Mapping the planting dates: An effort to retrive crop phenology information from MODIS NDVI time series in Africa

Obtaining a better understanding of the regional variation in local crop calendars; including the number of cropping seasons per year, and the planting and harvesting dates of specific crops, is important to a large number of organizations and individuals who are concerned with production, marketing, processing and trade of food and feed products. It is important to seed and input suppliers, for example, since these resources must be available to farmers in sufficient quantities, at the right time and in the right location. Furthermore, as crop growth models are increasingly used to explore the likely effects of different input and management interventions on crop yield, a scarce item of required input data is an estimate of the specific planting windows that best reflect local farming practices. Despite the importance of such information, there are limited sources of such data in a regional and global context. The situation is even worse in developing country, especially in Africa. The objective of this study is to evaluate various existing crop calendar data products and to design a strategy to geo-reference and harmonize the best of them using spatial extensions. Existing datasets are digitized and geo-referenced using ArcGIS tools. Spatial analysis is applied to disaggregate and examine the relationships among different data sources, and crop Cropping Calendars (assembled in part by use of remotely sensed data products) are finally generated at a pixel scale for sub-Saharan Africa.