Search

Maize lethal necrosis (MLN) is a severe disease in maize that significantly reduces yields by up to 90% in maize-growing regions such as Kenya and other countries in Africa. The disease causes chlorotic mottling of leaves and severe stunting which leads to plant death. The spread of MLN in the maize-growing regions of Kenya has intensified since the first outbreak was reported in September 2011. In this study, the RapidEye (5 m) imagery was combined with field-based data of MLN severity to map three MLN severity levels in Bomet County, Kenya. Two RapidEye images were acquired during maize stem elongation and inflorescence stages, respectively, and thirty spectral indices for each RapidEye time step were computed. A two-step random forest (RF) classification algorithm was used to firstly create a maize field mask and to predict the MLN severity levels (mild, moderate, and high). The RF algorithm yielded an overall accuracy of 91.0%, representing high model performance in predicting the MLN severity levels in a complex cropping system. The normalized difference red edge index (NDRE) was highly sensitive to MLN detection and demonstrated the ability to detect MLN-caused crop stress earlier than the normalized difference vegetation index (NDVI) and the green normalized difference vegetation index (GNDVI). These results confirm the potential of the RapidEye sensor and machine learning to detect crop disease infestation rates and for use in MLN monitoring in fragmented agro-ecological landscapes. [ABSTRACT FROM AUTHOR]

Understanding the detailed timing of crop phenology and their variability enhances grain yield and quality by providing precise scheduling of irrigation, fertilization, and crop protection mechanisms. Advances in information and communication technology (ICT) provide a unique opportunity to develop agriculture-related tools that enhance wall-to-wall upscaling of data outputs from point-location data to wide-area spatial scales. Because of the heterogeneity of the worldwide agro-ecological zones where crops are cultivated, it is unproductive to perform plant phenology research without providing means to upscale results to landscape-level while safeguarding field-scale relevance. This paper presents an advanced, reproducible, and open-source software for plant phenology prediction and mapping (PPMaP) that inputs data obtained from multi-location field experiments to derive models for any crop variety. This information can then be applied consecutively at a localized grid within a spatial framework to produce plant phenology predictions at the landscape level. This software runs on the 'Windows' platform and supports the development of process-oriented and temperature-driven plant phenology models by intuitively and interactively leading the user through a step-by-step progression to the production of spatial maps for any region of interest in sub-Saharan Africa. Maize (Zea mays L.) was used to demonstrate the robustness, versatility, and high computing efficiency of the resulting modeling outputs of the PPMaP. The framework was implemented in R, providing a flexible and easy-to-use GUI interface. Since this allows for appropriate scaling to the larger spatial domain, the software can effectively be used to determine the spatially explicit length of growing period (LGP) of any variety. [ABSTRACT FROM AUTHOR]