Search

Summarizing information from large bodies of scientific literature is an essential but work-intensive task. This is especially true in environmental studies where multiple factors (e.g., soil, climate, vegetation) can contribute to the effects observed. Meta-analyses, studies that quantitatively summarize findings of a large body of literature, rely on manually curated databases built upon primary publications. However, given the increasing amount of literature, this manual work is likely to require more and more effort in the future. Natural language processing (NLP) facilitates this task, but it is not clear yet to which extent the extraction process is reliable or complete. In this work, we explore three NLP techniques that can help support this task: topic modeling, tailored regular expressions and the shortest dependency path method. We apply these techniques in a practical and reproducible workflow on two corpora of documents: the Open Tension-disk Infiltrometer Meta-database (OTIM) and the Meta corpus. The OTIM corpus contains the source publications of the entries of the OTIM database of near-saturated hydraulic conductivity from tension-disk infiltrometer measurements (https://github.com/climasoma/otim-db, last access: 1 March 2023). The Meta corpus is constituted of all primary studies from 36 selected meta-analyses on the impact of agricultural practices on sustainable water management in Europe. As a first step of our practical workflow, we identified different topics from the individual source publications of the Meta corpus using topic modeling. This enabled us to distinguish well-researched topics (e.g., conventional tillage, cover crops), where meta-analysis would be useful, from neglected topics (e.g., effect of irrigation on soil properties), showing potential knowledge gaps. Then, we used tailored regular expressions to extract coordinates, soil texture, soil type, rainfall, disk diameter and tensions from the OTIM corpus to build a quantitative database. We were able to retrieve the respective information with 56 % up to 100 % of all relevant information (recall) and with a precision between 83 % and 100 %. Finally, we extracted relationships between a set of drivers corresponding to different soil management practices or amendments (e.g., “biochar”, “zero tillage”) and target variables (e.g., “soil aggregate”, “hydraulic conductivity”, “crop yield”) from the source publications' abstracts of the Meta corpus using the shortest dependency path between them. These relationships were further classified according to positive, negative or absent correlations between the driver and the target variable. This quickly provided an overview of the different driver–variable relationships and their abundance for an entire body of literature. Overall, we found that all three tested NLP techniques were able to support evidence synthesis tasks. While human supervision remains essential, NLP methods have the potential to support automated evidence synthesis which can be continuously updated as new publications become available.

Adopting soil and crop management practices that conserve or enhance soil structure is critical for supporting the sustainable adaptation of agriculture to climate change, as it should help maintain agricultural production in the face of increasing drought or water excess without impairing environmental quality. In this paper, we evaluate the evidence for this assertion by synthesizing the results of 34 published meta-analyses of the effects of such practices on soil physical and hydraulic properties relevant for climate change adaptation in European agriculture. We also review an additional 127 meta-analyses that investigated synergies and trade-offs or help to explain the effects of soil and crop management in terms of the underlying processes and mechanisms. Finally, we identify how responses to alternative soil–crop management systems vary under contrasting agro-environmental conditions across Europe. This information may help practitioners and policymakers to draw context-specific conclusions concerning the efficacy of management practices as climate adaptation tools. Our synthesis demonstrates that organic soil amendments and the adoption of practices that maintain “continuous living cover” result in significant benefits for the water regulation function of soils, mostly arising from the additional carbon inputs to soil and the stimulation of biological processes. These effects are clearly related to improved soil aggregation and enhanced bio-porosity, both of which reduce surface runoff and increase infiltration. One potentially negative consequence of these systems is a reduction in soil water storage and groundwater recharge, which may be problematic in dry climates. Some important synergies are reductions in nitrate leaching to groundwater and greenhouse gas emissions for nonleguminous cover crop systems. The benefits of reducing tillage intensity appear much less clear-cut. Increases in soil bulk density due to traffic compaction are commonly reported. However, biological activity is enhanced under reduced tillage intensity, which should improve soil structure and infiltration capacity and reduce surface runoff and the losses of agro-chemicals to surface water. However, the evidence for these beneficial effects is inconclusive, while significant trade-offs include yield penalties and increases in greenhouse gas emissions and the risks of leaching of pesticides and nitrate. Our synthesis also highlights important knowledge gaps on the effects of management practices on root growth and transpiration. Thus, conclusions related to the impacts of management on the crop water supply and other water regulation functions are necessarily based on inferences derived from proxy variables. Based on these knowledge gaps, we outlined several key avenues for future research on this topic.

Determining soil hydraulic properties is of major concern in various fields of study. Although stony soils are widespread across the globe, most studies deal with gravel-free soils, so that the literature describing the impact of stones on the hydraulic conductivity of a soil is still rather scarce. Most frequently, models characterizing the saturated hydraulic conductivity of stony soils assume that the only effect of rock fragments is to reduce the volume available for water flow, and therefore they predict a decrease in hydraulic conductivity with an increasing stoniness. The objective of this study is to assess the effect of rock fragments on the saturated and unsaturated hydraulic conductivity. This was done by means of laboratory experiments and numerical simulations involving different amounts and types of coarse fragments. We compared our results with values predicted by the aforementioned predictive models. Our study suggests that it might be ill-founded to consider that stones only reduce the volume available for water flow. We pointed out several factors of the saturated hydraulic conductivity of stony soils that are not considered by these models. On the one hand, the shape and the size of inclusions may substantially affect the hydraulic conductivity. On the other hand, laboratory experiments show that an increasing stone content can counteract and even overcome the effect of a reduced volume in some cases: we observed an increase in saturated hydraulic conductivity with volume of inclusions. These differences are mainly important near to saturation. However, comparison of results from predictive models and our experiments in unsaturated conditions shows that models and data agree on a decrease in hydraulic conductivity with stone content, even though the experimental conditions did not allow testing for stone contents higher than 20 %.