Temporal dynamics of vegetation indexes and their correlation with precipitation

Name: ROSANE GOMES DA SILVA
Type: MSc dissertation
Publication date: 26/02/2016
Advisor:

Namesort descending Role
ALEXANDRE ROSA DOS SANTOS Advisor *

Examining board:

Namesort descending Role
ALEXANDRE ROSA DOS SANTOS Advisor *
NILTON CESAR FIEDLER Internal Examiner *

Summary: Forests are areas of great environmental importance because its ecosystem enables the maintenance of various species of fauna and contributes to the quality of soil and water resources. Climate changes are main change agents in vegetation dynamics, influencing the distribution, structure and function of vegetation, which suggests a devaluation in the form of goods and services, extending the impact on socio-economic and ecosystem. In this context, increasingly important become research to study the dynamics of vegetation behavior and its relationship with the weather. The objective of this study was to analyze the behavioral trend of vegetation in atlantic forest biome, by the MODIS vegetation indices and their correlation with the variability of monthly data satellite precipitation TRMM. The research was realized with the study area the Caparaó National Park and part of its buffer zone in the state of Espirito Santo. We used NDVI and EVI data MODIS sensor, MOD13Q1 product, the period from 2001 to 2014, totaling 322 images and satellite precipitation monthly data TRMM, the same period, totaling 168 images. The analysis of the interannual trends of time series of vegetation index were performed by the methodologies of linearity , linear correlation, linear trend, Mann Kendall monotonic trend, median trend Theil-Sen and analysis of temporal profiles. The seasonal trend was verified by seasonal trend analysis (STA) and transform inverse wavelet Haar. By means of linear modeling techniques, expressed by R and R² calculated, the correlation between the data precipitation and vegetation index was studied. With the generation of the temporal profiles of the IVs, it was observed that there was a decrease in greenness, particularly in areas WHERE the vegetation has become more vigorous. This result was similar to study the interannual trends indicate that decrease in IR values for both monotonic trend Mann Kendall and for the median tendency, being a non-linear behavior in accordance with the correlation methodologies linear, linear and linear trend. According to the seasonal trend analysis were identified two seasonal cycles in the study area, an annual cycle and a semi-annual cycle. This result was the same found by wavelet transform to the EVI. For NDVI and precipitation was observed seasonal pattern of behavior by the wavelet transform. As for the correlation of vegetation indices with precipitation, larger correlation values were to 0.7 to 0.6 and R to R². However, in most of the area, especially considering the PARNA Caparaó, these values were very low. Thus, other factors may have influenced the dynamics of the behavior changes of vegetation in the period considered.

Key words: remote sensing, meteorological variables, atlantic forest, protect areas.

Access to document

Acesso à informação
Transparência Pública

© 2013 Universidade Federal do Espírito Santo. Todos os direitos reservados.
Av. Fernando Ferrari, 514 - Goiabeiras, Vitória - ES | CEP 29075-910