RESUMO Considerando que o estudo da fenologia vegetal permite estabelecer relações entre o padrão fenológico da vegetação e as alterações causadas pela variabilidade climática, o objetivo deste trabalho foi obter métricas fenológicas para as tipologias de vegetação de pastagens predominantes no Rio Grande do Sul (latitudes 27º 05’ e 33º 45’ S e longitudes 49° 43’ e 57º 39’ W) e avaliar o padrão de distribuição espaço-temporal destas métricas sob influência da variabilidade climática subtropical. A obtenção de 11 métricas fenológicas foi realizada com base na série temporal do EVI (Enhanced Vegetation Index) do sensor MODIS (Moderate Resolution Imaging Spectroradiometer), para o período de 2001 a 2014, por meio do programa Timesat. Identificou-se a presença de dois grupos espaciais de tipologias campestres no Estado do Rio Grande do Sul, um na região centro-sul e outro na região nordeste, ao longo do litoral e no oeste do Estado. Observou-se ainda que o padrão fenológico da vegetação de pastagens do Rio Grande do Sul é controlado pela sazonalidade da vegetação associada, principalmente, à variação da temperatura do ar.

ABSTRACT Considering that plant phenology studies allow establishing relationships between phenological patterns of vegetation and changes caused by climate variability, the aim of this study was to obtain phenological metrics for the predominant grassland typologies in Rio Grande do Sul State, Brazil (latitudes 27º 05’ and 33º 45’ S and longitudes 49° 43’ and 57º 39’ W) and to evaluate the spatial-temporal distribution pattern of these metrics under the influence of the subtropical climate variability. The phenological metrics were obtained based on the time series of the EVI (Enhanced Vegetation Index), of the sensor MODIS (Moderate Resolution Imaging Spectroradiometer), for the period from 2001 to 2014, through the Timesat program. Eleven phenological metrics were extracted, identifying the presence of two spatial distribution patterns of grass-dominated typologies in the Rio Grande do Sul state, Brazil, one located in the south-central region and the other located in the northeast, along the coast and in the western portion of the state. In addition, it was also observed that the phenological pattern of the grassland vegetation of the state of Rio Grande do Sul, Brazil, is controlled by the seasonality of vegetation, mainly associated with the variations in air temperature.

ABSTRACT: This study aimed to characterize the average seasonal pattern of the vegetation in southern grassland in Brazil, and the variability found in the time series of vegetation indices. It also sought to identify similarities in the seasonal pattern of different grassland typologies. Moderate Resolution Imaging Spectroradiometer (MODIS), Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) images from Feb to Dec, 2000 to 2014 were analyzed for ten regions. The grassland typologies studied showed EVI and NDVI profiles consistent with the seasonal dynamics of grassland vegetation under the influence of a subtropical climate, with highest values in the indices during the warm seasons of the year (spring and summer) and lowest in the colder seasons (autumn and winter). Considering the values of EVI and NDVI, grassland typologies were allocated to four groups with similar temporal profiles. Among the groups formed from the EVI index it is possible to identify differences between grassland typologies during the autumn and winter, while the NDVI showed differences only in winter as compared to the other seasons.

Com este trabalho objetivou-se identificar a dinâmica fenológica das principais tipologias vegetais do Rio Grande do Sul, para o período de 2000 a 2010, utilizando-se dados de Enhanced Vegetation Index, através da transformada de ondaleta. A identificação da fenologia em ciclos ou padrões sazonais em séries temporais de índices de vegetação, obtidos por sensores orbitais, permite a observação de anomalias e os efeitos de mudanças climáticas ou ambientais. Um perfil temporal do Enhanced Vegetation Index foi construído para o Rio Grande do Sul e retiradas amostras para quatro tipologias vegetais: campo nativo, floresta ombrófila mista, cultivo de soja e de arroz. Essas amostras foram submetidas à transformada de ondaleta, que permitiu a decomposição da série e apresentação dos dados em relação ao tempo e frequência com que os eventos fenológicos ocorreram. Os dados apresentaram regularidade na dinâmica das tipologias vegetais testadas, com ciclos anuais de maior vigor e crescimento vegetal nas estações de primavera e verão e menor no outono e inverno.

The objective of this study was the identification of the phenology dynamics of the main types of vegetation of Rio Grande do Sul state, for the period from 2000 to 2010, using Enhanced Vegetation Index data through the wavelet transform. The identification of cycles or seasonal patterns in time series of vegetation indices obtained by orbital sensors allows the observation of anomalies and effects of climate and environmental change. A temporal profile of Enhanced Vegetation Index was built for the Rio Grande do Sul region, where samples of the four main plant typologies were selected: native grassland, mixed ombrophilous forest, soybean and rice crop. These samples were submitted to the wavelet transform, which allowed the decomposition of the series and presentation of data in relation to time and frequency with which the phenological events have occurred. The data showed regularity in the dynamics of vegetation types tested, with annual cycles of plant growth and higher Enhanced Vegetation Index values in spring and summer and lower Enhanced Vegetation Index values in autumn and winter.