Abstract
Efficient, more accurate reporting of maize (Zea mays L.) phenology, crop condition, and progress is crucial for agronomists and policy makers. Integration of satellite imagery with machine learning models has shown great potential to improve crop classification and facilitate in-season phenological reports. However, crop phenology classification precision must be substantially improved to transform data into actionable management decisions for farmers and agronomists. An integrated approach utilizing ground truth field data for maize crop phenology (2013–2018 seasons), satellite imagery (Landsat 8), and weather data was explored with the following objectives: (i) model training and validation—identify the best combination of spectral bands, vegetation indices (VIs), weather parameters, geolocation, and ground truth data, resulting in a model with the highest accuracy across years at each season segment (step one) and (ii) model testing—post-selection model performance evaluation for each phenology class with unseen data (hold-out cross-validation) (step two). The best model performance for classifying maize phenology was documented when VIs (NDVI, EVI, GCVI, NDWI, GVMI) and vapor pressure deficit (VPD) were used as input variables. This study supports the integration of field ground truth, satellite imagery, and weather data to classify maize crop phenology, thereby facilitating foundational decision making and agricultural interventions for the different members of the agricultural chain.
Highlights
Efficient, more accurate reporting of maize (Zea mays L.) phenology, crop condition, and progress is crucial for agronomists and policy makers
Combining satellite data features permit the generation of different vegetation indices (VIs), such as the Normalized Difference Vegetation Index (NDVI)[11], Enhanced Vegetation Index (EVI)[12], Green Chlorophyll Vegetation Index (GCVI)[13], Global Vegetation Moisture Index (GVMI)[14] and Normalized Difference Water Index (NDWI)[15], among many others
All possible combinations between variables were tested. This resulted in two models with the best performances, one composed by 5 vegetation indices and only one weather parameter, vapor pressure deficit (VPD), where the accuracy assessment ranged from 86 to 98%; and the other model composed only by latitude, longitude and doy, with an accuracy of 100% for almost all the season segments—years (Fig. 2)
Summary
More accurate reporting of maize (Zea mays L.) phenology, crop condition, and progress is crucial for agronomists and policy makers. The aim of this research study was to evaluate a classification of satellite-derived maize crop phenology and integrate in-season weather information to develop a classification model benchmarked with field survey data To achieve this overarching goal, we established the following objectives: (i) model training and validation—understand how different variables affect the model performance and identify the best combination of spectral features, weather parameters, geolocation and ground truth data, resulting in a model with the highest accuracy across years at each season segment (step one); and (ii) model testing—post-selection model performance evaluation for (a) each phenology class with unseen data (hold-out cross-validation); (b) temporal transferability; (c) spatial transferability (step two)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
