Mapping the stock and spatial distribution of aboveground woody biomass in the native vegetation of the Brazilian Cerrado biome

FOREST ECOLOGY AND MANAGEMENT

卷:499

Article

Zimbres, Barbara¹  Rodriguez-Veiga, Pedro^2,3  Shimbo, Julia Z.¹  Bispo, Polyanna da Conceicao^2,4  Balzter, Heiko^2,3  Bustamante, Mercedes^5,6  Roitman, Iris^5,6  Haidar, Ricardo⁷  Miranda, Sabrina⁸  Gomes, Leticia^5,6  Carvalho, Fabricio Alvim⁹  Lenza, Eddie¹⁰  Maracahipes-Santos, Leonardo¹  Abadia, Ana Clara¹⁰  do Prado Junior, Jamir Afonso¹¹  Mendonca Machado, Evandro Luiz¹²  Dias Gonzaga, Anne Priscila¹³  Nunes Santos Terra, Marcela de Castro¹⁴  de Mello, Jose Marcio¹⁴  Soares Scolforo, Jose Roberto¹⁴  Rodrigues Pinto, Jose Roberto¹⁵  Alencar, Ane¹ 

[1] Amazon Environm Res Inst IPAM, BR-71503505 Brasilia, DF, Brazil

[2] NERC Natl Ctr Earth Observat NCEO, Leicester LE1 7RH, Leics, England

[3] Univ Leicester, Ctr Landscape & Climate Res, Sch Geog Geol & Environm, Leicester LE1 7RH, Leics, England

[4] Univ Manchester, Sch Environm Educ & Dev, Dept Geog, Oxford Rd, Manchester M13 9PL, Lancs, England

[5] Univ Brasilia UnB, Dept Ecol, BR-70910900 Brasilia, DF, Brazil

[6] Brazilian Res Network Global Climate Change Rede, BR-70910900 Brasilia, DF, Brazil

[7] Fed Univ Tocantins UFT, BR-77001090 Palmas, TO, Brazil

[8] Goias State Univ UEG, BR-76190000 Palmeiras De Goias, Go, Brazil

[9] Fed Univ Juiz de Fora UFJF, Dept Bot, BR-36037000 Juiz De Fora, MG, Brazil

[10] Mato Grosso State Univ UNEMAT, BR-78690000 Nova Xavantina, MT, Brazil

[11] Fed Univ Uberlandia UFU, BR-38402020 Uberlandia, MG, Brazil

[12] Fed Univ Vales Jequitinhonha & Mucuri UFVJM, Dept Forest Engn, BR-39100000 Diamantina, MG, Brazil

[13] Fed Univ Vales Jequitinhonha & Mucuri UFVJM, Dept Geog, BR-39100000 Diamantina, MG, Brazil

[14] Fed Univ Lavras UFLA, Dept Forest Sci, BR-37200000 Lavras, MG, Brazil

[15] Univ Brasilia UnB, Dept Forest Engn, BR-70910900 Brasilia, DF, Brazil

关键词: ALOS-2 PALSAR-2;    Savanna;    Landsat;    LiDAR;    Machine learning;    Synthetic Aperture Radar (SAR);   

DOI  :  10.1016/j.foreco.2021.119615

来源: Elsevier

【 摘 要 】

The Brazilian Cerrado biome consists of a highly heterogeneous tropical savanna, and is one of the world's biodiversity hotspots. High rates of deforestation, however, place it as the second-largest source of carbon emissions in Brazil. Due to its heterogeneity, biomass and carbon stocks in the Cerrado vegetation are highly variable, and mapping and monitoring these stocks are not a trivial effort. To address this challenge, we built an aboveground woody biomass (AGWB) model for the Cerrado biome using 30-m resolution optical satellite imagery (Landsat-5 and Landsat-8), 25-m resolution SAR imagery (ALOS and ALOS-2), and a set of plot-based and LiDAR-derived AGWB estimates (n = 1858) from a wide network of researchers in Brazil. We implemented both a Classification and Regression Tree (CART) and a Random Forest (RF) algorithm to model AGWB over the native vegetation in the year 2019 (as classified by MapBiomas) in the Cerrado. The RF algorithms resulted in a slightly better result (R-2 = 53%; rel. RMSE = 57%) than the CART model (R-2 = 45%; rel. RMSE = 63%), but our map shows an underestimation of very high AGWB (negative bias over 200 t ha(-1)) and a slight overestimation of low AGWB (positive bias), especially in the RF model (bias of 1.19 t ha(-1) against 0.86 t ha(-1) for the CART model). We believe we have contributed to knowledge on the woody biomass stocks in the biome, especially in the predominant savanna woodlands, which is where the highest current rates of conversion take place in the Cerrado.

【 授权许可】

Free