Knowing where, when, and how much rice is planted and harvested is crucial information for understanding the effects of policy, trade, and global and technological change on food security. We developed RiceAtlas, a spatial database on the seasonal distribution of the world’s rice production. It consists of data on rice planting and harvesting dates by growing season and estimates of monthly production for all rice-producing countries. Sources used for planting and harvesting dates include global and regional databases, national publications, online reports, and expert knowledge. Monthly production data were estimated based on annual or seasonal production statistics, and planting and harvesting dates. RiceAtlas has 2,725 spatial units. Compared with available global crop calendars, RiceAtlas is nearly ten times more spatially detailed and has nearly seven times more spatial units, with at least two seasons of calendar data, making RiceAtlas the most comprehensive and detailed spatial database on rice calendar and production.

Weeds are a crucial threat to agriculture, and in order to preserve crop productivity, spreading agrochemicals is a common practice with a potential negative impact on the environment. Methods that can support intelligent application are needed. Therefore, identification and mapping is a critical step in performing site-specific weed management. Unmanned aerial vehicle (UAV) data streams are considered the best for weed detection due to the high resolution and flexibility of data acquisition and the spatial explicit dimensions of imagery. However, with the existence of unstructured crop conditions and the high biological variation of weeds, it remains a difficult challenge to generate accurate weed recognition and detection models. Two critical barriers to tackling this challenge are related to (1) a lack of case-specific, large, and comprehensive weed UAV image datasets for the crop of interest, (2) defining the most appropriate computer vision (CV) weed detection models to assess the operationality of detection approaches in real case conditions. Deep Learning (DL) algorithms, appropriately trained to deal with the real case complexity of UAV data in agriculture, can provide valid alternative solutions with respect to standard CV approaches for an accurate weed recognition model. In this framework, this paper first introduces a new weed and crop dataset named Chicory Plant (CP) and then tests state-of-the-art DL algorithms for object detection. A total of 12,113 bounding box annotations were generated to identify weed targets (Mercurialis annua) from more than 3000 RGB images of chicory plantations, collected using a UAV system at various stages of crop and weed growth. Deep weed object detection was conducted by testing the most recent You Only Look Once version 7 (YOLOv7) on both the CP and publicly available datasets (Lincoln beet (LB)), for which a previous version of YOLO was used to map weeds and crops. The YOLOv7 results obtained for the CP dataset were encouraging, outperforming the other YOLO variants by producing value metrics of 56.6%, 62.1%, and 61.3% for the mAP@0.5 scores, recall, and precision, respectively. Furthermore, the YOLOv7 model applied to the LB dataset surpassed the existing published results by increasing the mAP@0.5 scores from 51% to 61%, 67.5% to 74.1%, and 34.6% to 48% for the total mAP, mAP for weeds, and mAP for sugar beets, respectively. This study illustrates the potential of the YOLOv7 model for weed detection but remarks on the fundamental needs of large-scale, annotated weed datasets to develop and evaluate models in real-case field circumstances.

Precise phenological calendars, for each cultivated species and variety, are necessary both to highlight anomalous agronomic situations and to feed crop models. This study, conducted in the Italian rice area, focuses on the evaluation of the contribution of remote sensing satellite data to providing phenological information on rice cropping systems. A time series of 5 years (2001–2005) of the Moderate Resolution Imaging Spectrometer (MODIS) Normalized Difference Vegetation Index (NDVI) 16-day composites was analysed with the TIMESAT program in order to automatically retrieve key phenological information such as the start of season (emergence), peak (heading) and end of season (maturity). The procedure involved two steps: (1) interpolation and smoothing of MODIS NDVI temporal profile and (2) the analysis of a temporal signal for the extraction of the phenological metrics. The remote sensing estimates were evaluated using information regarding cultivated variety, sowing dates, management and production directly acquired from rice farmers. A good correlation (R ² = 0.92, n = 24) has been observed between estimates derived from satellites and estimates produced with the traditional Growing Degree Days (GDD) method based on thermal unit accumulation. Improved estimates of the maturity stage were obtained using a procedure that integrates satellite and GDD methods; however its application requires spatially distributed information on the cultivated varieties. Satellite derived maps of the retrieved phenological parameters showed an intra-seasonal pattern related to different cultivated varieties. Inter-seasonal analysis allowed the anomalous behaviour of the year 2003 to be highlighted, characterized by rapid growth at the beginning of the spring and an early senescence. The results confirm the potential of remotely sensed data for the monitoring of crop status and for the forcing of crop models in a spatially distributed way.

Identifying managed flooding in paddy fields is commonly used in remote sensing to detect rice. Such flooding, followed by rapid vegetation growth, is a reliable indicator to discriminate rice. Spectral indices (SIs) are often used to perform this task. However, little work has been done on determining which spectral combination in the form of Normalised Difference Spectral Indices (NDSIs) is most appropriate for surface water detection or which thresholds are most robust to separate water from other surfaces in operational contexts. To address this, we conducted analyses on satellite and field spectral data from an agronomic experiment as well as on real farming situations with different soil and plant conditions. Firstly, we review and select NDSIs proposed in the literature, including a new combination of visible and shortwave infrared bands. Secondly, we analyse spectroradiometric field data and satellite data to evaluate mixed pixel effects. Thirdly, we analyse MODIS data and Landsat data at four sites in Europe and Asia to assess NDSI performance in real-world conditions. Finally, we test the performance of the NDSIs on MODIS temporal profiles in the four sites. We also compared the NDSIs against a combined index previously used for agronomic flood detection. Analyses suggest that NDSIs using MODIS bands 4 and 7, 1 and 7, 4 and 6 or 1 and 6 perform best. A common threshold for each NDSI across all sites was more appropriate than locally adaptive thresholds. In general, NDSIs that use band 7 have a negligible increase in Commission Error over those that use band 6 but are more sensitive to water presence in mixed land cover conditions typical of moderate spatial resolution analyses. The best performing NDSI is comparable to the combined index but with less variability in performance across sites, suggesting a more succinct and robust flood detection method.