Graduate Thesis Or Dissertation
 

Sensitivity and Partitioning in Remote Sensing Based Evapotranspiration Models

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/k643b6454

Descriptions

Attribute NameValues
Creator
Abstract
  • Satellite Based evapotranspiration (ET) models have become a dominant means to estimate large-scale surface fluxes of water. Global and regional ET estimates are important parameters in many climate forecasts and hydrologic models. However, large scale partitioning of ET into soil evaporation, transpiration, and canopy interceptions remains largely unknown and modeled estimates have been shown to diverge strongly. This study examines three such remote sensing-based models: the Penman-Monteith model from the Moderate Resolution Imaging Spectroradiometer (PM-MOD), the Priestley-Taylor Jet Propulsion Laboratory model (PT-JPL), and the Global Land Evaporation Amsterdam Model (GLEAM). Modeled ET and component estimates are compared against a compiled dataset of field estimates using stable isotopes, sap flux, and other methods. Results are analyzed across land cover type, precipitation regime, and field methods. Overall, we find large deviations between field and modeled estimates of soil evaporation (RMSD = 90-114%, r2 = 0.14-0.25, N = 35), interception (RMSD = 62-181%, r2 = 0.39-0.85, N = 13), and transpiration (RMSD = 54–114%, r2 = 0.33-0.55, N = 35) compared to the deviations found in the total ET estimate (RMSD = 35-49%, r2 = 0.61-0.75, N = 35). We then conduct a Monte Carlo sensitivity analysis using varying degrees of parameter uncertainty to determine how forcing data error influences model estimates and to determine which parameters are the primary drivers of each model. We find large sensitivity and bias in component estimates that becomes mitigated when aggregated into a total ET estimate. The results also show that the total ET of PT-JPL, PM-MOD, and GLEAM is most sensitive to NDVI, RH, and net radiation, respectively. The results of each study suggest that the soil evaporation component exhibits large errors and may be culpable for errors in ET partitioning. These results suggest that future improvements to remote sensing-based ET component estimates will vastly improve the confidence of the total ET estimates and provide greater understanding in how water interacts with vegetation, climate, and society.
Contributor
License
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Rights Statement
Publisher
Peer Reviewed
Language
Embargo reason
  • Ongoing Research
Embargo date range
  • 2018-06-14 to 2020-07-15

Relationships

Parents:

This work has no parents.

In Collection:

Items

Thumbnail Title Date Uploaded Visibility Actions
file details TalsmaCarlJ2018.pdf 2018-06-15 Public Download