Arthropod decline in grasslands and forests is associated with landscape-level drivers

Article metrics

Abstract

Recent reports of local extinctions of arthropod species1, and of massive declines in arthropod biomass2, point to land-use intensification as a major driver of decreasing biodiversity. However, to our knowledge, there are no multisite time series of arthropod occurrences across gradients of land-use intensity with which to confirm causal relationships. Moreover, it remains unclear which land-use types and arthropod groups are affected, and whether the observed declines in biomass and diversity are linked to one another. Here we analyse data from more than 1 million individual arthropods (about 2,700 species), from standardized inventories taken between 2008 and 2017 at 150 grassland and 140 forest sites in 3 regions of Germany. Overall gamma diversity in grasslands and forests decreased over time, indicating loss of species across sites and regions. In annually sampled grasslands, biomass, abundance and number of species declined by 67%, 78% and 34%, respectively. The decline was consistent across trophic levels and mainly affected rare species; its magnitude was independent of local land-use intensity. However, sites embedded in landscapes with a higher cover of agricultural land showed a stronger temporal decline. In 30 forest sites with annual inventories, biomass and species number—but not abundance—decreased by 41% and 36%, respectively. This was supported by analyses of all forest sites sampled in three-year intervals. The decline affected rare and abundant species, and trends differed across trophic levels. Our results show that there are widespread declines in arthropod biomass, abundance and the number of species across trophic levels. Arthropod declines in forests demonstrate that loss is not restricted to open habitats. Our results suggest that major drivers of arthropod decline act at larger spatial scales, and are (at least for grasslands) associated with agriculture at the landscape level. This implies that policies need to address the landscape scale to mitigate the negative effects of land-use practices.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Fig. 1: Temporal trends in arthropod communities.
Fig. 2: Changes in the dominance of species.
Fig. 3: Landscape effects on arthropod decline in grasslands.

Data availability

This work is based on data from several projects of the Biodiversity Exploratories programme (DFG Priority Program 1374). All data used for analyses are publicly available from the Biodiversity Exploratories Information System (https://doi.org/10.17616/R32P9Q) at https://www.bexis.uni-jena.de/PublicData/PublicDataSet.aspx?DatasetId=25786. Raw data are publicly available from the same repository (with identifiers 21969, 22007, 22008, 19686 and 20366), or will become publicly available after an embargo period of five years from the end of data assembly to give the owners and collectors of the data time to perform their analysis. Any other relevant data are available from the corresponding author upon reasonable request.

References

  1. 1.

    Habel, J. C. et al. Butterfly community shifts over two centuries. Conserv. Biol. 30, 754–762 (2016).

  2. 2.

    Hallmann, C. A. et al. More than 75 percent decline over 27 years in total flying insect biomass in protected areas. PLoS ONE 12, e0185809 (2017).

  3. 3.

    Ceballos, G. et al. Accelerated modern human-induced species losses: entering the sixth mass extinction. Sci. Adv. 1, e1400253 (2015).

  4. 4.

    Thomas, J. A. et al. Comparative losses of British butterflies, birds, and plants and the global extinction crisis. Science 303, 1879–1881 (2004).

  5. 5.

    Biesmeijer, J. C. et al. Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands. Science 313, 351–354 (2006).

  6. 6.

    Potts, S. G. et al. Global pollinator declines: trends, impacts and drivers. Trends Ecol. Evol. 25, 345–353 (2010).

  7. 7.

    Dirzo, R. et al. Defaunation in the Anthropocene. Science 345, 401–406 (2014).

  8. 8.

    Conrad, K. F., Warren, M. S., Fox, R., Parsons, M. S. & Woiwod, I. P. Rapid declines of common, widespread British moths provide evidence of an insect biodiversity crisis. Biol. Conserv. 132, 279–291 (2006).

  9. 9.

    Maes, D. & Van Dyck, H. Butterfly diversity loss in Flanders (north Belgium): Europe’s worst case scenario? Biol. Conserv. 99, 263–276 (2001).

  10. 10.

    Stefanescu, C., Torre, I., Jubany, J. & Páramo, F. Recent trends in butterfly populations from north-east Spain and Andorra in the light of habitat and climate change. J. Insect Conserv. 15, 83–93 (2011).

  11. 11.

