Forecasting the Best Time to Act Against Crop Pests – Innovative Initiatives in Africa
When is the best time to act against important crop pests? The Pest Risk Information Service (PRISE) offers digital advice.
We rely on weather forecasts to decide whether to wear suncream or take an umbrella out that morning. Hayfever sufferers use tools such as pollen forecasts to decide to take an antihistamine, or avoid parks and gardens that day. Whilst we readily rely on these advisory tools and information to make decisions, we don’t give much thought to the huge amount of data and processing that delivers it into the palms of our hands.
However, arguably for those with most at stake, there is a gap in such tool provision. Climate change, particularly in Sub-Saharan Africa (SSA), is resulting in inter- and intra-annual variability in weather patterns, causing traditional and localised knowledge of what farming actions to take and when to be unreliable, which affects productivity. One of the largest problems SSA smallholder farmers face year-on-year is the threat of loss to pests and diseases, which contribute an average of 40% crop loss globally, and economic losses of $1.3 trillion (Diagne et al., 2021). Controlling pests is best done preventatively, as when damage symptoms are observed, it is often too late for effective control. However, preventative control measures are expensive and ecologically damaging (Barzman et al., 2015; FAO, 2021).
The question the Center for Agriculture and Bioscience International (CABI) Digital Development team posed to help address this problem was: what actionable advice can we give to smallholder farmers, based on their local weather conditions and in an easily useable format, that helps them reduce losses to pests? To test this, we undertook a series of activities that resulted in the Pest Risk Information Service (PRISE): an early-warning information system that provides farmers with alerts on the best time to intervene in a crop for effective and efficient pest management.
PRISE uses biological models, Earth Observation (EO)-derived weather data and agronomic expertise to generate location-specific alerts on the optimum ‘time-to-act’ (TTA) window, allowing farmers and other agricultural stakeholders to prepare at the beginning of the growing season, and take action when they are most likely to achieve maximum kill of the pest in question. PRISE was started in 2017 and included four countries in SSA: Ghana, Kenya, Malawi, and Zambia. During the development of PRISE, the focus was on a selection of major insect pests of mixed-maize farming systems because it is the predominant smallholder cropping system in the four countries (Food and Agriculture Organization, 2024) and prioritised by key government stakeholders.
Pest Models
Phenological development models were created for the pests listed in Table 1 and two types of models were produced: the first model was used to calculate the time and duration in degree-days (the number of days an insect takes to complete its lifecycle, bound by upper and lower temperature thresholds). Each calculation accounts for the appearance of each of a series of cohorts of the larval stage that result from the first periods of successive daily egg laying early in the season. These times were then used to estimate the common period when the larval stage was present from all the cohorts. This period was used as the window for the application of a management intervention.
The second model was more detailed and only developed for three important pests that attack cash crops: fall armyworm (larvae of the moth spodoptera frugiperda), tomato leafminer, and bean fly ( (ophiomyi). This model uses an economic or action threshold (the ‘sweet spot’ when damage symptoms or pest presence is deemed severe enough to take action and still controllable with an intervention) as the optimum TTA. For this model, the population development and growth of the larval stage of each species from crop planting were related to cumulative degree-days, specific to each species. This was to enable the prediction of when economic or action thresholds had been reached. Models were scientifically validated through a series of field trials across countries and seasons. The output of the models - the TTA alert – was then converted into a number of days to act from planting date. More detail on the model methodologies can be seen in (Day, 2023).
Table 1: Crops and pests included in PRISE (2017–2022)
This table lists pests that damage maize, beans and tomatoes and at which devlopmental stage (i.e. as larvae or fully grown insects) they should be controlled with what substances.
SMS-Alerts for Pest Control as Part of Good Agricultural Practice
Since it began in 2017, PRISE has reached over 2 million smallholder farmers. We have developed two pathways to disseminated TTA alerts: Firstly, via bespoke spatial and temporally specific SMS mobile alerts early during the growing season. Here are some examples:
These services provide relevant, timely good agricultural practice (GAP) messages such as advice on choice of seed variety, land preparation, planting, irrigation and nutrient management, harvesting, storage etc. It was decided to partner with platforms already doing this because it allows for augmenting advice already being serviced to farmers, reaching large numbers of users quickly, whilst remaining service agnostic. Currently, in partnership with KALRO in Kenya, we are reaching nearly 1 million farmers each month with location-specific TTA alerts. Secondly, we also reach farmers via PRISE bulletins: shared to agricultural extension advisors who engage directly with farmers, farmer groups, cooperatives and other groups. This method ensures PRISE is more likely to be used by the often hardest-to-reach farmers, as it ultimately relies on the existing traditional face-to-face communication pathways most trusted and relied upon by farmers. Both pathways encourage PRISE to be embedded in GAP already serviced to farmers.
