Recent reports of a dramatic decline in the abundance of insects recommend serious consequences for global ecosystems and human society. However, most of the evidence comes from Europe, leaving uncertainty about trends in insect populations around the world. We use more than 5,300 time series of insects and other arthropods, collected for four to 36 years at tracking sites representing another 68 controlled and herbal areas, to look for evidence of decreases in the United States. Some taxa and sites have shown minimal abundance and diversity, while others have increased or remained unchanged, resulting in trends of net abundance and sometimes indistinguishable biodiversity of zero. This lack of overall increase or decrease was constant in all arthropod feeding equipment and was similar for highly disturbed sites compared to herbal sites. The obvious strength of American arthropod populations is reassuring. However, this result does not decrease the need for continuous monitoring and may mask more sophisticated adjustments to the composition of species that, however, endanger ecosystem facilities provided by insects.
The data supporting the effects of this study (organized abundances of arthropods and estimated temporal trends) are found in the Dryad Data Repository (https://doi.org/10.5061/dryad.cc2fqz645).
The R code used to organize and analyze knowledge must be held in the Dryad knowledge repository (https://doi.org/10.5061/dryad.cc2fqz645).
A. R. Ives (University of Wisconsin-Madison) provided valuable recommendations on our analysis, and Dr. R. Strand (University of Georgia) and W. F. Fagan (University of Maryland) made suggestions for the article. We recognize the investment for USDA-NIFA-OREI 2015-51300-24155 and USDA-NIFA-SCRI 2015-51181-24292 at W.E.S.
Department of Entomology, University of Georgia, Athens, GA, USA
Michael S. Crossley, Amanda R. Meier and William E. Snyder
Department of Biology and Health Sciences, Hendrix College, Conway, AR, USA
Emily M. Baldwin, Lauren L. Berry, Leah C. Crenshaw, David H. Nichols, Krishna Patel, Sofia Varriano & Matthew D. Moran
Agricultural Research Service, U.S. Department of Agriculture, Urban, IL, USA.
Glen L. Hartman and Doris Lagos-Kutz
M.S.C., A.R.M., W.E.S. and M.D.M. conceived the concept of paper, and M.S.C. and A.R.M. analysis M.S.C., A.R.M., W.E.S., M.D.M., E.M.B., D.L.-K., G.L.H., L.L.B., L.C.C., D.H.N., K.P. And please. He helped collect and acquire knowledge M.S.C., W.E.S. and M.D.M. they mainly wrote the article, all the authors contributed to the final manuscript.
The authors claim to have competing interests.
Peer review information Peer peer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
(a) herbivores, (b) carnivores, (c) omnivores, (d) detrivores, parasites and (f) parasitoids. The panels on the right illustrate the average replacement in diversity measures and 95% confidence periods between LTER. Blue shading and police LTER sites reporting aquatic taxa.
The cases of the moustache are the medians (thick line), the percentiles 25 and 75 (edges of the box), the percentiles 95 (bigotes) and the outliers (circles).
Trends in the abundance of all taxa according to (a) criteria of filtering of moderate time series vs. (b) strict or moderate filtering criteria. (c) Box diagrams of abundance trends according to filtering criteria for quiet, moderate and strict time series. The comfort criteria required at least 4 years of counting, one of which would be another of 0 (n -5,328 of 6,501 trends remained). Moderate criteria required at least 8 years of counting, 4 of which would be 0 (n -2,266 trends remained). Strict criteria required at least 15 years of counting, 10 of which had to be others from 0, and that temporary self-correction had to be extended.
(a) Year of start of LTER site sampling. (b) LTER site-related human footprint index. The average HFI for U.S. sites It’s 7; LTER sites ranged from 1 to 38. (c) Average annual temperature at LTER sites. (d) Average cumulative annual precipitation at LTER sites.
Each variable’s contribution to the accuracy of the random forest classifier, explained as the percentage that accumulates in the average quadratic error (decrease in accuracy) when the variable is excluded from the resolution trees.
Abundance trends are averaged between LTTT where sampling years preded on 1990, from 1990 to 2000, distributed from 2000 to 2010 or after 2010. The effects were the same when trends were grouped according to the last sampling years (except that there were no definitive sampling years. 1990).
Points are the over-time replacement of a diversity metric in an LTER site. The regularity of the species has been calculated as the rate of regularity of pielou, and dominance constitutes the proportional frequency of the most abundant taxon. Light gray lines divide each chart into quadrants to help visualize sites where the replacement signal in diversity measures similar (top right, left rear) or opposite (top left, right back). Black dashes indicate maximum productivity with the compatibility line. The slopes are significant at the point of 5%, R2 – 0.36 for regularity in relation to wealth and R2 – 0.68 for regularity in relation to dominance.
The panel on the left illustrates the abundance trends separated by echoregion point I. The panel on the right represents the abundance trends separated by echoregion point II. The mustache boxes illustrate the quantums between the LTER sites. The mustache boxes illustrate trends between medium-sized insects (thick line), 25th and 75th percentiles (box edges), 95th percentiles (bigotes), and outliers (circles).
U.S. Human Footprint Index (Left Panel) and LTER (Right Panel) index values.
Points are the over-time replacement of a diversity metric in an LTER site. The dotted line indicates line 1: 1.
Additional tables 1/3.
LTER attributes included in analysis.
Reprints and permits
Received March 17, 2020
Accepted: 26 June 2020
Published: August 10, 2020
DOI: https://doi.org/10.1038/s41559-020-1269-4