Geosciences and statistics were applied to develop predictive models to study the areas of risk to soybean rust （Phakopsora pachyrhizi Sydow） in soybean （Glycine max L.）, coffee leaf rust （Hemileia vastatrix Berk...Geosciences and statistics were applied to develop predictive models to study the areas of risk to soybean rust （Phakopsora pachyrhizi Sydow） in soybean （Glycine max L.）, coffee leaf rust （Hemileia vastatrix Berk ＆ Br） in coffee and Black Sigatoka （Mycosphaerella fijiensis var. difformis） in banana, considering to Brazilian climatic characterization and distribution of soybean, coffee and banana crops in the period of observed data of 1950 to 2000 and A2 climate change scenarios of simulated data of 2020, 2050 and 2080. The technique of principal components allowed generating 1 variable based on 57 variables in order to determine an index explaining 87%, 88% and 90% of the variability of soybean, coffee and banana crops in Brazilian municipal districts. The climatic model of each disease was used to generate the zoning of the coffee rust, soybean rust and black sigatoka based on temperature and leaf wetness. Areas of favorability of the diseases were plotted inside to the main coffee, soybean and banana growing in Brazil. The applied methodology enabled to visualize the variation of favorable areas of epidemics according to the studied scenarios of climate change.展开更多
文摘Geosciences and statistics were applied to develop predictive models to study the areas of risk to soybean rust （Phakopsora pachyrhizi Sydow） in soybean （Glycine max L.）, coffee leaf rust （Hemileia vastatrix Berk ＆ Br） in coffee and Black Sigatoka （Mycosphaerella fijiensis var. difformis） in banana, considering to Brazilian climatic characterization and distribution of soybean, coffee and banana crops in the period of observed data of 1950 to 2000 and A2 climate change scenarios of simulated data of 2020, 2050 and 2080. The technique of principal components allowed generating 1 variable based on 57 variables in order to determine an index explaining 87%, 88% and 90% of the variability of soybean, coffee and banana crops in Brazilian municipal districts. The climatic model of each disease was used to generate the zoning of the coffee rust, soybean rust and black sigatoka based on temperature and leaf wetness. Areas of favorability of the diseases were plotted inside to the main coffee, soybean and banana growing in Brazil. The applied methodology enabled to visualize the variation of favorable areas of epidemics according to the studied scenarios of climate change.
