Use this URL to cite or link to this record in EThOS:
Title: Methodology and uncertainty in ET estimate from Landsat-7 ETM+ imagery
Author: Koloskov, Gleb
ISNI:       0000 0004 2745 4934
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2012
Availability of Full Text:
Full text unavailable from EThOS.
Please contact the current institution’s library for further details.
The thesis emphasizes the pivotal role that the Monin-Obukhov length (L) can play within the energy balance method for calculating evapotranspiration (ET) from remotely sensed multi spectral data. It focuses on the possibility of using L as a cornerstone parameter for calculations of sensible heat flux from the pixel data, which can be separated from the other unknowns and can be found by means of any general root finding method. The work goes on to show that all other output parameters can be calculated directly from the value of L that is established, and an analytical expression for the evaporative fraction found. A novel approach is proposed that allows a close approximation of the evaporative fraction to be derived directly without the need for time-consuming iterations within the main calculation step. The energy balance approach for calculating ET from remotely sensed data requires that near-surface air temperature be established from surface temperatures. To do this a linear calibration between surface and near surface temperature is required, which is provided by a calibration procedure by means of selection of a dry and a wet pixel within the image. The work demonstrates the importance of appropriate selection of the dry pixel temperature used for calibration on the ET estimate. It goes on to develop a procedure for refining the selection of the most appropriate hot pixel based on albedo, and water and vegetation indices. The proposed approach bounds the possible error to less than 13%. The work assesses the potential error in the estimate of ET associated with establishing surface temperature from remotely sensed data and looks at the potential value of remotely sensed surface temperature correction models. It shows that error arising from poor determination of surface temperature was less than 3.5% in most cases, although it could reach up to 9% in hot arid environments with small areas of transpiring vegetation.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available