Abstract Fire behavior prediction models can assist environmental agencies with fire prevention and control. This study aimed to adjust a fire prediction model for the state of Minas Gerais, Brazil. Using the R program and hotspots provided by the National Institute for Space Research (INPE) for 2010, prediction of the probability of fires through the Random Forest algorithm was conducted using the Bootstrapping method. The model generated a prediction map with global kappa value of 0.65. External validation was performed with hotspots in 2015. Results showed that 58% of the hotspots are in areas with ignition probability > 50%, being 24% of them in areas with 25-50% probability, and 17% in areas with < 25% probability. These results were considered satisfactory, demonstrating that the model is suitable for predicting fires.

ABSTRACT: Little is known about the groundwater recharge potential of weathered tropical catchments, where increasing water uptake is widespread to meet various water demands. This study aimed to estimate the volume of groundwater recharge of the Doce River Basin, Minas Gerais, Brazil. The BALSEQ model was applied to calculate the water balance over a period of two years (2007-2009). Evapotranspiration, runoff, and potential groundwater recharge (PGR) were calculated, using daily data on rainfall, potential evapotranspiration, and plant-available water. A soil survey was undertaken for all major soils occurring in the basin. Soils samples were used to determine hydraulic conductivity, bulk density, and water content at field capacity and at the permanent wilting point. Vegetation data were obtained from the literature and used to determine the following parameters: canopy interception, crop coefficient, and root depth. The estimated groundwater was spatially predicted using the Random Forests model with digital elevation, vegetation index, pedological, lithological, and climate maps. During the two years, an average of 32 % of rainfall was converted to groundwater. Annually, the percent of rainwater converted to groundwater varied between 27 and 48 % for all soil classes, highlighting the great temporal variability. The spatial prediction showed a volume of approximately 17,484 and 35,410 m3 of rainfall being converted to groundwater for the first and second year, respectively. The BALSEQ model showed a feasibility for the water balance calculation and can be reapplied for updating the groundwater maps of the Doce River Basin. These maps could then be used to guide land use planning programs, with the aim to protect water resources.

RESUMO A região do Quadrilátero Ferrífero (QF) em Minas Gerais apresenta notável geobiodiversidade, a despeito de uma longa história de pressões antrópicas (mineração, urbanização), mas ainda carece de estudos detalhados sobre a distribuição das formações vegetais nos diferentes substratos. No presente estudo, utilizou-se imagens Aster, levantamentos gamaespectométricos (Gama), mapeamentos geológicos existentes e trabalhos de campo, para a discriminação e quantificação dos remanescentes da vegetação sobre substratos ferruginosos e outros no QF. Utilizou-se o algoritmo da Máxima Verossimilhança (MaxVer) para a classificação das tipologias de vegetação associadas aos substratos predominantes (canga-ferruginosos, filitos e quartzitos), assim denominados: Campo Rupestre Aberto, Campo Rupestre Arbustivo, Capão Florestal, Cerrado Stricto Sensu, Campo Cerrado, Florestas Estacionais, Pastagens e Reflorestamentos, sendo os três últimos não discretizados quanto ao substrato. A utilização das imagens Aster apenas não permitiu uma separação confiável de substratos ferruginosos dos não ferruginosos, sendo necessária a utilização de produtos de processamento de dados Gama e de geologia. A integração entre as bases de informação (Aster-MaxVer + Gama + geologia) permitiu o mapeamento da vegetação associada a três tipos de substratos: ferruginoso, potencialmente ferruginoso e não-ferruginoso. O mapeamento revelou que a vegetação sobre substrato ferruginoso e potencialmente ferruginoso recobrem, respectivamente, 8,7% e 6,9% no QF. Estes dados preliminares inéditos permitem análises mais detalhadas do padrão de distribuição e fragmentação das fitofisionomias sobre substratos ferruginosos, com grande importância para as estratégias de conservação da geobiodiversidade do QF, e necessitam de checagens de campo e reconhecimento da vegetação, por meio de técnicas inovadoras.

ABSTRACT The Iron Quadrangle (IQ) region in Minas Gerais is remarkably geobiodiverse, despite a long history of anthropogenic pressures such as mining and urbanization, but still lacks detailed studies on the distribution of its remaining native vegetation in different substrates. In this study, we utilized Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) images, besides Gamma-spectrometry (Gamma) survey data associated with existing geological mapping (GM) and extensive fieldwork, to discriminate and quantify remnants of vegetation on ferruginous substrates in the IQ. The Maximum Likelihood (ML) algorithm was used to classify the vegetation types, thus named: open Rupestrian Field, shrubby Rupestrian Field, Capão Forest, Cerrado stricto sensu, Cerrado Field, Seasonal Forests, Pastures and Reforestation (the latter three regardless of substrate type) associated with the predominant substrates (ferruginous ironstone, phyllites, and quartzite). The use of ASTER images alone did not allow a reliable separation of ferruginous and non-ferruginous substrates, but the integration of all different data (ASTER-ML + Gamma + GM) allowed the provisional mapping of the vegetation associated with ferruginous substrates, potentially ferruginous and non-ferruginous substrates. The resulting map shows that the vegetation on ferruginous and potentially ferruginous substrates cover 8.7% and 6.9% of the IQ, respectively. The detailed analysis of the distribution and fragmentation of phytophysiognomies on ferruginous substrates is of great importance for developing strategies to conserve the geobiodiversity of the IQ, and need to be further refined by checking and field mapping by novel approaches.