, J. J. Luo, M. Latif, J. Marotzke, U.

Mikolajewicz, and E. Roeckner (2006), Ocean circulation and tropical variability in the coupled

model ECHAM5/MPI-OM, Journal of Climate, 19(16), 3952-3972.

K-1 model developers (2004) K-1 coupled model (MIROC) description, K-1 technical report, 1.

In: Hasumi H, Emori S (eds) Center for Climate System Research, University of Tokyo, p 34

Kalnay, E., et al. (1996), The NCEP/NCAR 40-year reanalysis project, Bulletin of the American

Meteorological Society, 77(3), 437-471.

Liang, X., D. P. Lettenmaier, E. F. Wood, and S. J. Burges (1994), A SIMPLE

HYDROLOGICALLY BASED MODEL OF LAND-SURFACE WATER AND ENERGY

FLUXES FOR GENERAL-CIRCULATION MODELS, Journal of Geophysical Research-

Atmospheres, 99(D7), 14415-14428.

Lohmann, D., R. NolteHolube, and E. Raschke (1996), A large-scale horizontal routing model to

be coupled to land surface parametrization schemes, Tellus Series a-Dynamic Meteorology and

Oceanography, 48(5), 708-721.

Lohmann, D., E. Raschke, B. Nijssen, and D. P. Lettenmaier (1998), Regional scale hydrology:

I. Formulation of the VIC-2L model coupled to a routing model, Hydrological Sciences Journal-

Journal Des Sciences Hydrologiques, 43(1), 131-141.

Maurer, E. P., A. W. Wood, J. C. Adam, D. P. Lettenmaier, and B. Nijssen (2002), A long-term

hydrologically based dataset of land surface fluxes and states for the conterminous United States,

Journal of Climate, 15(22), 3237-3251.

McCabe, G. J., and D. M. Wolock (2008), Joint variability of global runoff and global sea

surface temperatures, Journal of Hydrometeorology, 9(4), 816-824.

Mitchell, K. E., et al. (2004), The multi-institution North American Land Data Assimilation

System (NLDAS): Utilizing multiple GCIP products and partners in a continental distributed

hydrological modeling system, Journal of Geophysical Research-Atmospheres, 109(D7).

Nash, L. L., and P. H. Gleick (1991), SENSITIVITY OF STREAMFLOW IN THE

COLORADO BASIN TO CLIMATIC CHANGES, Journal of Hydrology, 125(3-4), 221-241.

Pierce, D. W., T. P. Barnett, B. D. Santer, and P. J. Gleckler (2009), Selecting global climate

models for regional climate change studies, Proceedings of the National Academy of Sciences of

the United States of America, 106(21), 8441-8446.

15

Rajagopalan, B., K. Nowak, J. Prairie, M. Hoerling, B. Harding, J. Barsugli, A. Ray, and B.

Udall (2009), Water supply risk on the Colorado River: Can management mitigate?, Water

Resources Research, 45.

Reynolds, C. A., T. J. Jackson, and W. J. Rawls (2000), Estimating soil water-holding capacities

by linking the Food and Agriculture Organization soil map of the world with global pedon

databases and continuous pedotransfer functions, Water Resources Research, 36(12), 3653-3662.

Serreze, M. C., M. P. Clark, R. L. Armstrong, D. A. McGinnis, and R. S. Pulwarty (1999),

Characteristics of the western United States snowpack from snowpack telemetry (SNOTEL)

data, Water Resources Research, 35(7), 2145-2160.

Sheppard, P. R., A. C. Comrie, G. D. Packin, K. Angersbach, and M. K. Hughes (2002), The

climate of the US Southwest, Climate Research, 21(3), 219-238.

Smith, L. P., and C. W. Stockton (1981), RECONSTRUCTED STREAM-FLOW FOR THE

SALT AND VERDE RIVERS FROM TREE-RING DATA, Water Resources Bulletin, 17(6),

939-947.

Thornton, P. E., and S. W. Running (1999), An improved algorithm for estimating incident daily

solar radiation from measurements of temperature, humidity, and precipitation, Agricultural and

Forest Meteorology, 93(4), 211-228.

Washington, W. M., et al. (2000), Parallel climate model (PCM) control and transient

simulations, Climate Dynamics, 16(10-11), 755-774.

Wood, A. W., L. R. Leung, V. Sridhar, and D. P. Lettenmaier (2004), Hydrologic implications of

dynamical and statistical approaches to downscaling climate model outputs, Climatic Change,

62(1-3), 189-216.

Widmann, M., and C. S. Bretherton (2000), Validation of mesoscale precipitation in the NCEP

reanalysis using a new gridcell dataset for the northwestern United States, Journal of Climate,

13(11), 1936-1950.

Yapo P, HV Gupta and S Sorooshian (1996), Automatic Calibration of Conceptual Rainfall-

Runoff Models: Sensitivity to Calibration Data, Journal of Hydrology, 181(1-4), pp. 23-48