Pest risk and diseases

Crop pests and diseases are widespread globally and are estimated cause global productivity losses of up to 40%. Pest outbreaks are devastating, respect no political boundaries and are becoming increasingly unpredictable due to climate change and are exacerbated by globalisation, international trade and population growth. The losses caused by crop pests impact farmer livelihoods and international food supply chains globally. Whilst these losses threaten food safety in developed countries, the impacts of these losses are more prolific in developing countries. This is mainly due to the heavy reliance of rural economies and community development on agricultural productivity, and as such, crop losses have been found to hamper the pursuit of food security, improved nutrition and the ending of poverty.

Crop pests and diseases are widespread globally and are estimated cause global productivity losses of up to 40%. Pest outbreaks are devastating, respect no political boundaries and are becoming increasingly unpredictable due to climate change and are exacerbated by globalisation, international trade and population growth. The losses caused by crop pests impact farmer livelihoods and international food supply chains globally. Whilst these losses threaten food safety in developed countries, the impacts of these losses are more prolific in developing countries. This is mainly due to the heavy reliance of rural economies and community development on agricultural productivity, and as such, crop losses have been found to hamper the pursuit of food security, improved nutrition and the ending of poverty.

The development of these pests and diseases is strongly linked to environmental conditions, with temperature, rainfall and humidity amongst some of the most important factors. Analysing these parameters allows pest development to be modelled and predicted. Traditionally, methods to do this have used environmental data obtained from a network of in situ meteorological stations, but their sparse geographic coverage limits the applicability of this approach, particularly when applied to large rural areas. To overcome these issues, there is a growing recognition that Earth Observation (EO) data should be used in this sector, due to its ability to provide well-calibrated and spatially continuous data.

Assimila is working with CABI, a global leader in the field of international development and agriculture, and numerous other partners, to combine EO and meteorological data with biological models in an attempt to reduce crop losses due to pests and work towards achieving a number of the UN’s Sustainable Development Goals. There are two main approaches that we are currently developing.

PRISE

The first forms the core of a Pest Risk Information SErvice (PRISE). PRISE is a 5-year project funded by the UK Space Agency’s International Partnership Programme (IPP) currently being conducted in five Sub-Saharan African countries (Kenya, Ghana, Zambia, Malawi and Rwanda).  This project aims to predict a farmer’s change in pest risk throughout the growing season so that effective management advice can be provided to a smallholder farmer in a timely manner, resulting in an improvement in yields. An example of this approach for Fall Armyworm, an invasive caterpillar/moth that devastates maize, uses land surface temperature data to drive a biological model that predicts the change in larval population, so that when a critical level is reached, actionable advice can be sent to farmers (outputs shown in figure 1).

PRISE warning map showing Fall Army Worm emergence
BioSuccess

BioSuccess is being used in Colombia (for Coffee Berry Borer) and China (for Locusts and Wheat Rust) to support the transition from using chemical pesticides to biopesticides which are safer and more environmentally friendly.

Biopesticides are living organisms and their efficacy is strongly linked to environmental conditions. Assimila is using EO data to understand when the most suitable time for biopesticide application is given the predicted life stage of the pest, and to forecast how effective this application would be. These models estimate the growth of the biopesticide development under the environmental conditions at a given time. The example shows this, where the EO derived surface temperature has been converted into the hourly growth rate of the biopesticide. This rate is accumulated over time, until the cumulative amount reaches 1, at which point 90% of the pest population has been killed by the biopesticide. In the example, this point is 26 days after the application of the biopesticide.

Environmental model predicting the effective growth of a biopesticide
Locust infected with a biopesticide