TY - JOUR

T1 - How well does random forest analysis model deforestation and forest fragmentation in the Brazilian Atlantic forest?

AU - Zanella, Lisiane

AU - Folkard, Andrew Martin

AU - Blackburn, George Alan

AU - Carvalho, Luis

N1 - The final publication is available at Springer via http://dx.doi.org/10.1007/s10651-017-0389-8

PY - 2017/12

Y1 - 2017/12

N2 - We assessed the value of applying random forest analysis (RF) to relating metrics of deforestation (DF) and forest fragmentation (FF) to socio-economic (S-E) and bio-geophysical (BGP) factors, in the Brazilian Atlantic Forest of Minas Gerais, Brazil. A vegetation-monitoring project provided land cover maps, from which we derived DF and FF metrics. An ecologic-economical zoning project provided more than 300 S-E and BGP factors. We used random forest analysis (RF) to identify relationships between these sets of variables, and compared its performance in this task to that of a more traditional multiple linear regression approach. We found that RF modelled relatively-well variance in all metrics used (the rate of deforestation, the amount of forest, and the density and isolation of forest patches), presenting a better performance when compared to the classical approach. RF also identified geographical location and topographic factors as being most closely associated with patterns of DF and FF. Both analyses found factors associated with economic productivity, social institutions, accessibility and exploration to have little relationship with metrics. RF was better at explaining variations in rates of deforestation, remaining forest and patch patterns, than the multiple linear regression approach. We conclude that RF provides a promising methodology for elucidating the relationships between land use and cover changes with potential drivers.

AB - We assessed the value of applying random forest analysis (RF) to relating metrics of deforestation (DF) and forest fragmentation (FF) to socio-economic (S-E) and bio-geophysical (BGP) factors, in the Brazilian Atlantic Forest of Minas Gerais, Brazil. A vegetation-monitoring project provided land cover maps, from which we derived DF and FF metrics. An ecologic-economical zoning project provided more than 300 S-E and BGP factors. We used random forest analysis (RF) to identify relationships between these sets of variables, and compared its performance in this task to that of a more traditional multiple linear regression approach. We found that RF modelled relatively-well variance in all metrics used (the rate of deforestation, the amount of forest, and the density and isolation of forest patches), presenting a better performance when compared to the classical approach. RF also identified geographical location and topographic factors as being most closely associated with patterns of DF and FF. Both analyses found factors associated with economic productivity, social institutions, accessibility and exploration to have little relationship with metrics. RF was better at explaining variations in rates of deforestation, remaining forest and patch patterns, than the multiple linear regression approach. We conclude that RF provides a promising methodology for elucidating the relationships between land use and cover changes with potential drivers.

KW - Land use and land cover change

KW - Machine-learning technique

KW - Minas Gerais State

KW - Socioeconomic and biogeophysical factors

KW - Stepwise multiple regression

KW - Tropical forests

U2 - 10.1007/s10651-017-0389-8

DO - 10.1007/s10651-017-0389-8

M3 - Journal article

VL - 24

SP - 529

EP - 549

JO - Environmental and Ecological Statistics

JF - Environmental and Ecological Statistics

SN - 1352-8505

IS - 4

ER -