Remote sensing of fuel moisture content from ratios of narrow-band vegetation water and dry-matter indices

REMOTE SENSING OF ENVIRONMENT

卷:129

Article

Wang, Lingli¹  Hunt, E. Raymond, Jr.²  Qu, John J.¹  Hao, Xianjun¹  Daughtry, Craig S. T.² 

[1] George Mason Univ, Coll Sci, Dept Geog & GeoInformat Sci, Environm Sci & Technol Ctr, Fairfax, VA 22030 USA

[2] ARS, USDA, Hydrol & Remote Sensing Lab, Beltsville, MD USA

关键词: Fuel moisture content;    Spectral reflectance;    Normalized Dry Matter Index;    Normalized Difference Infrared Index;    PROSPECT;    SAIL;   

DOI  :  10.1016/j.rse.2012.10.027

来源: Elsevier

【 摘 要 】

Fuel moisture content (FMC) is an important variable for predicting the occurrence and spread of wildfire. Because FMC is calculated from the ratio of canopy water content to dry-matter content, we hypothesized that FMC may be estimated by remote sensing with a ratio of a vegetation water index to a vegetation dry-matter index. Four vegetation water indices, six dry-matter indices, and the resulting water/dry-matter index ratios were calculated using simulated leaf reflectances from the PROSPECT model. Two water indices, the Normalized Difference Infrared Index (NDII) and the Normalized Difference Water Index (NDWI), were more correlated with leaf water content than with FMC, and were not correlated with leaf dry-matter content. Two dry-matter indices, the Normalized Dry Matter Index (NDMI) and a recent index (unnamed) were correlated to leaf dry matter content, were inversely correlated with FMC, and were not correlated with water content. Ratios of these water indices and these dry-matter indices were highly and consistently correlated with FMC. Ratios of other water indices with other dry-matter indices were not consistently correlated with FMC. The ratio of NDII with NDMI was strongly related to FMC by a quadratic polynomial equation with an R-2 of 0.947. Spectral reflectance data were acquired for single leaves and leaf stacks of Quercus alba, Acer rubrum, and Zea mays; the relationship between FMC and NDII/NDMI had an R-2 of 0.853 and was almost identical to the equation from the PROSPECT model simulations. For the SAIL model simulations, the relationship between NDII/NDMI and FMC at the canopy scale had an R-2 of 0.900, but the quadratic polynomial equation differed from the equations determined from the PROSPECT simulations and spectral reflectance data. NDMI requires narrow-band sensors to measure the effect of dry matter on reflectance at 1722 nm whereas NDII may be determined with many different sensors. Therefore, monitoring FMC with NDII/NDMI requires either a new sensor or a combination of two sensors, one with high temporal resolution for monitoring water content and one with high spectral resolution for estimating dry-matter content. Published by Elsevier Inc.

【 授权许可】

Free