Frost Damage Detection in Sugarcane Crop Using Modis Images and Srtm Data

Brazil is the largest world producer of sugarcane which is used to produce almost equal proportions of either sugar (food) or ethanol (biofuel). In recent years sugarcane crop production has increased fast to meet the growing market demand for sugar and ethanol. This increase has been mainly due to expansion in crop area, but sugarcane production is also subjected to several factors that influence both the agricultural crop yield (tons of stalks/ha) and the industrial yield (kg of sugar/ton of stalks). Sugarcane is a semi-perennial crop that experiences major growth during spring and summer seasons with large demands for water and high temperatures to produce good stalk formation (crop yield). The harvest is performed mainly during fall and winter seasons when water availability and temperature should be low in order to accumulate sucrose in the stalks (industrial yield). These favorable climatic conditions for sugarcane crop are found in several regions in Brazil, particularly in São Paulo state, which is the major sugarcane producer in Brazil being responsible for almost 60% of its production. Despite the favorable climate in São Paulo state there is a certain probability of frost occurrence from time to time that has a negative impact on sugarcane crop, particularly on industrial yield, reducing the amount of sugar in the stalks; having consequences on price increase and product shortage. To evaluate the impact of frost on sugarcane crop, in the field, on a state level, is not a trivial task; however, this information is relevant due to its direct impact on the consumer market. Remote sensing images allow a synoptic view and present great potential to monitor large sugarcane plantations as has been done since 2003 in São Paulo state by the Canasat Project with Landsat type images (http://www.dsr.inpe.br/laf/canasat/en/). Images acquired from sensors with high temporal resolution such as MODIS (Moderate-Resolution Imaging Spectroradiometer) present the potential to detect the impact of climatic effects, such as frost, on crop growth, which is relevant information to evaluate the negative impact on sugarcane production. Thus, the objective of the present study is to detect the impact of the frost occurred on 28 June 2011 in the sugarcane production region of São Paulo state, using MODIS images acquired on board of Terra and Aqua satellites before and after the frost event. Also, Landsat type images were used to map the harvested sugarcane fields up to the frost event based on a sugarcane crop map for year 2011. The remaining sugarcane fields available for harvest in 2011 were monitored with the MODIS images acquired on 17, 19, 27, 28 June and 8 and 9 July, to detect frost damage. Field work was conducted shortly after frost occurrence to identify sugarcane fields with frost damage for training and validation purposes. MODIS images transformed to vegetation indices and morphometric variables extracted from SRTM (Shuttle Radar Topography Mission) data are being analyzed to detect and quantify the damage of the frost from 28 July 2011 on sugarcane crop.