Degree Type

Dissertation

Date of Award

2017

Degree Name

Doctor of Philosophy

Department

Geological and Atmospheric Sciences

Major

Geology

First Advisor

Kristie J. Franz

Abstract

Water resource planners require dynamic operational forecast information for making timely and accurate decisions, driving the necessity for hydrologic models that can account for physically explicit processes at the watershed scale. In operational applications, long-term average potential evapotranspiration (PET) inputs have been the standard; however, bias in the long-term average that underestimates an “observed” PET does not typically reflect watershed conditions. As a result, model simulations can lead to uncertainty in hydrologic processes such as evapotranspiration (ET), soil moisture, and discharge. Accurate input of PET that impact the simulated ET, is one vital component for replicating the overall water balance, and can ultimately assist in better decision-making. Over the last fifteen years, remote sensing data, specifically from Moderate Resolution Imaging Spectroradiometer (MODIS) sensors aboard NASA’s Terra and Aqua satellites, have become readily available as MODIS products are part of increasingly vital tools in modeling various Earth system processes. Robust testing through three separate analyses was conducted of a MODIS-derived mean daily PET for input to the current spatially lumped operational hydrologic forecast model.

First, we updated the long-term PET demand curves with a MODIS-derived mean daily PET and tested this product for fifteen forecast basins in the Upper Mississippi and Red River watersheds of the National Weather Service (NWS) North Central River Forecast Center (NCRFC). The updated PET demand curves were next input to the current spatially lumped operational Sacramento Soil Moisture Accounting (SAC-SMA) hydrologic forecast model. Overall results indicated potential for using a MODIS-derived mean daily PET demand as input into the SAC-SMA.

Second, a dynamic, daily-varying MODIS-derived mean daily PET (MODIS-PET) was tested against the current operational practice of climatological PET inputs to the SAC-SMA. The daily MODIS-PET performed as well as the climatological PET and the model simulated ET suggested that the dynamic MODIS-PET may produce a more physically realistic representation of ET processes in the lumped SAC-SMA model.

Third, we conducted sensitivity analysis of the response to PET resolutions of temporal variability to the operational streamflow prediction model. The PET inputs included the 8-day, Monthly, Seasonal, and Annual mean values. We examined the impact of the PET inputs in systems with different hydrologic controls by evaluating the modeled fluxes of discharge and ET, and the soil water states for the two climate regions. Overall, only the simulations of the Annual PET exhibited sensitivity for all hydrologic controls. Estimating PET inputs that are scientifically reasonable, operationally accessible, and that can highlight hydrologic controls that impact accurate forecasting are critical to demonstrating the full benefit of a satellite data for future operational use.

DOI

https://doi.org/10.31274/etd-180810-5111

Copyright Owner

Angela Loraine Bowman

Language

en

File Format

application/pdf

File Size

134 pages

Share

COinS