Measuring Agricultural Land–Use Intensity — A Global Analysis Using a Model–Assisted Approach

As future demand for agricultural produce rises, we must develop methods to increase production and satisfy our food needs.  The conversion of land into farms is a solution, but comes with the price of habitat destruction, altered CO2 output and input, and land erosion.  A more sensible solution would be to improve production on farmlands we currently have.  Dietrich et al. (2012) investigates and compares the agricultural land–use intensities for 10 world regions and 12 crops.  The authors use the Lund–Potsdam–Jena dynamic global vegetation model with managed Land (LPJmL) to project reference yields that are compared to current observed yields.  The results show parts of Russia, Asia, and Africa having low agricultural land–use intensities, whereas Eastern U.S., Western Europe, and parts of China have high agricultural land–use intensities.  Measuring for agricultural land–use intensity differs from the more commonly used method of calculating gap yields.  In light of the author’s results and analysis, this paper shows the value in using the t-factor for measuring land-use intensity which can be more accurate than other measuring methods.—Anthony Li
Dietrich, J. P., Schmitz, C., Mueller, C., Fader, M., Lotze-Campen, H., Popp, A., Measuring agricultural land-use intensity — A global analysis using a model-assisted approach.  Ecological Modeling May 10th, 2012

In calculating land–use intensities, the authors introduce a new measure called the t–factor.  The t–factor is the ratio between actual, observable yield to a reference, calculated yield under well–defined management and technology conditions.  The t–factor is independent of the physical environment and is proportional to the agricultural land–use intensity.  The reference yield used to calculate the t–factor can be either deduced from models or statistical analysis.  In this study, the authors use the LPJmL model to project the reference yield.  The LPJmL model simulates the growth, production, and phenology of plant and functional crop types, and reports it as maximum leaf area index, scaling factor from simulated leaf-level photosynthesis to field scale, and harvesting index.  Actual yields were compiled from the Food and Agricultural Organization of the United Nations Statistics (FAOSTAT), but were first applied to the same LPJmL model in order to account for discrepancies due to bias or systematic errors.  The authors calculated t–factors for 10 global regions and 12 different crop types for 2000, and omitted any data for crops that produced less than 0.1% of the global crop production.
The authors compared variations in t–factor, homogeneity in t–factor and the t–factor itself between the regions of the world.  Europe had the highest total t–factor, as well as the highest crop–specific t–factors for wheat, millet, field peas, and rapeseed.  The Middle East and North Africa ranked lowest in total t–factors, but was closely followed by Africa despite the fact that Africa had more crops with the lowest t–factor.  Africa and Europe had the least variation in their t–factors, which contributed to their lowest and highest t–factors, respectively.  In contrast, the Pacific Organization for Economic Cooperation and Development and the Middle East and North Africa showed the strongest variations in t–factor between crops.  In terms of homogeneity in t–factors across each country, Ireland, the United Kingdom, France, Germany, and Sweden show homogeneous values at a high t–factor while Madagascar and Mozambique had homogeneous values at a low t–factor. 
The t–factor measure used in this study differs from other measures of land–use intensity by eliminating the environmental component from observed actual yields via a reference yield.  Other typical, input–oriented measures determine land–use intensity by measuring individual drivers of land–use intensity such as fertilizer use, labor use, and machinery.  Since the t–factor is independent of natural conditions, it can be used as a measure of yield differences due to human activity.  If sample region A has twice the t–factor than sample region B, then it will have twice the yield due to human activities.  The same could be said for the reverse; assuming they have the same t–factors, if the physical conditions in region A are half as good as in region B, then the yields in both regions will be equal.  While the t–factor estimates total land–use intensity which cannot be attributed to individual factors, input–oriented measures record the relevance of certain individual inputs to the attributing factors of overall land–use intensity.  The authors make the point that low agricultural land–use intensities do not necessarily mean higher yield increases in the future.  Factors such as political instability or weak governing bodies may inhibit a nation’s opportunity to improve yields.  While these results may not accurately represent yield increases in specific countries, it is still representative of global agricultural conditions as the results are similar with FAO/OECD yield growth projections.  Without taking into account political stability or effectiveness of governing body, regions such as Africa, South and Eastern Europe, Russia, South Asia, and Latin America show long-term chances for yield increases.  This paper introduces a new method, the t–factor, for measuring agricultural land–use intensity that does not take into account environmental conditions.  Because of this, the t–factor is proportional to agriculture land–use intensity and is a good measurement if we want to calculate land–use intensity solely as a result of human activity.

