Managing soil functions for a sustainable bioeconomy—Assessment framework and state of the art

Bioeconomy strategies have been adopted in many countries around the world. Their sustainable implementation requires a management of soils that maintains soil functions and avoids land degradation. Only then, ecosystem services can be maintained and resources used efficiently. We present an analytical framework for impact assessment that links policy and technology driving forces for soil management decisions to soil processes, soil functional changes, and their impacts on ecosystem services and resource use efficiency, both being targets that have been set by society and are anchored in bioeconomy policy strategies and sustainable development goals. Although the resource use efficiency concept has a long‐term tradition, most studies of agricultural management do not address the role of soils in their efficiency assessment. The concept of ecosystem services has received increasing attention over the last years; however, its link to soil functions and soil management practices is still not well established. This study is the first to conceptually link the socioeconomic processes of external drivers for soil management with the natural processes of soil functions and connect them back to impacts on the social system. Application of the framework helps strengthen the science‐policy interface and to systemically assess and compare the opportunities and threats of soil management practices from the perspective of goals set by society at different spatial and temporal scales. Insights gained in this way can be applied in stakeholder decision‐making processes and used to inform the design of governance instruments aimed at sustainable soil management within a bioeconomy.

dietary changes and a globally increasing demand for food (Alexander et al., 2015) and, in many countries, a transition from fossil-based economies towards bioeconomies (Fund, El-Chichakli, Patermann, & Dieckhoff, 2015). Bioeconomy strategies strive for the agricultural production of more nonfood products, including energy and materials to substitute fossil resources, in addition to food products. However, ensuring a balanced system of soil functions and thereby avoiding soil degradation is important, not only to sustain soil fertility in the long term but also to meet other targets set by society, such as climate change mitigation and the support of regulating ecosystem services (flood control, cooling, buffering of weather extremes, and biodiversity) and cultural ecosystem services (recreation, heritage, and amenity;Dominati, Patterson, & Mackay, 2010). Additionally, biomass production needs to follow resource use efficiency considerations to avoid jeopardizing resource saving purposes inherent in the bioeconomy strategies (Scarlat, Dallemand, Monforti-Ferrario, & Nita, 2015). The term "sustainable intensification" (Garnett et al., 2013) was introduced to signify the challenge for agricultural management to increase production without increasing environmental pressures.
This goal can only be achieved when agricultural resources are utilized efficiently and when ecological interactions are intensified to stimulate the inherent capacity of soil to produce biomass (Tittonell, 2014).
Operationalizing the concept of sustainable intensification from the perspective of soils requires improvements in at least two research fields: soil management-soil process interactions and impact assessment. Impact assessment is a means to synthesize scientific knowledge to inform policy and management (Carpenter, Bennett, & Peterson, 2006). It implies a linkage of the socioeconomic system of target setting and decision-making with the natural system of biological, physical, and chemical process interactions. It weaves an analytical thread from driving forces to management decisions to soil reactions, soil function changes, and their impacts on targets set by society. Such targets are usually defined in strategic policy documents such as the bioeconomy strategy and integrate across social, economic, and environmental targets. The knowledge gained by impact assessment supports stakeholders in making decisions in various contexts of decision-making, such as soil management (de Olde, Oudshoorn, Sørensen, Bokkers, & de Boer, 2016), governance and policy formulation (Podhora et al., 2013), or research design for sustainable development (Bond & Pope, 2012). To the best of our knowledge, no systematic, dynamic approach to an impact assessment of soil management and soil functions on targets set for a sustainable bioeconomy exists yet.
Although there is comprehensive, albeit not well systemized, evidence about the impacts of soil management on soil processes (Vogel et al., 2018), the linkage to targets set by society has only recently emerged in scientific and policy debate. Here, two partly interlinked assessment concepts stand out: ecosystem services and resource use efficiency. The concept of ecosystem services aims to demonstrate the value or contributions of nature to human societies (Costanza et al., 1997;Díaz et al., 2018;Haines-Young & Potschin, 2013;Millennium Ecosystem Assessment [MEA], 2005; The Economics of Ecosystems and Biodiversity [TEEB], 2010). The value can be seen as arising from the interaction of biotic and abiotic processes, and it specifically refers to the "final" outputs of ecological systems, that is, the goods and services directly consumed or used by people (Haines-Young & Potschin, 2013). Resource use efficiency is defined as the ratio of benefits generated by a product or process divided by the amount of resources used for that purpose (Di Maio, Rem, Baldé, & Polder, 2017). Both concepts are relevant to linking soil functions to bioeconomy targets and internationally acknowledged sustainable development goals (SDGs;General Assembly, 2015). Among the manifold linkages between soil functions and SDGs (Keesstra et al., 2016), three SDGs pertain most profoundly to soil functions from an ecosystem service perspective: SDG 2, target 2.4 (sustainable provision of food and regulating and maintenance services); SDG 13, target 13.2 (climate action-regulating services), strengthened by the "4 pour 1000"-initiative (Ministère de l'Agriculture, 2015); and SDG 15, target 15.3 (restoring degraded soils and achieving a land-degradation neutral world). One SDG pertains most profoundly to soil functions from a resource use efficiency perspective: SDG 12, target 12.2 (sustainable and efficient use of natural resources).
