Abstract

Landsat Level-1 products are delivered as quantized and calibrated scaled Digital Numbers (DN). The Level-1 DN data can be rescaled to Top-of-Atmosphere (TOA) reflectance applying radiometric rescaling coefficients. Currently, the Level-1 product is the standard data product of the Landsat sensors. The more recent Level-2 data products contain surface reflectance values, i.e., reflectance as it would be measured at ground level in the absence of atmospheric effects; in the near future, these products are anticipated to become the standard products of the Landsat sensors. The purpose of this paper is to present a radiometric performance evaluation of Level-1 and Level-2 data products for the Landsat-7 Enhanced Thematic Mapper Plus (ETM+) and Landsat-8 Operational Land Imager (OLI) sensors. TOA reflectance and derived surface reflectance values from both data products were evaluated and compared to in situ measurements from eight test sites located in Turkey, Brazil, Chile, the United States, France, and Namibia. The results indicate an agreement between the ETM+ and OLI Level-1 TOA reflectance data and the in situ measurements of 3.9% to 6.5% and 3.9% to 6.0%, respectively, across all spectral bands. Agreement between the in situ measurements and both Level-2 surface reflectance data products were consistently decreased in the shorter wavelength bands, and slightly better in the longer wavelength bands. The mean percent absolute error for Level-2 surface reflectance data ranged from 3.3% to 10% for both Landsat sensors. The significant decay in agreement with the data collected in situ in the short wavelength spectral bands with Level-2 data suggests issues with retrieval of aerosol concentration at some sites. In contrast, the results indicate a reasonably accurate estimate of water vapor in the mid-infrared spectrum. Lastly, despite the less reliable performance of Level-2 data product in the visible spectral region (compared with Level-1 data) in both sensors, the Landsat-8 OLI Level-2 showed an improvement of surface reflectance product over all spectral bands in common with the Landsat-7 ETM+ Level-2 data.

Highlights

  • The Earth Observation (EO) sensors systems provide extensive information about the long-term patterns and changes in the Earth’s land, sea, and atmospheric systems in a relatively timely manner

  • Landsat Level-1 data are distributed as quantized and calibrated scaled digital numbers, convertible to TOA reflectance by applying the radiometric rescaling coefficients provided with the Landsat product metadata

  • Level-1 data are the standard product of the Landsat sensors

Read more

Summary

Introduction

The Earth Observation (EO) sensors systems provide extensive information about the long-term patterns and changes in the Earth’s land, sea, and atmospheric systems in a relatively timely manner. An EO system digital image is a two-dimensional array of pixels with an intensity value known as its Digital Number (DN). The step is applying an atmospheric correction, i.e., removing scattering and absorption effects caused by the atmosphere [4,5]. Such correction is significant in cases where multi-temporal sensor images are to be compared and analyzed. After these procedures, the image pixel values represent an estimate of the surface reflectance or Bottom-of-Atmosphere (BOA) reflectance level

Objectives
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call