Emerging nutrient management databases and networks of networks will have broad applicability in future machine learning and artificial intelligence applications in soil and water conservation

Authors: Delgado, Jorge; Vandenberg, Bruce; Neer, Donna; D Adamo, Robert

Submitted to: Journal of Soil and Water Conservation
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/25/2019
Publication Date: 11/18/2019
Citation: Delgado, J.A., Vandenberg, B.C., Neer, D.L., D Adamo, R.E. 2019. Emerging nutrient management databases and networks of networks will have broad applicability in future machine learning and artificial intelligence applications in soil and water conservation. Journal of Soil and Water Conservation. 74(6):113A-118A. https://doi.org/10.2489/jswc.74.6.113A.
DOI: https://doi.org/10.2489/jswc.74.6.113A

Interpretive Summary: Nutrient management, including management of nitrogen inputs, was a key part of the Green Revolution, which increased global agricultural production and helped feed the world in the latter half of the 20th century. While nitrogen is often applied to increase yields, and is indeed essential to meet the increased production demands that inevitably follow population growth, an extensive number of studies from regions throughout the globe have reported negative impacts related to nitrogen losses to the environment (Smith et al. 2018; Follett and Walker, 1989; Hey, 2002; Hey et al 2005; Greenhalch, et al 2003;Delgado and Follett 2010; Juergens-Gschwind 1989; Dubrovsky et al. 2010; Glebe 2006).

Technical Abstract: This special issue about the USDA-ARS AgCROS/NUOnet has a series of papers that present data related to soil and water conservation efforts. These papers cover areas related to nitrogen balances, runoff and leaching (Vadas et al., Sainu et al); nitrogen management in different tillage systems (Balkcom et al.), nutrient distributions in grazed pastures (Franzluebbers et al.), runoff and nutrient losses from conventional and conservation tillage systems (Endale et al.), increasing infiltration into saturated buffers (Jaynes et al.), a conservation planning and evaluation tool (White et al), nitrogen sources and rates (Mikha et al.) and integration of data for a sustainable food system (Finley et al.). Data from some of these papers have already been uploaded to NUOnet or to the USDA-ARS National Agricultural Library. The goal of this special issue is to present data about soil and water conservation and show the potential of the diversity of the data that could eventually form part of NUOnet and AgCROS, including data ranging from research plots to regional to national assessments. The future of machine learning and artificial intelligence applications in soil and water conservation is here, and we will be able to participate in the next revolution in precision agriculture and agriculture in general, which will be driven by SPAE using databases, networks of networks, and systems of systems (Delgado et al. 2019), and will have data similar to the kind presented in this special issue that will follow the FAIR (Findable, Accessible, Interoperable, Reusable) Data Principles described by Wilkinson (2016). These USDA-ARS databases and networks of networks will aid in the application of machine learning and artificial intelligence in soil and water conservation, contributing to reduced erosion and offsite transport of nitrogen and phosphorus, increased cycling of macro- and micro-nutrients, and greater adaptation to the 21st century’s greatest challenge to agricultural sustainability: a changing climate.