    Schuch, S., Wesche, K. & Schaefer, M. Long-term decline in the abundance of leafhoppers and planthoppers (Auchenorrhyncha) in Central European protected dry grasslands. Biol. Conserv. 149, 75–83 (2012).

  12. 12.

    Schuch, S., Bock, J., Krause, B., Wesche, K. & Schaefer, M. Long-term population trends in three grassland insect groups: a comparative analysis of 1951 and 2009. J. Appl. Entomol. 136, 321–331 (2012).

  13. 13.

    Brooks, D. R. et al. Large carabid beetle declines in a United Kingdom monitoring network increases evidence for a widespread loss in insect biodiversity. J. Appl. Ecol. 49, 1009–1019 (2012).

  14. 14.

    Desender, K., Dekoninck, W., Dufrêne, M. & Maes, D. Changes in the distribution of carabid beetles in Belgium revisited: have we halted the diversity loss? Biol. Conserv. 143, 1549–1557 (2010).

  15. 15.

    Bowler, D. E., Heldbjerg, H., Fox, A. D., de Jong, M. & Böhning-Gaese, K. Long-term declines of European insectivorous bird populations and potential causes. Conserv. Biol. 33, 1120–1130 (2019).

  16. 16.

    Siemann, E. & Weisser, W. W. (eds) Insects and Ecosystem Function (Springer, 2004).

  17. 17.

    Shortall, R. C. et al. Long-term changes in the abundance of flying insects. Insect Conserv. Divers. 2, 251–260 (2009).

  18. 18.

    Geiger, F. et al. Persistent negative effects of pesticides on biodiversity and biological control potential on European farmland. Basic Appl. Ecol. 11, 97–105 (2010).

  19. 19.

    Ewald, J. A. et al. Influences of extreme weather, climate and pesticide use on invertebrates in cereal fields over 42 years. Glob. Chang. Biol. 21, 3931–3950 (2015).

  20. 20.

    Gossner, M. M. et al. Land-use intensification causes multitrophic homogenization of grassland communities. Nature 540, 266–269 (2016).

  21. 21.

    Haddad, N. M., Haarstad, J. & Tilman, D. The effects of long-term nitrogen loading on grassland insect communities. Oecologia 124, 73–84 (2000).

  22. 22.

    Penone, C. et al. Specialisation and diversity of multiple trophic groups are promoted by different forest features. Ecol. Lett. 22, 170–180 (2019).

  23. 23.

    Brittain, C. A., Vighi, M., Bommarco, R., Settele, J. & Potts, S. G. Impacts of a pesticide on pollinator species richness at different spatial scales. Basic Appl. Ecol. 11, 106–115 (2010).

  24. 24.

    de Jong, F. M. W., de Snoo, G. R. & van de Zande, J. C. Estimated nationwide effects of pesticide spray drift on terrestrial habitats in the Netherlands. J. Environ. Manage. 86, 721–730 (2008).

  25. 25.

    Thomas, J. A. et al. The quality and isolation of habitat patches both determine where butterflies persist in fragmented landscapes. Proc. R. Soc. Lond. B 268, 1791–1796 (2001).

  26. 26.

    Blüthgen, N. et al. A quantitative index of land-use intensity in grasslands: Integrating mowing, grazing and fertilization. Basic Appl. Ecol. 13, 207–220 (2012).

  27. 27.

    Kahl, T. & Bauhus, J. An index of forest management intensity based on assessment of harvested tree volume, tree species composition and dead wood origin. Nat. Conserv. 7, 15–27 (2014).

  28. 28.

    Tscharntke, T. et al. Landscape moderation of biodiversity patterns and processes – eight hypotheses. Biol. Rev. Camb. Philos. Soc. 87, 661–685 (2012).

  29. 29.

    Chao, A. et al. Rarefaction and extrapolation with Hill numbers: a framework for sampling and estimation in species diversity studies. Ecol. Monogr. 84, 45–67 (2014).

  30. 30.

    Schenker, N. & Gentleman, J. F. On judging the significance of differences by examining the overlap between confidence intervals. Am. Stat. 55, 182–186 (2001).

  31. 31.

    Fischer, M. et al. Implementing large-scale and long-term functional biodiversity research: the Biodiversity Exploratories. Basic Appl. Ecol. 11, 473–485 (2010).