Recent Land Use Change in the Western Corn Belt Threatens Grasslands and Wetlands

The recent boom in the biofuel industry, in part due to incentives that promote the conversion of grassland to corn and soybean cropping, is reshaping the landscape of the US Corn Belt. Wright et al. (2013) sought to study the extent to which this land use conversion is occurring, and what its implications may mean for the environment.  The researchers used the National Agricultural Services (NASS) Cropland Data Layer (CDL) to examine the rate at which grasslands have been converted into corn/soy cultivation over five states of the Western Corn Belt: North and South Dakota, Nebraska, Minnesota, and Iowa.  The authors considered the agronomic and environmental attributes of lands on which grassland conversion was occurring, as well as the effects on nearby waterfowl nesting sites, and included these in the results as well.  The results of this study show that the rate at which land was being converted has not been seen in the US since the advent of the mechanization of US agriculture in the 1920s.  The implications of this rate are bleak as it threatens waterfowl populations, soil quality, and water resources.  The authors recommend we shift to biofuels produced from perennial feedstocks, as these fuels have desirable traits with respect to net energy and greenhouse gas balances and wildlife conservation. —Anthony Li
Wright, C. K., Wimberly, M. C., 2013. Recent land use change in the Western Corn Belt threatens grasslands and wetlands.  Proceedings of the National Academy of Sciences of the United States of America published ahead of print February 19, 2013

The authors acquired land cover data from 2006 to 2011 of the Western Corn Belt from the NASS CDL.  They selected this year range because the extent of the data recording goes back to 2006. The NASS CDL uses land cover data acquired from satellite imagery and maps agricultural land cover at a very high crop-type specificity.  Using the 2006 NASS CDL data and comparing it with the 2011 NASS CDL on a per-pixel basis allowed the researchers to observe a general grass-dominated land cover be converted into a general corn/soy cultivation land.  In order to see if the land use data derived from the NASS CDL was representative of long-term land cover change region-wide, they performed a trend analysis of grassland conversion in North Dakota and Iowa.  The analysis showed that the data were representative.  The researchers also took note of the agronomic and environmental attributes of the lands in which NASS CDL recorded data on.  Lastly, the authors examined the relationship between grassland conversion and lands protected under the Conservation Reserve Program (CRP).  The CRP “pays farmers to establish and maintain grassland cover on retired cropland in exchanged for a fixed rental payment over a fixed period,” but in recent years with the rise of corn and soybean prices as well as a projected consistently high commodity prices, more farmers have not been renewing their CRP contracts.  By examining this relationship, the authors were able to see which recently converted areas were formerly protected by the CRP, showing some insight in the farmer’s reasons for changing crop.
The results showed that across the Western Corn Belt, there was a net decline in grass-dominated land cover totaling near 530,000 ha, more than 1.3 million acres, from 2006 to 2011.  This change in land cover was concentrated in South Dakota and Iowa.  The rates at which grassland is being converted to corn/soy is comparable to the deforestation rates in Brazil, Malaysia, and Indonesia.  The authors make the comparison that the current rates of grassland conversion have not been seen in the Corn Belt since the advent of agriculture’s mechanization in the 1920’s.  Grassland conversion is also occurring dangerously close to the Prairie Pothole Region, a wetland region that acts as a climate-change refugia for North American waterfowl.  The current rate of grassland conversion threatens one of the few breeding grounds of waterfowl.  The authors found that grassland conversion was concentrated on relatively high quality lands in Minnesota and the Dakotas, suggesting that the local landowners are seeking higher rates of return by swapping to corn and soybean cultivation.  This trend has become increasingly consistent due to the emerging market of corn/soy production and its rate of return.  In Iowa, they found grassland conversion was occurring on less suitable land, reflecting the lack of high quality land for soybean/corn cultivation.  Similar to Iowa, Nebraska was also shown to have used unsuitable land for crop production, suggesting that both these states will have to acquire more resource-intensive irrigation practices to sustain the soy/corn crops.  The authors also predicted that fewer landowners will be renewing their CRP contracts as the higher rates of return for soybean/corn cultivation is more economically viable.
While this paper shows the rate at which the biofuel industry has grown, it also shows the daunting implications for such a growth. Grassland conversion into corn/soy production is characterized by high erosion risk and vulnerability to drought.  This grassland conversion also threatens waterfowl populations, as the soy/corn fields encroach upon diminishing waterfowl breeding sites.  The grassland conversion also effects the soil’s carbon sequestration ability.  The authors predict that with the reductions in soil sequestration caused by grassland conversion, “more than three decades of biofuel substitution” will be required to counteract this.  In the face of all this the researchers suggest an alternative, saying that biofuels derived from perennial feedstocks are more efficient with respect to net energy and greenhouse gas balances as well as wildlife conservation.

