Search

After 15 years, the Système Pour l'Observation de la Terre (SPOT)-VEGETATION (VGT) program reached the end of its life in May 2014 and was replaced by the Project for On-Board Autonomy-Vegetation (PROBA-V) mission. Exploiting the period of overlap between instruments, this study compares the normalized difference vegetation index (NDVI) of two instruments from the point of view of the user interested in operational crop monitoring. The comparison is performed for Morocco, Algeria, and Tunisia, where NDVI is used to derive anomaly maps, temporal profiles, and cereal yield forecasts. A relevant scatter due to unexplained unsystematic variability exists between anomaly values. A mismatch between anomaly classes is observed for 20%–30% of the crop area. However, when the NDVI is averaged over cropland and administrative units to derive temporal profiles, the two data sources show a high agreement. Results for yield estimation comparison indicate an overall high agreement, and both the (null) hypotheses that the model predictions and the root mean square error (RMSE) in yield estimation are not different, when using PROBA-V instead of SPOT-VGT, cannot be rejected in all cases for Morocco and Algeria. On the contrary, in Tunisia, where RMSE is lower using PROBA-V, the hypothesis of no difference in RMSE is rejected. These findings therefore indicate that yield estimation performances are not affected (Morocco and Algeria) or improved (Tunisia) by the source transition. Finally, despite the same nominal spatial resolution, the different spatial quality of the sensors was found to have an effect on yield estimation in areas characterized by sharp transitions between cropland and desert. [ABSTRACT FROM PUBLISHER]

This report shows the results of the first multi-year spatially distributed sensitivity analysis carried out on two complex agroecological models. Wheat is the staple food of 1.5 billion people worldwide. Projected trends in wheat global demand reveal risks of food security over the next decades. Systems for large-area crop monitoring and yield forecasting are needed to support agricultural policies, especially in developing countries. Among those crop systems, the most sophisticated ones are based on crop simulation models. Published reports of sensitivity analyses performed on different crop models show that parameters related to leaf area expansion are often considered as the most important. Here we show that, on the contrary, photosynthesis parameters are more relevant under the conditions explored. We carried out the sensitivity analysis on the models WOFOST and CropSyst for wheat simulation in Morocco. Due to the high number of model runs to be performed, a two-step procedure was adopted: The Morris method was used to identify parameters with a negligible effect, and then the Sobol method was applied on remaining parameters. Environmental and management information were obtained from the European Commission MARS database. Our results show that photosynthesis parameters explained more than 75 % of the total output variance for CropSyst and more than 70 % for WOFOST. On the contrary, parameters related to leaf area expansion were less relevant. Geographical patterns shown by sensitivity analysis results under heterogeneous conditions can help breeders to select specific plant traits, in order to develop phenotypes suitable for specific conditions, e.g., varieties with a higher level of thermal adaptation in the Southern regions. [ABSTRACT FROM AUTHOR]

Abstract: In Morocco, no operational system actually exists for the early prediction of the grain yields of wheat (Triticum aestivum L.). This study proposes empirical ordinary least squares regression models to forecast the yields at provincial and national levels. The predictions were based on dekadal (10-daily) NDVI/AVHRR, dekadal rainfall sums and average monthly air temperatures. The Global Land Cover raster map (GLC2000) was used to select only the NDVI pixels that are related to agricultural land. Provincial wheat yields were assessed with errors varying from 80 to 762kgha−1, depending on the province. At national level, wheat yield was predicted at the third dekad of April with 73kgha−1 error, using NDVI and rainfall. However, earlier forecasts are possible, starting from the second dekad of March with 84kgha−1 error, at least 1 month before harvest. At the provincial and national levels, most of the yield variation was accounted for by NDVI. The proposed models can be used in an operational context to early forecast wheat yields in Morocco. [Copyright &y& Elsevier]