Food Security Through Earth Observation
In 2008 and 2010, agricultural droughts in Russia, Ukraine, China and Australia, combined with the increased demand for biofuel and reactive trade policies, resulted in unprecedented grain prices and market instability. Between 2009 and 2010, the global price of wheat and corn increased by 80 percent. Amid deepening concern over food supply and demand, the 2009 U.N. World Food Summit made a Declaration on Food Security, calling for increased international cooperation to “improve the quality of national agricultural statistics and early warning and forecasting systems for food insecurity and vulnerability.”
In June, agricultural ministers from the world’s leading economies agreed to the G-20 “Action Plan on Food Price Volatility and Agriculture,” which launched two international initiatives:
- The Agricultural Market Information System, to gather timely information on food stocks and trade for producer and exporter countries, led by the U.N. Food and Agriculture Organization.
- The Global Agricultural Geo-Monitoring Initiative, to improve global monitoring of agricultural production and food security using Earth observation, led by the Group on Earth Observations (GEO). This GEO-GLAM initiative will build on the current GEO Agricultural Monitoring Task, providing enhancements and transparency to global production monitoring and food security systems, strengthening existing national capacities to use Earth observation data and building new capabilities where there are gaps. If it is to succeed, this initiative will need a major injection of resources for implementation and a serious commitment from national governments and the international Earth observation organizations (e.g. the Committee on Earth Observation Satellites and the World Meteorological Organization).
Crises often have the positive result of government agencies and international organizations working together on a coordinated response. In 1972, the year that the first Landsat was launched, a severe agricultural drought in Russia forced the Soviet Union into foreign grain purchases, resulting in the “Great Grain Robbery.” This crisis, the lack of spatially explicit information on Soviet agricultural production, the resulting instability in the grain markets and the availability of the new Earth observation satellite, led NASA, the U.S. Department of Agriculture (USDA) and the National Oceanic and Atmospheric Administration to develop the Large Area Crop Inventory Experiment in 1974, aimed at developing in-season crop production forecasts. It was followed by the Agriculture and Resources Inventory Surveys Through Aerospace Remote Sensing program in 1978. These programs provided the methodological underpinning for the satellite component of global agricultural monitoring undertaken by the USDA Foreign Agricultural Service and the National Agricultural Statistical Service — mapping of cropland area and crop type, estimation of area planted, assessment of within-season crop condition, and estimation of production and yield.
Subsequent incremental improvements in the spectral, spatial and temporal resolutions of Earth observation (EO) systems — including the Advanced Very High Resolution Radiometer, Moderate Resolution Imaging Spectroradiometer, Thematic Mapper, Jason (reservoir and lake levels), Tropical Rainfall Measuring Mission and a number of geostationary systems (rainfall estimation), and a continued partnership between NASA and USDA — have led to improvements in U.S. operational agricultural monitoring. Recently, the Land Atmosphere Near-real-time Capability for EOS providing global data within two hours of acquisition, has improved the timeliness of Moderate Resolution Imaging Spectroradiometer data delivery, a critical aspect of Earth observation for agricultural monitoring. Recent research in Canada and Europe using synthetic aperture radar data is also generating promising results for monitoring crop development.
The failure of the Landsat 7 Scan Line Corrector in 2003 emphasized the fragility of U.S. Earth observation systems to meet the needs of operational users and obliged the USDA to purchase international moderate-resolution data. The resulting use of data from the Advanced Wide-Field Sensor on India’s Resourcesat highlighted the benefits of having multiple international Earth observation assets on orbit and higher revisit times at moderate resolution.
There are currently four systems for global agricultural monitoring, all using Earth observation data:
- The USDA Foreign Agricultural Service’s Crop Explorer.
- The European Commission’s Monitoring of Agriculture with Remote Sensing.
- CropWatch, developed by the Chinese Academy of Sciences’ Institute for Remote Sensing Applications.
