New study assesses impact of agricultural research investments on biodiversity, land use

"At the global level, we see a reduction in cropland use from these technology improvements leading to gains in terrestrial carbon stock and avoided loss of threatened plant and animal species," reported the team led by Purdue's Uris Baldos, research associate professor of agricultural economics.

The study is the first to undertake a fine-scale analysis back to the early 1960s. The analysis incorporated global data from approximately 100,000 grid cells. Each cell covers an area measuring 27.2 square kilometers (10.5 square miles) at the equator. Grid cells farther north and south of the equator become smaller because of the Earth's curvature.

"You need that spatial resolution to get at the biodiversity question, because biodiversity is not evenly spread around," said study co-author Thomas Hertel, Distinguished Professor of Agriculture. In another first, the study revealed how agricultural land-use changes have affected biodiversity. The analysis found that, globally, reduced agricultural land use resulting from improved crop varieties saved 1,043 animal and plant species.

Saved plant species numbered 818, along with 225 amphibian, bird, mammal and reptile species. "We find that roughly 80% of the avoided losses in plant species are located within 31 out of 34 biodiversity hot spots which are mapped in our model," Baldos and his co-authors reported.

Agriculture covers about 37% of the world's land area and generates one-fourth of greenhouse gas emissions that humans produce. The study found that improved crop varieties reduced the amount of cropland area from 1961 to 2015. Global croplands decreased by more than 39 million acres, while crop production increased by 226 million metric tons. Crop prices, meanwhile, dropped by nearly 2% as a result of the improved crop varieties.

The study also quantified the impact of new crop varieties developed by the Consultative Group on International Agricultural Research (CGIAR), a global innovation network of 15 centers that marked its 50th anniversary in 2021.

"Globally, CGIAR technologies contributed roughly 47% of the total production gains from adoption of improved crop varieties in developing countries" from 1961 to 2015, Baldos and his co-authors reported. These CGIAR technologies also significantly reduced cropland use, greenhouse gas emissions and biodiversity loss.

The researchers generated their results with Purdue's global model of agriculture, land use and the environment, called the Simplified International Model of agricultural Prices, Land use, and the Environment -- Gridded, or SIMPLE-G. The model incorporated a novel decades-long dataset of variety adoption and farm-level crop yields provided by co-author Keith Fuglie, an economist at the U.S. Department of Agriculture Economic Research Service.

"For this version of SIMPLE-G, the key drivers are population growth and productivity growth," Baldos said. Included in the model is crop production at the grid-cell level with input factors such as fertilizer, labor and water.

The researchers incorporated satellite data on terrestrial carbon and cropland availability into their SIMPLE-G model. The open-access book "SIMPLE-G: Gridded Economic Approach to Sustainability Analysis of the Earth's Land and Water Resources" offers various versions of the model. SIMPLE-G is among an array of models developed by Purdue's Global Trade Analysis Project.

Data constraints led previous studies to focus on national- and continental-scale regions in assessing the historical land-use effects of agricultural advancements. Those studies found that improving the application of agricultural technology made farming more profitable in areas such as sub-Saharan Africa and Latin America while fostering deforestation and other environmentally harmful impacts in certain locations.

But there's also a global aspect, as Purdue agricultural economists have repeatedly studied over the years. "Improved technology generally saves resources at the global level because you're feeding more or less the same amount of people and doing it more efficiently," Hertel noted.

Previous research, including a 2014 PNAS study by Hertel and Baldos, examined the impact of improved agricultural technology on land use and greenhouse gas emissions. It was the first such study that presented data from running an agricultural economics model backward in time over multiple decades, as well as forward, the way climate scientists have done for years.

The 2014 study, like the new one, ran a scenario backward to 1961 and then forward, with and without the new crop varieties. "If we take away the technology, what would things look like then?" Hertel said. In the new work, he added, "We're repeating some of those innovations from that earlier study, but now with the fine-scale analysis that lets us get at the biodiversity and the terrestrial carbon."

The earlier study discussed the prospective impact of a green revolution in Africa. "These improved varieties have had a big impact in Africa. That's a good news story," Hertel said.

Many of the early Green Revolution gains pertained to Asia and some of Latin America. But with recent funding from various private and governmental organizations, research institutes in sub-Saharan Africa have begun developing new varieties of regionally important crops such as tubers and legumes.

"In the past, those countries didn't have national research institutes that could adapt the improved varieties to local conditions," Baldos said. The private sector, he noted, also developed new varieties over more recent decades, such as genetically modified soybeans in South America.

The research team concluded that ongoing research investment "can help sustain agricultural productivity growth across the world, strengthening global food security and mitigate agriculture's environmental footprint in the coming decades."

This research was supported by the U.S. Agency for International Development's Bureau for Resilience and Food Security and the USDA Economic Research Service.