A comprehensive assessment would therefore require a combination of the two concepts of ecosystem services and resource use efficiency.
The objective of this paper is to construct a robust analytical framework for an impact assessment of soil management and soil functions and to compile state-of-the-art knowledge of its analytical components, including a thorough discussion of opportunities and limits for the application of the framework.
The analytical framework combines the two concepts of resource use efficiency and ecosystem services into an assessment that links driving forces and management decisions to soil reactions, soil function changes, and their impacts on bioeconomy and sustainable development targets. We have developed this framework in the context of the National German research program BonaRes-"Soil as a Sustainable Resource for the Bioeconomy" (www.bonares.de).

| ANALYTICAL FRAMEWORK FOR IMPACT ASSESSMENT
The analytical framework combines the driver-pressure-stateimpact-response (DPSIR) framework (Gabrielsen & Bosch, 2003) with the five steps of impact assessment (Helming, Diehl, Geneletti, & Wiggering, 2013). DPSIR was developed for the assessment of relations between human activities and the environment. The specific strength of the concept lies in its adaptability to various impact areas, objectives, and scales of analysis (Tscherning, Helming, Krippner, Sieber, & Paloma, 2012). For example, it has been proposed for soil system analysis by Bouma, de Vos, Sonneveld, Heuvelink, and Stoorvogel (2008) and Schjønning et al. (2015). Based on the DPSIR framework, we integrate the five steps of impact assessment ( Figure 1): (1) analysis of future trends and driving forces for soil management; (2) definition of human activities exerting pressures on ecological systems, which in our case is soil management; (3) analysis of the effects of human activities on the state of ecological systems, which are soil processes and soil functions; here, this analytical step concerns the soil system, depicts how soil processes are affected by soil management, and describes how soil processes impact the ensemble of soil functions; (4) assessment and valuation of direct and indirect impacts in the context of social, economic, and environmental targets; here, we use the resource use efficiency and ecosystem service concepts; and (5) elaboration of governance instruments-making use of assessment results-to provoke responses that counteract negative impacts and reinforce positive impacts. For example, (1) bioenergy policy (driver) can increase the demand for biomass and (2) motivate management change towards a higher share of maize in the crop rotation (pressure); (3) this narrower crop rotation affects the soil (e.g., changed and nutrient balance and altered composition of soil microbial communities) which in turn (4) has an impact on ecosystem services (increased nitrogen discharge, higher risk of water erosion) and resource use efficiency (yield per hectare, revenue per cost) with a consequence for the terrestrial ecosystem (SDG15). Based on an assessment of this causal chain, (5) policy makers may decide to revise the bioenergy policy (response). The five assessment steps are elaborated in subsequent sections, and state-of-the-art knowledge in the context of sustainable intensification is provided for each.
We particularly emphasize analytical Step 4, the impacts step. It evaluates the consequences of human activities on social and ecological systems from the perspective of targets set by society, thereby taking an anthropocentric viewpoint. In our framework, we focus on soil management practices as the human activities that trigger changes in soil functions, and the latter are understood to be the key linkages between ecological and societal systems. Depending on the system components of such linkages, impacts of human activities can be assessed by different value systems. Our framework has been developed to study the impacts of soil management practices in industrialized agricultural systems with low yield gaps and a low share of agriculture in the gross domestic product, as is the case in most industrialized countries. Targets set by society for such soil management systems are characterized by an ecosystem services perspective, that is, increasing biomass production while maintaining the contribution of soil functions to the other ecosystem services; and by resource use efficiency, that is, an increasing return from invested resources ( Figure 1). The latter is rarely considered in recent studies on the value of soil functions for society Schwilch et al., 2016;Stavi, Bel, & Zaady, 2016), or it is not as explicit as the ecosystem service perspective (Keesstra et al., 2016). While both assessment concepts are at first sight motivated by environmental perspectives, the economic perspective is inherent in either of them. Embedded in the natural capital concept (Costanza et al., 1997), ecosystem services are understood as natural assets that can be evaluated in terms comparable to economic activities and manufactured capital. The resource use efficiency concept directly considers the economic dimension through indicators describing a benefit-cost ratio. Costs may be monetary or arise in the form of resources required. A high resource use efficiency is generally required to achieve competitive production. It is important to note that, for other agricultural systems (e.g., small holder farming in developing countries), additional and/or different assessment perspectives are appropriate, for example, motivated by targets such as nutrition, poverty alleviation, or rural development.
