Date of Award

2021-12-01

Degree Name

Master of Science

Department

Engineering

Advisor(s)

Dr. Miguer Velez-Reyes

Abstract

The GOES-16 satellite ground segment team develops different levels of environmental products and makes them readily available to the public. The level 2 Land Surface Temperature (LST) product aims to provide a remote measurement of the Earth’s surface temperature every hour. LST is of fundamental importance to many aspects of the geosciences, for example, to study the Urban Heat Islands (UHI) effect. The product however, does not provide values for pixels where the satellite imager’s view is obstructed by clouds, as determined by a cloud mask intermediate product. The objective of this thesis is to estimate LST in the cloud masked pixels by taking advantage of the temporal resolution of the GOES-16 level two product and other remote sensing products related to the temperature of the area of interest. This is done by using two different methods; Long-Short Term Memory (LSTM) Neural Network and an Interpolation Method. LSTM is well-suited to make prediction based on time series data and the interpolation method uses neighbor pixels without cloud cover for the estimation of LST. The accuracy of the estimation was measured by reconstructing the images with the estimated values of LST and using R and RMSE as performance evaluation metrics. The results show that the interpolation method has better performance in the estimation of LST since the reconstructed images show better values in the performance evaluation metrics ranging between 0.3460 to 0.9880 for R and 0.2929 K to 0.8111 K for RMSE. The performance evaluation metrics of the images that were reconstructed using LSTM range between -0.0859 to 0.5548 for R and 0.546 K to 1.9844 K for RMSE.

Language

en

Provenance

Received from ProQuest

File Size

99 p.

File Format

application/pdf

Rights Holder

Guadalupe Ortega

Download

Share

COinS