Abstract

MARTINEZ BARRON, René Alejandro et al. Modeling of biomass and aboveground arboreal carbon in forests of the state of Durango. Rev. mex. de cienc. forestales [online]. 2016, vol.7, n.35, pp.91-105. ISSN 2007-1132.

Forest ecosystems are particularly important as carbon reservoirs, since the trees retain large amounts of this element per surface area unit, compared to other types of vegetation. Within this context, the carbon reservoirs in various evaluated systems in the state of Durango, aged 5 to 60 years, were quantified; the biomass and aboveground arboreal carbon were spatially modeled using satellite imagery, and biomass (Mg ha-1) and aboveground arboreal carbon (Mg ha-1) maps of the selected area were generated. This study presents the relationships between biomass and carbon in forests obtained from the spectral data analysis by the Landsat TM (Thematic Mapper), in the State of Durango, Mexico. The multiple linear regression method was utilized to determine relationships and estimations. The results showed a good correlation between forest variables and spectral indices related to vegetation moisture. The generated models used for quantifying the biomass and aboveground carbon were those with the best fit and fewest errors. For the biomass, the root mean square error was RMSE = 54.74 Mg ha-1, B = 334.69 - 12.83(B2) +107.13(B3) + 8 250.39(NDVI) - 8 265.91(NDVI41) + 6 548.72(NDVI43), and for carbon, RMSE = 26.87 Mg ha-1, C = 164.71 - 55.23(B2) + 52.54(B3) + 4 052.05(NDVI) - 4 059.76(NDVI41) + 3 215.18(NDVI43). Landsat TM images provide valuable information to estimate the biomass and quantify the aboveground arboreal carbon.

Keywords : Forest biomass; aboveground arboreal carbon; spectral data; Landsat TM; reflectances; SPOT satellite.

        · abstract in Spanish     · text in English | Spanish     · English ( pdf ) | Spanish ( pdf )