Recent collapse of crop belts and declining diversity of US agriculture since 1840

Abstract

Over the last century, US agriculture greatly intensified and became industrialized, increasing in inputs and yields while decreasing in total cropland area. In the industrial sector, spatial agglomeration effects are typical, but such changes in the patterns of crop types and diversity would have major implications for the resilience of food systems to global change. Here, we investigate the extent to which agricultural industrialization in the United States was accompanied by agglomeration of crop types, not just overall cropland area, as well as declines in crop diversity. Based on countylevel analyses of individual crop land cover area in the conterminous United States from 1840 to 2017, we found a strong and abrupt spatial concentration of most crop types in very recent years. For 13 of the 18 major crops, the widespread belts that characterized early 20th century US agriculture have collapsed, with spatial concentration increasing 15-fold after 2002. The number of counties producing each crop declined from 1940 to 2017 by up to 97%, and their total area declined by up to 98%, despite increasing total production. Concomitantly, the diversity of crop types within counties plummeted: in 1940, 88% of counties grew >10 crops, but only 2% did so in 2017, and combinations of crop types that once characterized entire agricultural regions are lost. Importantly, declining crop diversity with increasing cropland area is a recent phenomenon, suggesting that corresponding environmental effects in agriculturally dominated counties have fundamentally changed. For example, the spatial concentration of agriculture has important consequences for the spread of crop pests, agrochemical use, and climate change. Ultimately, the recent collapse of most agricultural belts and the loss of crop diversity suggest greater vulnerability of US food systems to environmental and economic change, but the spatial concentration of agriculture may also offer environmental benefits in areas that are no longer farmed.

Historical decrease in agricultural landscape diversity is associated with shifts in bumble bee species occurrence

“…Bee and agricultural census data are therefore compared at the county‐level. Because 95% of records were from 1890 and beyond, we are confident that county assignments are accurate, as changes in county geographical extent in this region were largely complete by 1890 (Crossley et al 2020). As both data sources (GBIF and BBW) are largely comprised of incidental occurrence records, have expert‐verified species identifications, and are geo‐referenced, we determined that combining them was appropriate.…”

“…To assess measures of agricultural intensification, we used county‐level ( n = 535) agricultural census data collected every 10 years that were projected and geographically corrected by Crossley et al (2020). For each county by census year, we used crop richness (range 0–18) and the proportion of county area in cropland (range 0–1) as our metrics of agricultural intensity.…”

“…To test the hypothesis that long‐term bumble bee occurrence trends are associated with patterns of agricultural intensification, we utilised an extensive data set of historical bumble bee museum records and modern citizen‐science surveys for the Midwest USA and combined this with a newly available digital data set of agronomic metrics compiled from the United States Census of Agriculture over the period 1840–2017 (Crossley et al 2020). We predicted that increasing agricultural intensification, as measured by the amount of cropland in a county, would be negatively associated with bumble bee occurrence (Duchenne et al, 2020; Goulson et al, 2008, 2015; Samuelson et al, 2018; Williams & Osborne, 2009), given that an increase in cropland extent is typically associated with a decrease in a natural or semi‐natural habitat.…”

Recent climate change is creating hotspots of butterfly increase and decline across North America

“…Effects of insecticides and herbicides, which are detrimental to insects directly by toxicity as well as indirectly via effects on host plants (Pleasants & Oberhauser, 2013), will also tend to be greater with increasing cropland and built land cover (Falcone et al, 2018; Meehan & Gratton, 2016). Lastly, given the rapid and dramatic change in the amount and location of cropland in the United States since the 1800s (Crossley et al, 2021; Waisanen & Bliss, 2002), the historical trend in proportion cropland was included to examine whether legacies of historical cropland expansion show any effects on butterfly abundance trends that are distinct from the amount of contemporary, intensively managed cropland (Foster et al, 2003). Contemporary land cover data were obtained from the North American Land Change Monitoring System (NALCMS; MRLC Consortium, 2020).…”

Farmer perspectives on benefits of and barriers to extended crop rotations in Iowa, USA

“…forages (Sulc & Tracy, 2007). However, socio-economic and technological changes in agriculture over the last 75 years have led to a reduction in the area planted to crops such as oat and hay and an increase in that used for soybean (Aguilar et al, 2015;Hijmans et al, 2016;Crossley et al, 2021).…”