Figure 3
Example TTA values generated for Phthorimaea absoluta (tomato leafminer) in August 2022 spatially averaged over administrative districts in the four PRISE countries. Colour represent the number of days between planting and recommended time to act, and the scale varies between different countries due to differences in temperature ranges.
Impact on Crops and Income
CABI is continually interested to understand and improve the depth and breadth of our work’s impact. Results from a working paper (Khonje, 2024 in prep) from a 2023 randomised controlled trial study examine the effectiveness of delivering PRISE time-to-act pest alerts and messages on good agricultural practices (GAP) among 1,300 tomato farmers in Kenya via lead farmers within farming communities. Two treatment groups received pest information on: (i) tomato GAP only, and (ii) GAP plus time-to-act pest alerts. After 15 weeks of intervention, we find that compared to control farmers, the first group increased yields (kg/ha) by 29% and total net income by 52%. The second group increased yield by 45% and total net income by 62%. The second group also reduced pesticide spraying roughly by 11%, frequency of spraying by 14%, and pesticide costs by 12%.
This is an exciting set of findings for our work, and corroborates earlier findings related to the benefits of PRISE. We are now looking forward to scaling PRISE to new countries, particularly across Sub-Saharan Africa, working with national research partners to increase the range of pests in the PRISE portfolio, and with dissemination partners to extend the reach of PRISE time-to-act alerts to more farmers.
Next steps and partnerships
The PRISE project was funded from 2017 to 2022 by the UK Space Agency International Partnerships Programme and from 2022 by the CABI PlantwisePlus program as a collaboration between CABI and Assimila LTD. Among the donors of CABI are ministries and development agencies from Great Britain, Australia, Canada, the Netherlands and Switzerland.
Partnerships are key to PRISE. We are looking for national public and private agro-advisory services to join us as implementing partners to integrate PRISE models into their platforms. The models we have already created (see Table 1) can be applied anywhere and we are looking to scale these to new countries, initially focussing on other East and Southern Africa as the models are well validated for these regions and require little work to establish.
We are also exploring an approach to training national agricultural researchers to develop additional pest models that are important challenges to your farming community. This includes agricultural, horticultural and perennial crops, and covers both insects and pathogens.
We are interested to work on the development of innovative models that go beyond pest alerts. Using historical climatology data, we can support decision making throughout the production process and beyond the farmgate, such as appropriate site selection, yield modelling, drought and flooding risk, as well as input distribution, financial inclusion services and more.
All views expressed in the Welternährung are those of the authors and do not necessarily reflect the view or policies of the editorial board or of Welthungerhilfe.
References:
Barzman, M., Bàrberi, P. Birch, A.N.E., Boonekamp, P., Dachbrodt-Saaydeh, S., Graf, B., Hommel, B., Jensen, J.E., Kiss, J., Lamichhane, J.R., Messéan, A., Moonen A.C., Ratnadass, A., Ricci, P., Sarah, J.L. and Sattin, M. (2015). Eight principles of integrated pest management. Agronomy for Sustainable Development 35: 1199–1215.
Day, C., Murphy, S.T., Styles, J., Taylor, B., Beale, T., Holland, W., Williams, F., Shaw, A., Finegold, C., Oronje, M., Oppong-Mensah, B., Phiri, N., Lowry, A., Finch, E.A., Mahony, J., Wood, S., Durocher-Granger, L., Chacha, D., Maczey, N., Gonzalez-Moreno, P., Thomas, S.E., Beeken, J., Lewis, J., Saldana, G.L., Duah, S., Bundi, M., Wasilwa, L., Amata, R., Musila, R., Mutisya, D., Gitonga, C.K., Kalama, P., Nyasani, J.O., Matimelo, M., Mgomba, H., Gaitu, C., Ocloo, C., Adjei-Mensah, I., Ohene-Mensah, G., Nboyine, J.A. and Susuwele, B. (2024). Forecasting the population development of within-season insect crop pests in Sub-Saharan Africa: The Pest Risk Information Service. Journal of Integrated Pest Management 15(1): 7.
FAO, IFAD, UNICEF, WFP and WHO. (2024). The state of food security and nutrition in the World 2024 – Financing to end hunger, food insecurity and malnutrition in all its forms. Rome. doi.org/10.4060/cd1254en.
FAO. (2021). The impact of disasters and crises on agriculture and food security: 2021. Rome. https://doi.org/10.4060/cb3673en.
Khonje, M. G., Tambo, J. A., Chacha, D.,Mbugua., Holland, W. Lowry, A. Beale, T., Taylor, B., Day, C. Williams, F. (2024 in prep) Pest information and farm performance: Experimental evidence from Kenya. American Journal of Agricultural Economics.