  32. 32.

    R Core Team. R: A Language and Environment for Statistical Computing (2018).

  33. 33.

    Schall, P. et al. The impact of even-aged and uneven-aged forest management on regional biodiversity of multiple taxa in European beech forests. J. Appl. Ecol. 55, 267–278 (2018).

  34. 34.

    Gossner, M. M. et al. A summary of eight traits of Coleoptera, Hemiptera, Orthoptera and Araneae, occurring in grasslands in Germany. Sci. Data 2, 150013 (2015).

  35. 35.

    Birkhofer, K. et al. Land-use type and intensity differentially filter traits in above- and below-ground arthropod communities. J. Anim. Ecol. 86, 511–520 (2017).

  36. 36.

    Rogers, L. E., Hinds, W. T. & Buschbom, R. L. A General weight vs. length relationship for insects. Ann. Entomol. Soc. Am. 69, 387–389 (1976).

  37. 37.

    Ellenberg, H. et al. Zeigerwerte von Pflanzen in Mitteleuropa. Scr. Geobot. 18, 248 (1991).

  38. 38.

    Fahrig, L. Rethinking patch size and isolation effects: the habitat amount hypothesis. J. Biogeogr. 40, 1649–1663 (2013).

  39. 39.

    Elston, D. A., Moss, R., Boulinier, T., Arrowsmith, C. & Lambin, X. Analysis of aggregation, a worked example: numbers of ticks on red grouse chicks. Parasitology 122, 563–569 (2001).

  40. 40.

    Hilmers, T. et al. Biodiversity along temperate forest succession. J. Appl. Ecol. 55, 2756–2766 (2018).

Download references

Acknowledgements

We thank T. Lewinsohn, S. Meyer and V. Wolters for their comments and suggestions for the analyses; M. Lutz, J. Bartezko, P. Freynhagen, I. Gallenberger, M. Türke, M. Lange, T. Kahl, E. Pašalić, E. Sperr, K. Kremer and all student helpers for conducting arthropod sampling in the field and laboratory; R. Achtziger, E. Anton, T. Blick, B. Büche, M.-A. Fritze, R. Heckmann, A. Kästner, F. Köhler, G. Köhler, T. Kölkebeck, C. Morkel, F. Schmolke, T. Wagner and O. Wiche for arthropod species identification; C. Seilwinder and R. Honecker for GIS work; the managers of the three Exploratories (K. Wells, S. Renner, K. Reichel-Jung, S. Gockel, K. Wiesner, K. Lorenzen, A. Hemp and M. Gorke) for their work in maintaining the site and project infrastructure; C. Fischer and S. Pfeiffer for giving support through the central office; A, Ostrowski, M. Owonibi and J. Nieschulze for managing the central database; and D. Hessenmöller, I. Schöning, F. Buscot and the late E. Kalko for their role in setting up the Biodiversity Exploratories project. The work has been funded by the DFG Priority Program 1374 ‘Infrastructure-Biodiversity-Exploratories’. Field work permits were issued by the responsible state environmental offices of Baden-Württemberg, Thüringen and Brandenburg.

Author information

S.S., J.M. and N.K.S. conceived the idea for the manuscript; M.M.G., N.K.S., S.S., D.A., W.W.W., T.N., S.W., P.S., C.A., J.B., J.V., D.P. and M.F. collected and processed data; S.S., J.M., M.M.G. and W.W.W. defined the final analysis; S.S., N.K.S., C.P., P.S. and M.M.G. analysed the data; S.S. and W.W.W wrote the first manuscript draft and finalized the manuscript. All authors discussed the analyses and commented on the manuscript.

Correspondence to Sebastian Seibold.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Peer review information Nature thanks Simon Leather and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Extended data figures and tables

Extended Data Fig. 1 Effects of weather variables on arthropod numbers.