- The U.N. Food and Agriculture Organization’s Global Information and Early Warning System.
Most countries have their own national agricultural monitoring systems, which traditionally utilize rainfall data, sample field measurements, agricultural statistics and, in some cases, agro-meteorological modeling. The use of Earth observation data is in various stages of development and integration in these systems. For food insecure countries lacking effective agricultural (crops and rangeland) monitoring, international programs such as the U.S. Agency for International Development’s Famine Early Warning Systems Network, the World Food Program and the Global Information and Early Warning System provide operational monitoring of food supply utilizing Earth observation, with information on food availability, market prices and livelihoods.
The GEO was formed in 2003 with a vision of “a world where decisions and actions are informed by coordinated, comprehensive and sustained Earth observations.” In 2007 the GEO established the Global Agricultural Monitoring Task and the associated Community of Practice, focused on monitoring agricultural production, food security and agricultural land use change. From the ensuing Community of Practice workshops, it became apparent that: Earth observation needs are more or less common across the different agricultural monitoring systems; there are no standard methods; the variety of Earth observation data policies, formats and access systems present an obstacle to uptake; there is no coordinated acquisition of satellite data for agricultural areas by the space agencies or data providers; and the rain gauge networks critical for crop modeling are generally in decline and particularly in developing countries. There also was recognition that capacity building is urgently needed to transfer existing Earth observation capabilities to operational agencies producing national agricultural statistics, especially in Africa.
Four community activities were initiated under the GEO Task: articulation of observation requirements for agricultural monitoring; a common database allowing comparison of national production estimates from different systems; a series of regional workshops to share operational methods and new research findings and develop best practices; and an international Joint Experiment for Crop Assessment and Monitoring, focused on comparing data, methods and models for a series of established, well-instrumented agricultural experimental sites around the world. The Committee on Earth Observation Satellites is coordinating the provision of data from different Earth observation systems for these experimental sites. Although progress has been made under GEO, these efforts are constrained by a reliance on voluntary contributions and best efforts from the Community of Practice.
Food crises such as the food riots in Algeria and Tunisia this year and the tragic famine in the Horn of Africa point to the critical need to deliver timely food security information to decision-makers. The effect of increasing fuel prices on the price of food, through increased transportation and fertilizer costs, along with projections of increased frequency of extreme events and climate variability, would indicate that the issue of food security will be with us for some time to come. In addition, the World Food Summit stated that agricultural output will have to increase by 70 percent to feed the projected world population of 9 billion by 2050.
Recent and planned missions, including the Visible Infrared Imager Radiometer Suite, Landsat Data Continuity Mission and Soil Moisture Active Passive (U.S.), Spot-Vegetation and RapidEye (Europe), Medium Resolution Imaging Spectrometer and the Sentinels (European Space Agency), Resourcesat (India), and Huan Jing, Feng Yun and China-Brazil Earth Resources Satellite (China, Brazil), along with private sector, fine spatial resolution systems, offer a clear opportunity through international cooperation to put in place an effective global agricultural monitoring system of systems. Experience from the meteorology domain has proved this is feasible through long-term coordinated commitment of national, regional and international actors and close interaction with downstream users.
The Action Plan on Food Price Volatility and Agriculture will be submitted to G-20 leaders in November. We will see then whether there is the political will to support the new international initiative to improve agricultural monitoring and commitment from government agencies and international organizations to work together on implementing a coordinated response.
Chris Justice is a professor and chairman of the geography department at the University of Maryland, College Park, and co-chairman of the GEO Global Agricultural Monitoring Task. Wu Bingfang is a professor and assistant director at the Institute of Remote Sensing Applications, Chinese Academy of Sciences, Beijing, and co-chairman of the GEO Global Agricultural Monitoring Task. Pascal Kosuth is a researcher at Cemagref-CIRAD-ENGREF, Montpellier, France, and co-lead of the GEO-GLAM initiative.