Search USGS⌕ Search
Home/Archive/USGS

USGS · 70216895

Quantifying and addressing the prevalence and bias of study designs in the environmental and social sciences

Alec P. Christie·David Abecasis·Mehdi Adjeroud·Juan C. Alonso·Tatsuya Amano·Alvaro Anton·Barry P. Baldigo·Rafael Barrientos·Jake E. Bicknell·Deborah A. Buhl·Just Cebrian·Ricardo S. Ceia·Luciana Cibils-Martina·Sarah Clarke·Joachim Claudet·Michael D. Craig·Dominique Davoult·Annelies De Backer·Mary K. Donovan·Tyler D. Eddy·Filipe M. Franca·Jonathan P.A. Gardner·Bradley P. Harris·Ari Huusko·Ian L. Jones·Brendan P. Kelaher·Janne S. Kotiaho·Adrià López-Baucells·Heather L. Major·Aki Maki-Petays·Beatriz Martinez-Lopez·Carlos A. Martin·Philip A. Martin·Daniel Mateos-Molina·Robert A. McConnaughey·Michele Meroni·Christoph F. J. Meyer·Kade Mills·Monica Montefalcone·Norbertas Noreika·Carlos Palacin·Anjali Pande·C. Roland Pitcher·Carlos Ponce·Matthew J. Rinella·Ricardo Rocha·Maria C. Ruiz-Delgado·Juan J. Schmitter-Soto·Jill A. Shaffer·Shailesh Sharma·Anna A. Sher·Doriane Stagnol·Thomas Stanley·Kevin D.E. Stokesbury·Aurora Torres·Oliver Tully·Teppo Vehanen·Corinne Watts·Qingyuan Zhao·William J. Sutherland
View original source ↗Export metadata ↓

Abstract

Building trust in science and evidence-based decision-making depends heavily on the credibility of studies and their findings. Researchers employ many different study designs that vary in their risk of bias to evaluate the true effect of interventions or impacts. Here, we empirically quantify, on a large scale, the prevalence of different study designs and the magnitude of bias in their estimates. Randomised designs and controlled observational designs with pre-intervention sampling were used by just 23% of intervention studies in biodiversity conservation, and 36% of intervention studies in social science. We demonstrate, through pairwise within-study comparisons across 49 environmental datasets, that these types of designs usually give less biased estimates than simpler observational designs. We propose a model-based approach to combine study estimates that may suffer from different levels of study design bias, discuss the implications for evidence synthesis, and how to facilitate the use of more credible study designs.

Explore related subjects

Nature CommunicationsArticle

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Alec P. Christie, David Abecasis, Mehdi Adjeroud, Juan C. Alonso, Tatsuya Amano, Alvaro Anton, Barry P. Baldigo, Rafael Barrientos, Jake E. Bicknell, Deborah A. Buhl, Just Cebrian, Ricardo S. Ceia, Luciana Cibils-Martina, Sarah Clarke, Joachim Claudet, Michael D. Craig, Dominique Davoult, Annelies De Backer, Mary K. Donovan, Tyler D. Eddy, Filipe M. Franca, Jonathan P.A. Gardner, Bradley P. Harris, Ari Huusko, Ian L. Jones, Brendan P. Kelaher, Janne S. Kotiaho, Adrià López-Baucells, Heather L. Major, Aki Maki-Petays, Beatriz Martinez-Lopez, Carlos A. Martin, Philip A. Martin, Daniel Mateos-Molina, Robert A. McConnaughey, Michele Meroni, Christoph F. J. Meyer, Kade Mills, Monica Montefalcone, Norbertas Noreika, Carlos Palacin, Anjali Pande, C. Roland Pitcher, Carlos Ponce, Matthew J. Rinella, Ricardo Rocha, Maria C. Ruiz-Delgado, Juan J. Schmitter-Soto, Jill A. Shaffer, Shailesh Sharma, Anna A. Sher, Doriane Stagnol, Thomas Stanley, Kevin D.E. Stokesbury, Aurora Torres, Oliver Tully, Teppo Vehanen, Corinne Watts, Qingyuan Zhao, William J. Sutherland. 2020-12-11. Quantifying and addressing the prevalence and bias of study designs in the environmental and social sciences. https://doi.org/10.1038/s41467-020-20142-y

