RESUMO
Changes in land use and land cover (LULC) have significant implications for biodiversity, ecosystem functioning, and deforestation. Modeling LULC changes is crucial to understanding anthropogenic impacts on environmental conservation and ecosystem services. Although previous studies have focused on predicting future changes, there is a growing need to determine past scenarios using new assessment tools. This study proposes a methodology for LULC past scenario generation based on transition analysis. Aiming to hindcast LULC scenario in 1970 based on the transition analysis of the past 35 years (from 1985 to 2020), two machine learning algorithms, multilayer perceptron (MLP) and similarity weighted (SimWeight), were employed to determine the driver variables most related to conversions in LULC and to simulate the past. The study focused on the Aristida spp. grasslands in the Uruguayan savannas, where native grasslands have been extensively converted to agricultural areas. Land use and land cover data from the MapBiomas project were integrated with spatial variables such as altimetry, slope, pedology, and linear distances from rivers, roads, urban areas, agriculture, forest, forestry, and native grasslands. The accuracy of the predicted maps was assessed through stratified random sampling of reference images from the Multispectral Scanner (MSS) sensor. The results demonstrate a reduction of approximately 659 934 ha of native grasslands in the study area between 1985 and 2020, directly proportional to the increase in cultivable areas. The MLP algorithm exhibited moderate performance, with notable errors in classifying agriculture and grassland areas. In contrast, the SimWeight algorithm displayed better accuracy, particularly in distinguishing grassland and agriculture classes. The modeled map using SimWeight accurately represented the transitions between grassland and agriculture with a high level of agreement. By modeling the 1970s scenario using the SimWeight model, it was estimated that the Aristida spp. grasslands experienced a substantial reduction in grassland coverage, ranging from 9982.31 to 10 022.32 km2 between 1970 and 2020. This represents a range of 60.8%-61.07% of the total grassland area in 1970. These findings provide valuable insights into the driving factors behind land use change in the Aristida spp. grasslands and offer useful information for land management, conservation, and sustainable development in the region. The study's main contribution lies in the hindcasting of past LULC scenarios, utilizing a tool used primarily for forecasting future scenarios. Integr Environ Assess Manag 2023;00:1-16. © 2023 SETAC.
RESUMO
O presente trabalho teve como objetivo avaliar o potencial das variáveis geomorfométricas extraídas de dados SRTM (Shuttle Radar Topographic Mission) para identificação de tipos vegetacionais da região do interflúvio Madeira-Purus. As variáveis geomorfométricas (elevação, declividade, orientação de vertente, curvatura vertical e curvatura horizontal) foram confrontadas com o mapa de vegetação do projeto RADAMBRASIL através de análises de histogramas e análises discriminantes. Tais análises indicaram os grupos de classes vegetacionais que podem ser separados mais facilmente em contraste com outros grupos que ocorrem sob as mesmas condições topográficas. As variáveis de relevo mais importantes na distinção entre os tipos vegetacionais foram: a elevação, a declividade e a orientação de vertentes. Apesar dos dados geomorfométricos mostrarem potencial indicativo das classes de vegetação, as análises resultaram em discriminação em um nível aquém do detalhamento temático máximo apresentado pelos dados RADAM. Tal desempenho pode ser explicado pela incompatibilidade das escalas de variação exibidas entre os dados geomorfométricos em relação ao tamanho das unidades de mapeamento da vegetação, além da co-ocorrência de classes de respostas distintas ao relevo sob uma mesma associação de classes sob as mesmas classes na legenda. Com base nas análises discriminantes das variáveis geomorfométricas, foi possível o mapeamento da vegetação experimentalmente até o nível de subfisionomias.
The objective of this work was to assess the potential of geomorphometric variables, derived from SRTM (Shuttle Radar Topographic Mission) data, to help identify vegetation types in the Amazonian Madeira-Purus interfluvio region. A RADAMBRASIL project vegetation map was used as a reference and the geomorphometric variables (elevation, slope, aspect and profile and plan curvatures) were compared to the mapped units. Analyses indicated vegetation types easily discriminated, depending on the topographic position. The variables of elevation, slope and aspect were the most important for their high discrimination power of the vegetation types. Although geomorphometric data are recognized as having strong potential for characterizing vegetation, this was not shown in the results, due to the mismatching of variability scales between the two sources of data; large units tend to exhibit similar distribution patterns of geomorphometry, and comprise classes with different responses for geomorphometric constraints. Discriminant analyses of geomorphometric variables permitted vegetation mapping up to the sub-physiognomy levels.
