Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Publications at this Location » Publication #283126

Title: A data mining approach for sharpening satellite thermal imagery over land

Author
item Gao, Feng
item Kustas, William - Bill
item Anderson, Martha

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/22/2013
Publication Date: 10/26/2012
Citation: Gao, F.N., Kustas, W.P., Anderson, M.C. 2012. A data mining approach for sharpening satellite thermal imagery over land. Remote Sensing. 4:3287-3319.

Interpretive Summary: Thermal infrared band imagery provides key information for mapping land surface energy fluxes and evapotranspiration, and monitoring drought. Thermal infrared (TIR) imagery at fine resolution is required for field scale applications. However, thermal band imagery is normally acquired at a coarser pixel resolution compared to shortwave spectral bands. A robust data mining sharpening (DMS) approach has been developed to sharpen TIR imagery using spectral band data. The DMS approach was compared to a classic sharpening technique (TsHARP). Results show DMS outperforms TsHARP in three different test areas that represent rainfed agriculture, irrigated agriculture and a heterogeneous naturally vegetated landscape. This approach provides a feasible and cost effective solution for producing thermal imagery at fine spatial resolution for crop water use and drought monitoring that will greatly benefit the USDA National Agricultural Statistics Service (NASS) and Foreign Agricultural Service (FAS) in their operational applications.

Technical Abstract: Thermal infrared (TIR) imagery is normally acquired at coarser pixel resolution than that of shortwave sensors on the same satellite platform and often the TIR resolution is not suitable for monitoring crop conditions of individual fields or the impacts of land cover changes which are at significantly finer spatial scales. Consequently, thermal sharpening techniques have been developed to sharpen TIR imagery to shortwave band pixel resolutions, which are often at fine enough spatial resolutions for field applications. A classic thermal sharpening technique, TsHARP, uses a relationship between land surface temperature (LST) and Normalized Difference Vegetation Index (NDVI) developed empirically at the TIR pixel resolution and applied at the NDVI pixel resolution. However, recent studies show that unique relationships between temperature and NDVI may only exist for a limited class of landscapes, with mostly green vegetation and relatively homogeneous air and soil conditions. To extend application of thermal sharpening to more complex conditions, a new data mining sharpener (DMS) technique was developed. The DMS approach builds regression trees between TIR band brightness temperatures and shortwave spectral reflectances based on intrinsic sample characteristics. A comparison of sharpening techniques applied over a rainfed agricultural area in central Iowa, an irrigated agricultural region in the Texas High Plains, and a heterogeneous naturally vegetated landscape in Alaska indicates that the DMS outperformed TsHARP in all cases. The improvement varies with sharpening ratios (ratio of TIR pixel resolution to shortwave spectral band pixel resolution), image acquisition date, and the type of landscape. The mean absolute errors (MAE) from TsHARP ranged from 0.4 to 2 K, while the DMS approach yielded MAE between 0.3 to 1.6 K. In general, the DMS model provides better estimates of the sharpened LST in applications using Landsat TM, ETM+ and airborne data from different locations and seasons. The artificial box-like patterns in LST generated by the TsHARP approach are greatly reduced using the DMS scheme, especially for areas containing irrigated crops, water, thin clouds or terrain. While the DMS technique can provide fine resolution TIR imagery, there are limitations to the sharpening ratios that can be reasonably implemented. Furthermore, subpixel variability in LST that are uncorrelated with signals captured in the shortwave bands will not be recovered (e.g., due to sub-pixel variations in surface moisture conditions). Consequently, sharpening approaches cannot replace actual thermal band imagery at fine resolutions or missions that provide high quality thermal band imagery at high temporal and spatial resolution critical for many agricultural, land use, and water resource management applications.