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

The saline groundwater legacy of a large buried coastal paleo-estuary

Elevated groundwater salinity in coastal regions threatens the beneficial use of fresh groundwater. Coastal groundwater management typically focuses on preventing intrusion from modern sources of seawater; however, past geological processes can also leave a legacy of saline groundwater now hidden in the subsurface. Here, multiple extensive airborne electromagnetic surveys provide detailed evidence of residual salinity from a paleo-estuary filling a late Pleistocene incised valley impacting more than 10,000 km 2 that is now hidden beneath coastal Louisiana’s deltaic plain. Our results show that the three-dimensional pattern of saline groundwater beneath Louisiana mimics that of near-surface aquifers surrounding the modern Delaware Bay estuary, fingerprinting the signature of the past drowning of a large, incised valley of the Mississippi River following post-glacial sea-level rise. These findings demonstrate a new framework for understanding legacy sources of saltwater critical for managing stressed water resources along global coastlines.

Louisiana↗
USGS2026

Previous prescribed burns saved thousands of ancient sequoias during historically unprecedented wildfires

Wildfires in California’s Sierra Nevada during 2020–2021 killed giant sequoias ( Sequoiadendron giganteum ) at rates unseen for millennia, underscoring the vulnerability of highly fire-adapted trees to ongoing environmental change. Following a century of fire exclusion and fuel accumulation, the effectiveness of prescribed burns in reducing giant sequoia mortality from wildfire remained poorly quantified. Here we estimate mortality outcomes for 26,403 giant sequoias across 19 groves in Sequoia and Kings Canyon national parks following the Castle (2020) and KNP Complex (2021) wildfires using a Bayesian framework. We map tree mortality using a deep learning classifier integrating 3 m PlanetScope imagery, airborne lidar, and field observations. From an estimated 7,974 sequoia deaths (95% Bayesian credible interval (CI): 7,555–8,430), corresponding to 30.2% mortality (CI: 28.6–31.9%), we find previous prescribed burns (≤10 years prior) reduced mortality odds by 77% (CI: 69–83%), making treated trees nearly four times more likely to survive. Counterfactual simulations suggest that prescribed burns prevented at least 1,888 (CI: 1,487–2,302) deaths, and universal treatment would have saved an additional 3,888 (CI: 3,236–4,580) giant sequoias. These results show that prescribed burns substantially improve survival during extreme wildfires, offering guidance for conserving long-lived, fire-adapted forests under intensifying fire regimes.

California↗
USGS2026

Tectonically driven integration of the 4.8 Ma Colorado River USA tracked with detrital sanidine and fish genetics

The development of the continental-scale Colorado River system, western USA, from 8 to 4.8 Ma, is revealed using 60-40 Ma detrital sanidine tracer grains and fish phylogeny. Here we show that precursor paleoriver segments became integrated north to south as traced by 60-40 Ma sand grains that were derived from the north and sequentially appeared in the 25-8 Ma Browns Park Formation of Utah, 7-6 Ma upper Bidahochi Formation of Arizona, and 4.8 Ma Bouse Formation of the lower Colorado River and proto Gulf of California. This timing is mimicked by molecular clock estimates of divergence times among fish lineages. River integration was a response to headwater uplifts in the Yellowstone hotspot track and Rocky Mountains. 40 Ar/ 39 Ar ages refine the timing for mantle-drips that caused subsidence, then uplift, of depositional basins that influenced the integration pathway and tempo. The ~ 3-million-year timescale suggests that multiscale mantle-driven uplift, rather than lake spillover, was the primary driver for integration of the proto-Colorado River through Grand Canyon.

Colorado River↗