Closing the Gap: Global Potential for Increasing Biofuel Production Through Agricultural Intensification

In the past couple of decades, the global agricultural industry has seen a massive boom, in part due to a combination of fertilizers, pesticides, herbicides, smart management techniques, mechanization, irrigation, and optimized seed varieties and genetic engineering.  This jump in agriculture not only provides the opportunity to feed our growing population, but to also create ethanol and biodiesel to meet our energy demands.  Johnston et al. (2011) looked at the magnitude and spatial variation of new agricultural production potential from closing of ‘yield gaps’ for 20 major ethanol and biodiesel crops.  By using data sets of annual crop yields to determine the amount of additional biofuel produced from obtaining yield gaps up to the global median yield, the researchers deduced that approximately 112.5 billion liters of ethanol and 8.5 billion liters of biodiesel could be made.  While this shows an optimistic future for energy security, it also has a profound effect on policymakers and how individuals will determine goals of reaching a level of biofuel use.  —Anthony Li
Johnston M., Licker R., Foley J., Holloway T., Mueller N. D., Barford C., Kucharik C. 2011. Closing the gap: global potential for increasing biofuel production through agricultural intensification. Environmental Research Letters 6, 034028

The authors of this paper investigated 20 common biofuel and biodiesel crops, some notable ones include maize, rice, sugarcane and wheat for biofuel, or ethanol, and soybean, rapeseed, and oil palm for biodiesel crops.  The researchers obtained the M3 data set of global farming yields for these 20 crops and organized the data based on region.  With information on the average global yields of crops, the authors were able to calculate the yield gaps, which they defined for this study as the “difference between current agricultural yields and future potential based on climatic and biophysical characteristics of the growing region.”  They calculated the potential yields of biodiesel and ethanol if yield gaps of these crops were closed to multiple degrees, such as the global median or the 90thpercentile gap of what is completely attainable.  In order to observe the effects of unequal distributions of irrigation infrastructure and sustainable water resources on crop yields, the researchers re-ran their analysis with irrigated areas excluded.  In order to get a rough idea of what was needed to increase crop yield, the authors calculated the growing degree days for each crop, which is a measure of heat to predict plant development rates.
The researchers found that increasing yield gaps to the median global yield would result in 112.5 billion additional liters of ethanol and 8.5 billion liters of ethanol, while obtaining the 90th percentile gap would result in 450 billion liters of additional ethanol and 33 billion liters of biodiesel.  While the new tonnage varied considerably between biodiesel and ethanol, the overall percentage increase between the two were roughly equal, ranging from 10%–17%.  The majority of ethanol potential identified was attributable to maize, wheat, and rice crops, while the majority of biodiesel potential was attributable to soybean, rapeseed, and oil palm.  Biodiesel fuel production was generally more evenly distributed amongst its constituent crops, whereas ethanol fuel production was incredibly uneven between the crops.
The implications of this study in energy security are obvious, but they also provide a benefit to policymakers or anyone setting goals for biofuel use.  The research performed here shows policymakers how much additional biofuel we can expect from closing various yield gaps to different degrees, allowing them to make more accurate goals.  For example, The Renewable Fuel Standard Program Final Rule of the 2007 Energy Independence and Security Act made a goal for the US to blend 36 billion gallons of biomass-based fuels by 2022.  As ambitious as this goal was, this study showed that even if all the countries were to increase their biofuel crop yields to only the median level, there would still not be enough fuel to meet this goal.  This study is also useful in that it shows the biofuel and biodiesel distribution based on specific crop.  For ethanol, a very notable crop for fuel production was sugarcane.  While this may not mean that sugarcane produces the most ethanol of any other crop per mass, if we can identify whichever crop produces more fuel than others, we can focus our biofuel industry to take advantage of these specific crops.
In the face of our energy and food crisis, our nations should begin looking towards agriculture for potential solutions.  Johnston et al.’s study shows how much additional biofuel can be produced by closing various yield gap levels per crop.  This information will prove useful to governments seeking to implement goals of reaching certain levels of biofuel use and individuals such as farmers who want to capitalize on the most biofuel yielding crop.