Effects of mean winter temperature (November to February) and precipitation during the growing period (March to October) on biomass, abundance and number of species in arthropod communities in 30 forests (orange) and 150 grasslands (blue) across three regions of Germany. Dots represent raw data aggregated per site and year (n = 1,406 (grassland) or 266 (forest) independent samples). Dotted lines indicate non-significant (P ≥ 0.05) and solid lines indicate significant effects of weather variables (P < 0.05), based on linear mixed models that included year, local and landscape land-use intensity as covariates. Shaded areas represent confidence intervals. The effects of winter temperature and precipitation differed between forests and grasslands. In grasslands, arthropod numbers increased with increasing winter temperature and with increasing precipitation in the growing period; the effect of precipitation was weaker than the effect of winter temperature, and the effects of both weather variables were weaker than the effect of the year (Supplementary Table 1-1). In forests, arthropod numbers decreased with increasing winter temperature and with increasing precipitation in the growing period; the effects of the two weather variables were similarly strong, but slightly weaker than the effect of the year (Supplementary Table 1-1).

Extended Data Fig. 2 Contribution of individual years to overall trends.

a, To assess the contribution of individual years to the overall trend, we repeated the linear mixed models for overall biomass, abundance and number of species, and excluded one year each time. The distribution of t and z values for the effect of the year from subset models (white), and from the full models including all years (black), are shown (11 models for grasslands and 10 models for forests). Grey bars denote effect of the year 2008 (the year with the strongest effect on overall trend estimates). b, In addition, we tested whether the observed effect of year differed from a random expectation by randomizing the order of years 100× for forests and grasslands before modelling. The distribution of t and z values for the effect of the year from models with randomly ordered years (white) and models with the years ordered correctly (black) are shown (101 models each for grasslands and forests). Vertical dashed lines indicate levels of significance with P < 0.05. The results in a show that both weaker and stronger temporal trends could be detected when single years were excluded from the analysis, compared to the full model including all years. Results in b show that models with the years ordered randomly produced effects of the year that were normally distributed around zero, and only the models with years ordered correctly generated strong temporal trends. For a more detailed discussion, see Supplementary Information section 3.

Extended Data Fig. 3 Declines in gamma diversity of frequent species.

Estimated gamma diversity (total number of species across all grassland or forest sites) over time. Symbols and error bars shown mean and 95% confidence intervals for gamma diversity, calculated as incidence-based, bias-corrected diversity estimates (Chao’s BSS29, with 200 bootstrapping runs; Methods) for q = 1 and 2 (for q = 0, see Fig. 1). With increasing order q, the more-frequent species are more strongly weighted (q = 0 equals species richness, q = 1 equals the exponential of Shannon entropy and q = 2 equals the inverse of Simpson diversity; that is, only dominant species affect the diversity measure). This approach accounts for slight differences in site numbers between years caused by limited accessibility or failure of traps. Non-overlapping confidence intervals indicate a significant difference between two sampling years30. Figure 1 shows that gamma diversity declines in both forests and grasslands for q = 0. We find that in forests gamma diversity declines when only the more-common species are considered (q = 1 and q = 2), whereas in grasslands there is no overall decline when only the common species are considered. For a more detailed interpretation, see Supplementary Information section 4.

Extended Data Fig. 4 Effect of tree mortality on arthropod trends.

a, The relative change in the number of arthropod species between the first two and the final two study years was similar for managed (n = 19) and unmanaged (n = 9) forest sites (z = 0.648, P = 0.517, derived from a linear mixed model with relative difference in species number as response, harvesting category as fixed and region as random effect). Dots show raw data per site. Boxes represent data within the 25th and 75th percentile, black lines show medians, and whiskers show 1.5× the interquartile range. b, When considering actual tree mortality between forest inventories in 2009 and 2016, declines in the number of arthropod species were weaker at sites with higher tree mortality (z = 2.536, P = 0.011, derived from a linear mixed model with relative difference in species number as response, tree mortality as fixed and region as random effect). Dots show raw data per site. The blue line visualizes the significant relationship between the change in the number of arthropod species and tree mortality based on the linear mixed model, and the shaded area represents confidence intervals. This suggests that changes in habitat conditions and heterogeneity linked to tree mortality—such as increasing canopy openness, herb cover or deadwood availability—moderated declines in the number of arthropod species. More research is needed to identify mechanistic relationships. Tree mortality included both natural mortality and timber harvesting. Forest sites had a stand age of, on average, 116 years (minimum of 30 years and maximum of 180 years) and therefore did not include overmature stands. Owing to stand age and because management was abandoned 20 to 70 years before this study started, natural tree mortality was low even in unmanaged stands. We expect increasingly positive effects of natural tree mortality and associated increased structural diversity and heterogeneity40 on arthropod trends with increasing stand age, but further research is required. In Germany, harvesting is usually conducted as shelterwood cutting. In our sites, the harvested amount over the course of our study reached a maximum of 1% of the standing volume per year. More intense harvesting systems (such as clear cutting), which lead to less heterogeneous habitat conditions, may not have similar moderating effects on arthropod declines.

