Analysis of raining streaks to manage cane crops, Saccharum officinarum, in the Cauca River Valley, Colombia

Abstract

The equatorial conditions of the Cauca River Valley determine that sugarcane can be sown and harvested throughout the year. That is, there are no defined dates for sowing and harvesting, unlike what happens in other countries (zafra). In this scenario, the rainfall in addition to impacting production via physiological and epidemiological effects, determine the probability of mechanized tasks. A postponed mechanized labor (or not executed), because of high humidity in the soil, affects the production. Such an effect on productivity is due to the asynchronism between the ideal moment to carry out a task and the moment when it is executed. The descriptive analysis of two monthly climatic indicators derived from the daily precipitation data and the probabilistic analysis of one of them are proposed to prioritize agricultural management activities in situ. The probabilistic analysis was based on the parameterization of a negative binomial model truncated at zero for each month and for each climatic zone of the region under analysis. The results show that the average length of the streak and the average number of streaks for each month and each zone vary significantly. In this sense, the average values of these variables in the two rainiest areas of the region differ markedly, determining differential management for sugar cane, a crop of the maximum importance for the regional economy.