International Conference of Agricultural Economists 2015: A recap via datasets

Leah Bevis, Tanvi Rao, and Andrew Simons are all PhD candidates at Cornell’s Dyson School. Andrew and Leah are currently on the job market.

We recently attended the International Conference of Agricultural Economists (ICAE) 2015 in Milan. This conference was particularly exciting, as it only happens every three years and draws both junior and senior agricultural economists from across the globe. It also draws economists from all types of organizations: the World Bank, USAID, USDA, IFPRI and other CGIAR institutes, the Joint Research Center of the European Commission, the FAO, and universities across the world. This diverse set of economists clearly commanded a diverse set of interests and skills. Moreover, they worked with valuable and often publically available datasets.

Instead of highlighting papers that caught our interest, as has been done on this blog in the past (AERE 2015, MWIEDC 2015, etc.) we offer a new type of conference “recap,” bringing attention to  some of the diverse and valuable datasets we became aware of at the conference that may be of interest to our readers. Here are a few that we noticed, which conveniently fall into four major categories:

• Nationally representative household panel data sets.
  • The LSMS-ISA datasets. This is a bandwagon well-jumped-on, but so valuable that we feel compelled to mention it. Dozens of papers presented work with one or more LSMS-ISA datasets, a group of nationally representative, publically available panel datasets in 8 African countries that include detailed information on agricultural practices and production.
    • Lesser known: the LSMS team is conducting a series of methodological experiments, where specific variables (e.g. plot size, soil quality, production quantity) are measured in a variety of ways, and the accuracy of these techniques is compared. See an example here. These additional experimental data will be made publically available at some point in the future.
  • The Mexican Family Life Survey. This publically available, nationally representative panel dataset tracks approximately 40 thousand Mexican individuals across 3 rounds: 2002, 2005-6, and 2009-12. It includes individuals who migrated within Mexico and those who immigrated to the United States, re-interviews with about 90 percent of the original families in the second two waves, and also gathers information about the family and split-off households of migrants/immigrants. A gargantuan tracking task, this dataset allows new, micro-level analysis on the dynamics around migration and well-being.

• Numerous temperature and precipitation datasets that can be overlaid onto other existing data.
  • Re-analysis datasets from the European Centre for Medium-Range Weather Forecasts (ECMWF). While the ECMWF primarily forecasts weather, they additionally provide re-analysis (modeled) datasets of recent weather patterns, such as the ERA-interim dataset, with atmospheric variables at varying intervals (e.g., hourly, daily, monthly) at a 79-km grid between 1979 and today.
  • Africa Rainfall Climatology version 2 (ARC2) from the National Oceanic and Atmospheric Administration (NOAA). This project provides historical re-analysis rainfall data for Africa between 1983 and 2012, gridded at ~0.1° spatial resolution (~10km), and provided in 10-day intervals.
  • Willmott and Matsuura’s Gridded Monthly Time Series V 4.01. These datasets provide monthly, interpolated temperature averages and monthly precipitation totals for the entire world, from 1900 to 2014, in grids size 0.5° latitude by 0.5°
  • NASA’s Modern-Era Retrospective Analysis for Research and Applications (MERRA). This project provides re-analysis data on a range of weather and climate variables, at a range of time-scales, gridded at 2/3° longitude and 0.5° latitude.
  • CHIRPS data from the Climate Hazards Group (CHG). The quasi-global rainfall data span 50° S to 50° N, gridded at 0.05° resolution. Data includes totals by day, pentad (6 pentads = 1 month) or dekad (3 dekads = 1 month), from 1981 until near-present.

• Soil databases that are georeferenced so they can also be overlaid with other datasets.
  • Harmonized World Soil Database from FAO, IIASA, ISRIC, ISSCAS, and JRC. This massive database provides interpolated soil quality estimates for the entire world, including nutrient availability and a number of other variables, in grids spaced at 30 arc sections (approximately 1 km).
  • African SoilGrids from AfSIS/ISRIC. This source provides data on a number of interpolated soil quality indicators (soil pH, sand, soil organic matter, cation exchange capacity, etc.) in 250-meter grids, for the African continent.

• Crop production/suitability datasets, often used to indicate productivity potential in a given spatial region.
  • IFPRI’s Spatial Production Allocation Model (SPAM) database compiles spatially disaggregated (gridded at 10×10 km resolution) crop production data for several countries. While this data can be overlaid with other gridded information for all sorts of detailed analysis, one session at the conference discussed overlaying SPAM data with Demographic and Health Surveys (DHS) data, which also provides geographic coordinates, for studying agriculture-nutrition linkages.
  • FAO’s Global Agro-Ecological Zones (GAEZ) data, which provides spatially referenced time series data as well as time-averaged data on climactic variables, crop suitability/productivity variables, and yields and production gaps.

Note: This post was updated on September 2, 2015.