Extended Data Fig. 5 Distribution of landscape-level land-use variables.

Data distribution of the cover of arable fields, grassland and forest within 1,000 m surrounding each of the 150 grassland and 30 forest sites for each region, and for all regions in total. ALB, Schwäbische Alb; HAI, Hainich-Dün; SCH, Schorfheide-Chorin.

Extended Data Fig. 6 Correlations among weather and among land-use variables.

a, b, Coefficients of pairwise correlations and PCAs for weather variables (a) and land-use variables (b). Temperature-related data are based on observed air temperature by weather stations at each site. Precipitation is derived from gauge-corrected radar observations (RADOLAN, Deutscher Wetterdienst). For each site and year, we calculated mean temperature (T mean), number of frost days (daily minimum temperature <0 °C; n frost), number of warm days (daily mean temperature >20 °C; n warm days) and precipitation sum in mm (precipitation) for three different periods: winter (November of the previous year to February; win), growing period (March to October; grow) and year (November of the previous year to October; year). The number of independent observations for weather variables was n = 1,406 (grasslands) or 266 (forests). Land-use variables include local land-use intensity (local LU) and cover of arable fields (A), grassland (G) and forest (F) at different spatial scales (250, 500, 1,000, 1,500 and 2,000 m). The number of independent observations for land-use variables equalled the number of sites; n = 150 (grasslands) or n = 30 (forests). On the basis of correlations and PCA results, we chose mean winter temperature and precipitation during the growing period, as well as cover of arable fields and cover of grassland, as ecologically meaningful and the least-correlated explanatory variables for modelling arthropod data.

Extended Data Fig. 7 Temporal patterns in weather conditions.

Temporal patterns of the sum of precipitation during the growing period (March to October) and mean winter temperature (November of the previous year to February) for 150 grassland and 30 forest sites (n = 1,406 (grassland) or 266 (forest) independent observations). Boxes represent data within the 25th and 75th percentile, black lines show medians and whiskers show 1.5× the interquartile range. A linear mixed model for each response variable, with year as a fixed effect and the site nested in the region as a random effect, indicate that winter temperature increased (grassland, z = 10.90, P ≤ 0.001; forest, z = 8.24, P ≤ 0.001) and precipitation during the growing period decreased during our study period (grassland, z = −6.53, P ≤ 0.001; forest, z = −8.44, P ≤ 0.001). We are currently not able to quantify whether and how much the observed trends in arthropod numbers were affected by changes in climatic conditions (Supplementary Information section 2).

Extended Data Fig. 8 Results from multiscale analysis.

Mean and s.d. of Pearson’s coefficients of correlation between arthropod numbers (biomass, abundance and number of species) and landscape-level land-use variables (cover of arable fields and cover of grassland) for radii of 250–2,000 m around 150 grassland sites and 30 forest sites. Only data from a random subset of sites with non-overlapping buffers at the 2,000-m scale were used. The randomized subsampling of sites with non-overlapping buffers and the calculation of correlations was repeated 100 times (median number of sites per subsample was n = 18 (grassland) or 17 (forest)). The 1,000-m scale was used for modelling arthropod numbers for both grassland and forests because (i) the correlation coefficients appeared to plateau at this scale in grasslands, (ii) the range of landscape-level land-use variables at small spatial scales in forests was small and (iii) buffers of neighbouring plots overlapped more extensively at higher spatial scales.

Extended Data Table 1 Details on arthropod numbers

Supplementary information

Supplementary Information

This file contains Supplementary Sections 1-8 including Supplementary Tables and additional references.

Reporting Summary

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Seibold, S., Gossner, M.M., Simons, N.K. et al. Arthropod decline in grasslands and forests is associated with landscape-level drivers. Nature 574, 671–674 (2019) doi:10.1038/s41586-019-1684-3

Download citation

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.