Research priorities for global food security under extreme events

Extreme events, such as those caused by climate change, economic or geopolitical shocks, and pest or disease epidemics, threaten global food security. The complexity of causation, as well as the myriad ways that an event, or a sequence of events, creates cascading and systemic impacts, poses significant challenges to food systems research and policy alike. To identify priority food security risks and research opportunities, we asked experts from a range of fields and geographies to describe key threats to global food security over the next two decades and to suggest key research questions and gaps on this topic. Here, we present a prioritization of threats to global food security from extreme events, as well as emerging research questions that highlight the conceptual and practical challenges that exist in designing, adopting, and governing resilient food systems. We hope that these findings help in directing research funding and resources toward food system transformations needed to help society tackle major food system risks and food insecurity under extreme events.

•

We asked experts for top threats to global food security from extreme events

•

We find unresolved governance challenges underpin many of the key threats

•

We also asked experts for top outstanding research priorities on this topic

•

Coordination to design, adopt, and govern resilient food systems is needed

Heat waves, floods, droughts, pest outbreaks and diseases, financial crises, and human conflicts are threatening the production and supply of food around the world. These extreme events are on the rise, and our ability to prepare for them seems limited. Multiple events occurring at the same time compound the problem. Research teams and policymakers are developing solutions to improve the resilience of food systems, but this is often done in isolation—tackling one problem at a time. In this article we bring together food system experts to identify the top threats over the next two decades and priority research questions to address them. We find that unresolved governance challenges in international relations underpin many of the key threats and that coordinated research is needed to help design and adopt systems of governance for food systems that are resilient to extreme events in the future.

Spatial Correlations Don't Predict Changes in Agricultural Ecosystem Services: A Canada-Wide Case Study

Improving the management of multiple ecosystem services (e.g., food provision, water and air quality regulation, carbon storage, and erosion control) in agricultural landscapes is a critical challenge to improve food system sustainability. However, we currently lack spatially-resolved national-level assessments of the relationships among services in agricultural landscapes over time. This limits our ability to make decisions and predict how environmental changes or agricultural management actions will impact multiple services. How do multiple ecosystem services vary across both space and time, at regional-to-national scales? To address this question, we quantified eight indicators of four ecosystem services across 290 Canadian agricultural landscapes in 1996, 2001, and 2006. We observed consistent correlations between pairs of services across the 290 ecodistricts in each of the 3 years of our study. In particular, ecodistricts with high livestock production had low provision of most regulating services, while ecodistricts with high air quality (ammonia retention) also had high soil and water quality regulation services. However, these ‘snapshot’ correlations poorly predicted how pairs of services changed through time. Ecosystem service change from 1996–2001 to 2001–2006 (as measured by pairwise correlations) showed markedly different patterns than snapshot correlations. In particular, where livestock production increased between years, so did most regulating services. Ecosystem service bundles also showed similar divergent patterns. The distribution of ecosystem service “snapshot” bundles—sets of ecodistricts with similar levels of provision across multiple ecosystem services in a single year—was generally stable between 1996 and 2006; only 15% of ecodistricts changed bundle types in this time period. However, ecosystem service “change” bundles—sets of ecodistricts with similar changes in ecosystem service provision through time—were much more dynamic. Nearly 60% of ecodistricts exhibited a different set of ecosystem service changes from 2001 to 2006 compared to 1996 to 2001. Our results add to the growing evidence that relationships between services across space do not necessarily predict service change through time. Improved understanding of the spatial patterns and temporal dynamics of ecosystem services, and better understanding of underlying processes, is crucial to improve agricultural landscape management for multifunctionality and sustainability.

A coupled stochastic/deterministic model to estimate the evolution of the risk of water contamination by pesticides across Canada

Periodic assessments of the risk of water contamination by pesticides help decision makers improve the sustainability of agricultural management practices. In Canada, when evaluating the risk of water contamination by pesticides, 2 main constraints arise. First, because the area of interest is large, a pesticide transport model with low computational running time is mandatory. Second, some relevant input data for simulations are not known, and most are known only at coarse scale. This study aims to develop a robust methodology to estimate the evolution of the risk of water contamination by pesticides across Canada. To circumvent the 2 aforementioned issues, we constructed a stochastic model and coupled it to the 1-dimensional pesticide fate model Pesticide Root Zone Model (PRZM). To account for input data uncertainty, the stochastic model uses a Monte Carlo approach to generate several pesticide application scenarios and to randomly select PRZM parameter values. One hundred different scenarios were simulated for each of over 2000 regions (Soil Landscapes of Canada [SLC] polygons) for the years 1981 and 2006. Overall, the results indicated that in those regions in which the risk increased from 1981 to 2006, the increase in risk was mainly attributable to the increased area treated by pesticides or an increase in the number of days with runoff. More specifically, this work identifies the areas at higher risk, where further analyses with finer-scale input data should be performed. The model is specific for Canadian data, but the framework could be adapted for other large countries.