Nitrogen (N) is one of the most important limiting nutrients for sugarcane production. Conventionally, sugarcane N concentration is examined using direct methods such as collecting leaf samples from the field followed by analytical assays in the laboratory. These methods do not offer real-time, quick, and non-destructive strategies for estimating sugarcane N concentration. Methods that take advantage of remote sensing, particularly hyperspectral data, can present reliable techniques for predicting sugarcane leaf N concentration. Hyperspectral data are extremely large and of high dimensionality. Many hyperspectral features are redundant due to the strong correlation between wavebands that are adjacent. Hence, the analysis of hyperspectral data is complex and needs to be simplified by selecting the most relevant spectral features. The aim of this study was to explore the potential of a random forest (RF) regression algorithm for selecting spectral features in hyperspectral data necessary for predicting sugarcane leaf N concentration. To achieve this, two Hyperion images were captured from fields of 6–7 month-old sugarcane, variety N19. The machine-learning RF algorithm was used as a feature-selection and regression method to analyse the spectral data. Stepwise multiple linear (SML) regression was also examined to predict the concentration of sugarcane leaf N after the reduction of the redundancy in hyperspectral data. The results showed that sugarcane leaf N concentration can be predicted using both non-linear RF regression (coefficient of determination, R ² = 0.67; root mean square error of validation (RMSEV) = 0.15%; 8.44% of the mean) and SML regression models (R ² = 0.71; RMSEV = 0.19%; 10.39% of the mean) derived from the first-order derivative of reflectance. It was concluded that the RF regression algorithm has potential for predicting sugarcane leaf N concentration using hyperspectral data.

The fall armyworm, Spodoptera frugiperda (FAW), first invaded Africa in 2016 and has since become established in many areas across the continent where it poses a serious threat to food and nutrition security. We re-parameterized the existing CLIMEX model to assess the FAW global invasion threat, emphasizing the risk of transient and permanent population establishment in Africa under current and projected future climates, considering irrigation patterns. FAW can establish itself in almost all countries in eastern and central Africa and a large part of western Africa under the current climate. Climatic barriers, such as heat and dry stresses, may limit the spread of FAW to North and South Africa. Future projections suggest that FAW invasive range will retract from both northern and southern regions towards the equator. However, a large area in eastern and central Africa is projected to have an optimal climate for FAW persistence. These areas will serve as FAW ‘hotspots’ from where it may migrate to the north and south during favorable seasons and then pose an economic threat. Our projections can be used to identify countries at risk for permanent and transient FAW-population establishment and inform timely integrated pest management interventions under present and future climate in Africa.

Desert locust outbreak in East Africa is threatening livelihoods, food security, environment, and economic development in the region. The current magnitude of the desert locust invasion in East Africa is unprecedented and has not been witnessed for more than 70 years. Identifying the potential breeding sites of the pest is essential to carry out cost-effective and timely preventive measures before it inflicts significant damage. We accessed 9,134 desert locust occurrence records and applied a machine-learning algorithm to predict potential desert locust breeding sites in East Africa using key bio-climatic (temperature and rainfall) and edaphic (sand and moisture contents) factors. Ten days greenness maps from February 2020 to April 2020 were overlaid in model outputs to illustrate the temporal evolution of breeding site locations. This study demonstrated that vast areas of Kenya and Sudan, north eastern regions of Uganda, and south eastern and northern regions of South Sudan are at high risk of providing a conducive breeding environment for the desert locust. Our prediction results suggest that there is need to target these high-risk areas and strengthen ground surveillance to manage the pest in a timely, cost-effective, and environmentally friendly manner.

The present study is the first modeling effort at a global scale to predict habitat suitability of fall armyworm (FAW), Spodoptera frugiperda and its key parasitoids, namely Chelonus insularis, Cotesia marginiventris,Eiphosoma laphygmae,Telenomus remus and Trichogramma pretiosum, to be considered for biological control. An adjusted procedure of a machine-learning algorithm, the maximum entropy (Maxent), was applied for the modeling experiments. Model predictions showed particularly high establishment potential of the five hymenopteran parasitoids in areas that are heavily affected by FAW (like the coastal belt of West Africa from Côte d’Ivoire (Ivory Coast) to Nigeria, the Congo basin to Eastern Africa, Eastern, Southern and Southeastern Asia and some portions of Eastern Australia) and those of potential invasion risks (western & southern Europe). These habitats can be priority sites for scaling FAW biocontrol efforts. In the context of global warming and the event of accidental FAW introduction, warmer parts of Europe are at high risk. The effect of winter on the survival and life cycle of the pest in Europe and other temperate regions of the world are discussed in this paper. Overall, the models provide pioneering information to guide decision making for biological-based medium and long-term management of FAW across the globe.

Leaf area index (LAI) is an important biophysical trait for forest ecosystem and ecological modeling, as it plays a key role for the forest productivity and structural characteristics. The ground-based methods like the handheld optical instruments for predicting LAI are subjective, pricy and time-consuming. The advent of very high spatial resolutions multispectral data and robust machine learning regression algorithms like support vector machines (SVM) and artificial neural networks (ANN) has provided an opportunity to estimate LAI at tree species level. The objective of the this study was therefore to test the utility of spectral vegetation indices (SVI) calculated from the multispectral WorldView-2 (WV-2) data in predicting LAI at tree species level using the SVM and ANN machine learning regression algorithms. We further tested whether there are significant differences between LAI of intact and fragmented (open) indigenous forest ecosystems at tree species level. The study shows that LAI at tree species level could accurately be estimated using the fragmented stratum data compared with the intact stratum data. Specifically, our study shows that the accurate LAI predictions were achieved for Hymenocardia ulmoides using the fragmented stratum data and SVM regression model based on a validation dataset (R²_Val = 0.75, RMSE_Val = 0.05 (1.37% of the mean)). Our study further showed that SVM regression approach achieved more accurate models for predicting the LAI of the six endangered tree species compared with ANN regression method. It is concluded that the successful application of the WV-2 data, SVM and ANN methods in predicting LAI of six endangered tree species in the Dukuduku indigenous forest could help in making informed decisions and policies regarding management, protection and conservation of these endangered tree species.