We aggregated pairwise service observations to provide a system-wide visualization of cover crop assessments and common ES frameworks. In contrast to Bayesian network analyses, our network map is based on descriptive statistics. Ecosystem service bundles were mapped to observe common knowledge frameworks used in cover crop research, and consistent pathways of investigation . The Tableau ® Data Analytics and Visualization Software was used to develop network maps. Figures 2.5, 2.6 & 2.7 provide visualization of scientific observations of monitored ES, expressed in the form of nodes, scientific observations of cooccurring pairwise services, in the form of links, and research replications in the form of size and coloration of nodes and links. Nodes were mapped around a circle shape. The width of links followed a continuous scale, reflecting research coverage. The coloration of links was stepped in 3 increments on a 35 scale: increment 1 in yellow illustrates research coverages of 1-12%, increment 2 in orange illustrates coverage of 13-24% and increment 3 in red corresponds to 25-35% coverage. Figures 2.6 & 2.7 uses the same coloration increments as Figure 2.5, with shades of blue. Linkages are not to be read as correlations, rather they represent the pairwise co-occurrence of services within a given study. We define “common knowledge frameworks”, bato bucket as bundles of services which commonly cooccur within a same study and which have received substantial research attention .
The structures of these bundles describe sets of ecosystem services, which are commonly connected within scientific rationales. Thereby, we can visualize the construction of researchers’ mental models and experts’ shared conceptualization of cover crop functioning. Visually, specialized frameworks appear as clusters within the network of ecosystem services, and are defined as densely connected ecosystem services, represented by nodes, which are sparsely connected to other clusters of the network. These specialized clusters define mutually exclusive networks of ES linkages of high research coverage . We recorded 19 ecosystem services associated with cover cropping, in perennial agroecosystem literature. Our definition of ecosystem services was based on the Millennium Ecosystem Assessment , whose framework is based on works of Costanza et al. and Daily and based on the framework described by Schipanski et al. for reporting ecosystem services, provided by cover cropping, in agroecosystems. The ecosystem service approach allows for improved representation of dynamic cross-service interactions, as well as the integration of social and environmental perspectives within agroecosystem assessments . Although much of cover crop research predates the introduction of ‘ecosystem services’ as a concept, most studies have reported co-occurring ecosystem functions bundled in sets, thereby describing multifunctional processes . We surveyed cover crops ecosystem services to understand the broader benefits of cover cropping, beyond productivity and biodiversity. We recorded research replication and the distribution of studies in the reporting of individual services. Pest suppression services included above ground and below ground suppression of parasitic nematodes, insect pests and parasitic fungi, and included disease prevention.
Biodiversity included above ground metrics for plant and insect diversity as well as below ground soil fauna, nematode and microbial diversity. Biomass production referred to net primary production, including non-marketable crop growth as well as cover crop productivity.Of monitored services, 10 ecosystem services were regulating, 7 were supporting services and 3 were provisioning services. Whereas nutrient cycling was investigated in 36% of studies, greenhouse gas regulation was reported in <1% of studies . Descriptive statistics revealed a 10-fold research gap between the 5 most-commonly reported ES and the 7 least-reported ES colonization, pollination, wildlife habitat, NO3 – leaching and GHG regulation. In Figure 2.4, hot spots in research coverage reveal common scientific pathways and a systemization of knowledge production. The heterogenous distribution of research among pairwise ES linkages reveals a non-randomization of scientific observations and therefore, a valuation of knowledge. We observed limited or no representation of certain ES. Of 153 potential pairwise ES associations, nearly a third of ES-interactions remain unexplored . Cover crop assessments followed weighed systems of ES valuation, reflecting the different interests and goals of the studies with respect to sustainability . Our analysis revealed that 90% of articles could be categorized into two specialized frameworks of cover crop ES assessment: the first is a “nutrient management framework”, which valued linkages between soil resource management services : nutrient cycling, N mineralization, soil C, water dynamics, soil structure, soil retention, AMF colonization, NO3 – leaching and GHG regulation.
The second framework, referred to as the “biological management framework”, attributed greater value to linkages among biological services: pollination, wildlife habitat, pest suppression, beneficial insect conservation and weed suppression. Thus, we observe that cover crop knowledge is systematized by scientific domain. Between these two domains, service interactions were poorly explored, delimiting two distinct,mutually exclusive frameworks of cover crop assessment. Crop yields were reported in 49% of nutrient management studies but only 22% of biological management studies. Instead, biological management studies attributed greater value to biodiversity services, monitored in 68% of studies, compared to only 22% in nutrient management frameworks. Thereby, beyond the systemization of knowledge within specialized scientific domains, the two frameworks display separate approaches to sustainability: the first one focused on economic productivity and the second towards biological conservation.Our results reveal that cover crop studies conducted in perennial systems included on average 2 species per cover crop in addition to the cash crop and 4 ES per assessment. Of 638 distinct cover crops reported in the literature, 73% were single species. Furthermore, 43% of studies only studied one cover crop design, half of which were single species cover crops. Our results indicate that literature for perennial systems has relied heavily on a limited subset of species in its assessment of cover crop ES. Of 1,446 cover crop trials, ~80% used species belonging to either the Fabaceae, Poaceae or Asteraceae plant families . All other plant families appeared in ≤13% of trials . Of 441 tested species, the five most-studied species were used in 16% of trials. Overall, Trifolium was the most common plant genus, regardless of the intended use of the cover crop. We suggest that the reliance of research on only a handful of cover crop species reflects the limited availability of seed varieties. Our results highlight the need to expand breeding programs as well as seed distribution, so as to broaden the diversity of available cover crop seed options.While nutrient management frameworks are primarily based on supporting services, biological management frameworks are structured on regulating services. Thus, ES are fragmented categorically by their supporting and regulating value . Supporting value suggests an underlying paradigm of ableism and exists on a discrete scale, whereby ecosystem states are differentiated on the basis of an ability to serve. The paradigm of regulation is the preservation of multiple processes and ecosystem qualities: the purpose of regulation is one of moderation and balance, on a continuous scale. Thereby, we observe that each specialized framework abides to separate logics. The first logic, associated with nutrient management frameworks, favors profitability over biodiversity. It ascribes to an anthropogenic view of agriculture , whereby resource conservation is valued for its ability to sustain socio-economic growth. The second logic, associated with biological management, attributes a higher value to the preservation of nature and its biodiversity, in addition to continued productivity, thereby taking a more biocentric approach. Figure 2.7 further illustrates the divide in cover crop assessment structures, defined by researchers’ approach to yields. Comparing across frameworks, we notice strong similarities between productivity and nutrient management frameworks, and noticeable differences between conservation and biological management frameworks. Notably, only the biological management framework links biological processes to gains in net primary production. The fragmentation of scientific research across distinct sustainability paradigms leaves the immensely complex work of interpretation to practitioners. We suggest that addressing lags in the uptake of science-driven solutions requires negotiating a shared strategy, across differential value systems.
