Skip to main content
ARS Home » Pacific West Area » Tucson, Arizona » SWRC » Research » Publications at this Location » Publication #324416
Title: Agroecosystem research with big data and a modified scientific method using machine learning concepts

Author
item Moran, Mary
item Heilman, Philip - Phil
item Peters, Debra
item Holifield Collins, Chandra

Submitted to: Ecosphere
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/4/2016
Publication Date: 10/1/2016
Citation: Moran, M.S., Heilman, P., Peters, D.C., Holifield Collins, C.D. 2016. Agroecosystem research with big data and a modified scientific method using machine learning concepts. Ecosphere. 7(10):e01493. https://doi.org/10.1002/ecs2.1493.
DOI: https://doi.org/10.1002/ecs2.1493

Interpretive Summary: The new USDA Long-term Agro-ecosystem Research (LTAR) network has thrust agricultural research into uncharted territory by promoting a long-term, continental-scale coordinated research program. This network-wide research has inspired a modified scientific method where new ideas derive from individual scientists while the hypothesis, analytics, data and conclusions are developed in collaboration with the broader scientific community. This, in turn, supports a non-traditional system of credit for co-authors based on publication impact with less regard for author order. Examples of research using this new method have resulted in solid scientific contributions in high-impact journals, with high citation records and recent awards. The LTAR network has embraced this modified scientific method in its Shared Research Strategy (SRS) and Common Experiment to address the problematic issues of data quality, credit, efficiency, and impact in data intensive research. This report offers direction for the LTAR and other such networks going forward.

Technical Abstract: Long-term studies of agro-ecosystems at the continental scale are providing an extraordinary understanding of regional environmental dynamics. The new Long-Term Agro-ecosystem Research (LTAR) network (established in 2013) has designed an explicit research program with multiple USDA experimental watersheds, ranges and forests for cross-site studies. Here, we report results from studies using a modified scientific method implemented over the past five years with long-term data from USDA experimental sites in coordination with other networks. The results offer a compelling argument for the LTAR concept of combining bottom-up site-based expertise and top-down network-wide coordination to arrive at more accurate scientific conclusions. Simply put, without site-based expertise and cross-site communication, the interpretations and conclusions of these studies would have been incomplete, if not incorrect. Further, the up-front time commitment to data processing and analytics above the time dedicated to place-based studies increased the productivity of the team and the impact of the research, unlike the common perception that cross-site research might be less efficient. In turn, this supported a non-traditional system of credit for co-authors based on publication impact with less regard for author order. The LTAR network has embraced this modified scientific method in its Shared Research Strategy and Common Experiment to address the problematic issues of data quality, co-author credit, research efficiency, and scientific impact in data intensive research.