The proposed framework seeks to establish an analytical basis for the identification of indicators for impact assessment. The framework allows two modes of interaction between soil management (key pressure) and impacts: soil-borne interactions via changes in soil processes and soil functions (solid arrow in Figure 1) and management-induced interactions irrespective of changes in the soil system (thin arrows in Figure 1). The definition of system boundaries, including the governance level and spatial and temporal scales, is essential for the assessment of management impacts. The chosen boundaries must reflect whether relevant impacts are expected at the respective spatial or temporal scale to avoid misleading results. Inclusion of effects up to a global scale may be required to account for spill-over effects, such as indirect land use changes, that is, the impacts of management decisions in one region emerging in other regions. The temporal scale must be FIGURE 1 Analytical framework for impact assessment of agricultural soil management and soil functions. Numbers refer to the five steps of the DPSIR system: drivers (1), pressures (2), states (3), impacts (4), and responses (5) [Colour figure can be viewed at wileyonlinelibrary.com] sufficiently wide to account for an improvement or deterioration of soil functions and ecosystem services through management, which often only emerge after considerable time lags (Fremier et al., 2013).
The challenges of assessing the impacts of these spill-over effects are discussed in Section 8.

| AGRICULTURAL SOIL MANAGEMENT AND ITS DRIVING FORCES
We use the term agricultural soil management for practices that affect soil processes and soil functions, and that are thus a pressure on soil in terms of the DPSIR framework. Agricultural soil management is driven by socioeconomic, biophysical, and technological drivers, such as pol- Spatial patterns of cropping affect soils in multiple ways. For example, field transition zones and adaptations to landscape contours shorten overland flow lines and reduce erosion (Panagos et al., 2015).
The same is true for the emerging practice of strip-tillage, which may also have potentially positive impacts on natural pest antagonists, microclimate and biological interactions (Williams et al., 2016).
Crop species and varieties have different characteristics that are relevant to soil, such as root structure and effects on pest suppression.
In Europe, approximately 60% of crop land is covered with cereals, half of which is wheat (Eurostat, 2016;Eurostat, 2017a). Recent trends, including emerging pesticide resistance and diversified consumer demand, may lead to increased diversity of crop rotations (DLG-Vorstand, 2017;Last, Buchmann, Gilgen, Grant, & Shreck, 2015); however, more narrow rotations are also possible in some regions, for example, due to climate change (Troost & Berger, 2015).
New breeding technologies, such as CRISPR/cas9, may improve crop varieties but may also generate novel risks for ecosystems (Science for Environment Policy, 2016).
Mechanical pressures on soils mainly originate from machine use, including tillage and field traffic. For example, in the UK, France, and Germany, conservation tillage was practiced on approximately 40% of arable land in 2010, whereas in Scandinavian countries, it was practiced on 10-22% of arable land (Eurostat, 2017b). Cost saving is an established driver of reduced tillage (Flessa et al., 2012), whereas soil threats, particularly soil erosion (Techen, Ries, & Steinführer, 2015) and climate adaptation may drive its adoption in the future (Olesen et al., 2011). The weight of agricultural machines has increased for decades, leading to compaction and damage to soil structure (Schjønning et al., 2015). Emerging mitigation measures include decision-support systems that optimize field traffic corresponding to site characteristics (Han, Steward, & Tang, 2015) and smaller, autonomous machines (e.g., Conesa-Muñoz, Gonzalez-de-Soto, Gonzalez-de-Santos, & Ribeiro, 2015;ecoRobotix Ltd., 2016); however, their future adoption is associated with numerous uncertainties.