Much of the rationale behind the use of multi-species cover crops is based on the tested diversity-productivity theory, dutch bucket hydroponic whereby increased diversity is expected to increase cover crop primary productivity and ES, through higher resource use efficiency . In 1997, foundational work by Tilman et al. demonstrated that ecosystem processes not only depend on the identity of species but also the number of species within an ecosystem. It was shown that the integration of diverse mixtures of species within an ecosystem could support a suite of ecosystem services. At the time, it had been established that productivity could accrue with the addition of up to 5 species. Tilman’s later work demonstrated increases in ecosystem productivity with up to 16 species . Although these studies were conducted in natural systems, other studies suggest that intentionally increasing plant diversity through cover cropping could substantially augment agroecosystem services . Our results indicating an average of 2 species per cover crop in addition to the cash crop and 4 ES per assessment are in contradiction with Tilman’s diversity productivity theory. The limited use of polycultures may be attributed to the limited availability of improved seed varieties for cover cropping and existing lags in breeding programs for cover crop cultivar development. With unimproved seed varieties, the establishment of species-rich assemblages is especially challenging, due to poor germination and antagonistic interactions between species, limiting productivity. Yet, the use of simplified one- to two-species cover crops restrains our capacity to evaluate the full benefits of higher diversity and of the ecosystem services, supported by polycultures. The lack of marketability of certain ES limits research advancements, which in turn impact stakeholders’ valuation and their willingness to invest in these services. We observe that many of unexplored ES linkages pertain to conservation measures and that these ES associations have not been presently ascribed a marketable value. Although some ES linkages may not have enough substance for investigation in research, exploring some of these data gaps could bring valuable updates to our current conception of marketable ES. In his work, Swinton et al. discuss the importance of public engagement in valuing ecosystem services for agriculture . Swinton et al. suggest that the development of measurement systems is key to create markets and policies, which support the provision of ecosystem services and considers them as outputs . Frameworks have been developed specifically for cover cropping to account for ES bundles and co-occurrences . Beyond the definition of indicators and metrics, assessment structures need to account for dynamics between services in the form of synergies and tradeoffs. Frameworks such as the Sustainable Intensification Framework provide a set of methods, practices, and principles to account for ecosystem dynamics . This framework integrates social services in its environmental assessment, which evaluates provisioning, regulating and supporting services. Our results indicated that social and cultural services are especially under-represented in the scientific literature. Due to the low representation of knowledge diffusion and cultural services in the articles of our study, these services were not included within our final ES visualization . However, we recognize their importance and the esthetic, spiritual and cultural value of perennial agro-ecosystems. Dendoncker et al. suggest that the transition towards agroecological practices requires the development and use of integrated ES valuation systems in science. The Sustainable Intensification Framework proposes a set of methods to integrate socio-cultural services with the assessment of agro-ecosystem sustainability . Other frameworks and processes have also been developed to integrate different motivations for cover cropping: the ES assessment framework for agroecological transitions in practice , the goal-oriented indicator framework , the adaptive learning process and the multi-criteria, multi-stakeholder assessment for cover cropping . Although not yet applied to cover cropping, system-based modelling efforts have developed multilevel story lines and participatory socio-environmental modules to better represent different realities and perspectives within agro-environmental sustainability assessments . Our dataset reveals that distinct schools of thought have commonly adopted the language of “cover crops”. Due to existing differences in the selection of metrics, particularly yields and economic indicators, integrating each realm of research into tradeoff assessments and comprehensive frameworks may be particularly challenging. Although cover crop knowledge frameworks are intricately contextualized and systematized, the design of cover crops is not. Addressing barriers to cover crop adoption requires that seed design rationales be integrated within cover crop sustainability discourse. Indeed, formal literature rarely explicitly describes researchers’ processes of seed selection or clarify the chosen design criterion of tested cover crops. Our analysis reveals that, within ES assessments, primary production services have largely been recognized as the core component of ES provision. Biomass production was the most-reported ES, monitored in 64% of studies. In comparison, biodiversity was monitored in only 36% of studies .