Location: Hydrology and Remote Sensing Laboratory
Title: Remote sensing of fuel moisture content from canopy water indices and the normalized dry matter index Authors
Submitted to: Journal of Applied Remote Sensing (JARS)
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: November 12, 2012
Publication Date: December 3, 2012
Citation: Hunt, E.R., Wang, L., Qu, J.J., Hao, X. 2012. Remote sensing of fuel moisture content from canopy water indices and the normalized dry matter index. Journal of Applied Remote Sensing (JARS). 6:1-12. Interpretive Summary: Fuel moisture content (FMC) is one of the important variables for predicting the occurrence and spread of wildfire, and is measured by resource managers over large areas by a few point measurements. Remote sensing has potential to estimate FMC efficiently over large areas, but current methods are inadequate. Recently, the Normalized Dry Matter Index (NDMI) was developed for determining the dry mass per leaf area using high-spectral resolution data. We hypothesized that FMC could be predicted using the ratio of a vegetation water index to NDMI. We found that the ratio of the Normalized Difference Infrared Index (NDII) to NDMI was related to FMC using leaf and canopy simulation models and using measured leaf reflectances. The relationship between NDII/NDMI and fuel moisture content was the same among three species (white oak, red maple and maize) which indicates the proposed method is promising. However, the required satellite sensors for measuring FMC are in the planning stages, so it is unlikely that this proposed method could be implemented with the next few years.
Technical Abstract: An important variable for predicting the occurrence and spread of wildfire, fuel moisture content (FMC) is the ratio of foliar water content and foliar dry matter content. One approach for the remote sensing of FMC was to estimate the change in canopy water content over time by using a vegetation water index. Recently, the Normalized Dry Matter Index (NDMI) was developed for the remote sensing of dry matter content using high-spectral resolution data. Because FMC is a ratio, we hypothesized that FMC could be predicted using the ratio of a vegetation water index to NDMI. For leaf-scale simulations using the PROSPECT model, all water-index/NDMI ratios were significantly related to FMC with a second-order polynomial regression. For canopy-scale simulations using the SAIL model, two water-index/NDMI ratios, with numerators of the Normalized Difference Infrared Index (NDII) and the Normalized Difference Water Index (NDWI), predicted FMC with R2 values of 0.900 and 0.864, respectively. NDII/NDMI determined from spectral reflectance data had the best relationship with FMC, which was independent of species. Whereas the planned NASA mission Hyperspectral InfraRed Imager (HyspIRI) will have high spectral resolution and very high signal-to-noise properties, the planned 19-day repeat frequency will not be sufficient for monitoring FMC with NDII/NDMI. Since increased fire frequency is expected with climatic change, operational assessment of FMC at large scales will require polar-orbiting environmental sensors with narrow bands to calculate NDMI.