Back to search

South Dakota State University

Surface Reflectance Product Generated from MSS Using SMACAA and Validation with Respect to LEDAPS

Abstract

dc:description.abstract

Landsat MSS 1-5 sensors were the only sensors that provided systematic global multispectral space-based observation of the Earth’s surface from early 1970s, but they lack Level-2 Collection 2 surface reflectance (L2C2 SR) products. This research proposes a method of generating L2C2 SR products for the MSS 1-5 sensors going back to the 1970s. The primary motivation behind this project was to validate the SMACAA model ability to operate on MSS sensors. SMACAA is based on the inversion of MODTRAN and the automated application of SDSU meteorological database, and does not require a SWIR band. First, the performance of SMACAA model was evaluated by validating TM 5 SMACAA SR products against trusted TM 5 LEDAPS SR products. On average, TM 5 SMACAA & TM 5 LEDAPS products show agreement within an accuracy of 1-2 reflectance units. Agreement between both products over dark sites was consistent, but some inconsistency were observed over bright sites due to believed difference in the way AOD is retrieved over bright extended deserts between the two methods. Further, the SMACAA model was applied to MSS 5 datasets over selected sites, resulting in the generation of MSS 5 SR products. To ensure the spectral consistency between the two sensors, an appropriate Spectral Band Adjustment Factor (SBAF) was applied to MSS 5 SMACAA products, and cross validated with TM 5 LEDAPS products. The average difference between MSS 5 SMACAA and TM 5 LEDPAS generated SR products is within an accuracy of 0.005 reflectance units, with a precision of ±0.02. This indicates a high level of agreement between the two products. Additionally, both products agree well in an atmosphere where the aerosol optical depth varies from 0.6 to 0.16, but it starts to disagree above 0.16 aerosol. Over deep desert sites, both products might disagree with each other, indicating inconsistencies in the accurate retrieval of water vapor, which can be either overestimated or underestimated by both methods. Per-pixel total uncertainty obtained for SMACAA estimated MSS 5 SR using Monte Carlo Simulation, varies from 5.4-6.52% for all bands, except the green band, which can vary up to 7.4% depending upon different atmospheric conditions and reflectance levels. This implies that as brightness levels increases, the uncertainty decreases (~5.45%) in the green band. The feasibility of generating MSS 1-5 SR products using the proposed method has been successfully demonstrated, with potential for future applications in long-term monitoring of land surface changes.

Degree

thesis:*
Name thesis:degree_name
Master of Science (MS)
Level thesis:degree_level
Thesis - Open Access
Discipline thesis:degree_discipline
Electrical Engineering and Computer Science

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Shah, Rahul Kumar
Contributors dc:contributor
  • Larry Leigh

Rights

dc:rights

Identifiers

dc:identifier.*
Repository record dc:identifier
https://openprairie.sdstate.edu/etd2/858

Chain of custody

source
Harvested from
South Dakota State University
Base URL
openprairie.sdstate.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Shah, Rahul Kumar. Surface Reflectance Product Generated from MSS Using SMACAA and Validation with Respect to LEDAPS. Thesis - Open Access thesis, https://openprairie.sdstate.edu/etd2/858