BACKGROUND The oriental migratory locust is a major crop pest across eastern and south-eastern Asia. Metarhizium anisopliae is an effective biopesticide agent used for locust control, but its performance is temperature dependent, and thus can be more variable than chemical pesticide performance. To predict biopesticide performance for the control of the oriental migratory locust, we adapted a previous temperature-dependent model and validated it using field trial data. To increase the applicability of this model, we explored the use of readily available temperature variables, as well as our own satellite-derived canopy temperature variable, to run the model.RESULTS Compared to collected in situ temperature data, our canopy temperature variable most accurately represented the ambient temperature experienced by the locust. When the biopesticide performance model was run using this canopy temperature and compared to field trials results, the model predictions were more accurate than when the model was run with the other temperature variables. The accuracy of the biopesticide performance model was impacted by vegetation cover, but across the areas most associated with locust oviposition, growth and migration, the model predictions were satisfactorily accurate to guide biopesticide operational use.CONCLUSION We validated the model in six provinces in China, representing the three agro-ecological zones largely representative of the oriental migratory locust problem areas in China, Thailand, Cambodia and Vietnam. Whilst further validation work is needed, this model could be used in these countries to assess, at a fine spatial scale, the appropriateness of M. anisopliae for controlling the oriental migratory locust.

Smallholder farmers are the mainstay of the agricultural economies of sub-Saharan Africa (SSA), where they produce several crops, predominantly centered on maize. Smallholder productivity remains limited resulting from a range of confounding factors, but a primary cause is loss from pests and diseases, particularly insects. To improve productivity, recommendations for the mitigation of crop loss globally include early-warning and management systems for in-season indigenous pests. There are many early-warning systems in temperate regions; however, such systems are poorly established in Africa. This is in part due to the need for a combination of pest modeling, data handling and dissemination infrastructure, capacity, and resource provision. While each of these components is progressing in Africa, the means to successfully deploy such systems remain limited. To bridge this, the development of the Pest Risk Information Service (PRISE) began in 2017 for farmers in SSA. Implemented in Kenya, Ghana, Malawi, and Zambia, PRISE developed temperature-driven phenology models for major maize, bean, and tomato pests. Using downscaled and processed Earth Observation data to drive the models, PRISE partnered with African national agencies to communicate pre- and in-season pest alerts that forecast the time to act against key insect pests. Alerts were designed to be integrated into country-specific Good Agricultural Practice (GAP) recommendations to provide a complementary package to agricultural stakeholders. End line studies with farmers showed that those who received information about the target crops including PRISE pest forecasts, generally reported better outcomes in terms of reduced losses and increased incomes compared with farmers who did not.

Earth observation satellites are generating large amounts of data as they monitor the earthâ ˘A ´Zs environment. Large scale visualisation represents a powerful mechanism to explore and communicate this data; however, con-ventional tools and screens are limited by their size, while large shared screens are limited in their resolution. In this paper, we describe the motivations and design for a multi-panel video wall facility for the new International Space Innovation Centre. We further describe two visualisation applications which have been designed to demonstrate some of the potential of this wall to share data visualisations with a large audience, while providing a highly detailed view on the data

We report on a major expansion to JASMIN, a big data infrastructure upon which the UK Centre for Environmental Data Archival (CEDA) operates data centres and a major scientific analysis environment. The academic component of the facility for Environmental Monitoring from Space (CEMS) is hosted on JASMIN and continues to grow significantly, both in capability and its usage by the Earth Observation science community. We describe recent infrastructure upgrades to storage, compute and networking, and the deployment of a full private cloud. Dedicated data transfer nodes allow highperformance data flows to meet the needs of the climate and earth observation science communities. Over 20 EO science projects are currently active on JASMIN-CEMS, examples include near-real time atmospheric composition processing, global land surface products, and a collaborative research environment for land data assimilation (Optirad).