The main inputs that farmers apply to soils are mineral fertilizers, organic material, pesticides and, in many countries, water. The pressures on soil in these categories differ by the applied quantity, spatial and temporal distribution and characteristics of the substance. Technological developments improve the site specificity and accuracy of application, leading to reduced inputs (especially for pesticides) and increased efficiency (especially for fertilizers; Flessa et al., 2012;Meuli, 2016). Pesticide application can also be reduced by promoting antagonists of pests, for example, by adding organic amendments or inoculating soil or seeds with beneficial organisms. The latter is still rare, but researchers see significant advances and potential for this technology (Johnson et al., 2016;Kergunteuil et al., 2016).
Opportunities for sustainable intensification especially lie in qualitative management changes (Garnett et al., 2013), which improve the inherent production capacity of the soil and thereby improve resource use efficiency and minimize adverse impacts on ecosystem services.
However, new practices may in general also create novel threats to soil functions and ecosystem services, which must be better understood and assessed to support decision-making towards a sustainable bioeconomy.

| SOIL PROCESSES AND FUNCTIONS
From an anthropocentric viewpoint, soils have several functions, and Blum's classification (Blum, 1990) became very influential in policy (Baveye, Baveye, & Gowdy, 2016), for example, in the Soil Thematic Strategy of the European Union (European Commission, 2006). From the perspective of agricultural soil management that affects earth system functioning, five soil functions are critical to our study: biomass production, storing and filtering of water, storing and recycling of nutrients, habitat for organisms, and carbon storage. The performance of soils regarding these functions depends on inherent soil properties (e.g., texture, depth, mineralogy, and horizon sequence), geo-biophysical site conditions (e.g., relief, climate, and altitude), their management history, and actual soil processes as affected by soil management practices. Soil functions are considered to be integral properties that emerge from complex interactions between physical, chemical, and biological processes depending on these manifold factors. The quantification of soil functions is a prerequisite for any assessment, and it may lead to trade-offs but can also lead to potential complementarities (Reed et al., 2013). Every function can also be seen as providing a soil-related contribution to ecosystem services (Bouma, 2014;Keesstra et al., 2016). Thus, soil functions are the linkage from soil systems' processes to the valuation of performance or their services in the context of sustainable development.
All soil science disciplines have developed sophisticated methods to explore physical, chemical, and biological processes using a broad spectrum of observable soil attributes. However, it is not yet obvious how steadily improving insights can be synthesized into a better understanding of soil functioning (Kibblewhite, Ritz, & Swift, 2008).
There is a need to identify soil functional characteristics that integrate systemic knowledge about the complex, nonlinear interactions between soil components and processes on various temporal and spatial scales on the one hand and to link them to soil management practices on the other hand. Soil functional characteristics emerge from the interactions between soil components (e.g., minerals, roots, and organisms) and soil processes (e.g., physical, chemical, and biological processes). They are sensitive to soil management and may change at a time scale of days to months. Examples are water capacity, aggregate stability, macropores, organic matter, and functional group diversity. The typical range of such functional characteristics depends on the soil type and the inherent soil properties that are considered to be stable at the time scale of decades (Vogel et al., 2018). They in turn influence state variables (e.g., water content, biological activity, and temperature) that change very quickly within days. The research challenge for soil sciences is an in-depth exploration of the spatial and temporal dynamics of soil functional characteristics being the basis to derive meaningful indicators for soil functions based on well-defined and observable soil properties. For a recent review on soil quality indicators, see also Bünemann et al. (2018), who make a plea for novel indicators that can account for the multifunctionality of soils.

| RESOURCE USE EFFICIENCY
In the context of soil management, we define resource use efficiency as ratio between benefits generated by agricultural production processes and the amount of resources used. Only resources for which there is competing demand are considered (Di Maio et al., 2017).
Competition is context-specific and can be due to demand exceeding resource limitations or due to conflicts between resource use and other targets set by society (e.g., pesticide application conflicting with biodiversity preservation).
The concept of agricultural resource use efficiency is much older than the concept of soil functions, and few assessments documented in the literature explicitly address the role of soils. However, assessment results often implicitly reflect changes in soil functions due to the paramount role of soils for crop growth.
Within the context of research for sustainable bioeconomies, it will be necessary to determine how soil functions affect resource use efficiencies and additionally, to what degree soil management can increase efficiencies.
The spatial and temporal scope must be clearly defined before efficiency assessments because system boundaries can have strong implications for the final results. Leakage, cascade, and rebound effects should be considered (Lambin & Meyfroidt, 2011). The use of multiyear averages and considering production over several years is necessary to account for interannual yield variability and crop rotation effects (Preissel, Reckling, Schläfke, & Zander, 2015;Zhang et al., 2017). Farm-or field-level resource use efficiencies are of particular interest to practitioners, especially when they affect economic performance. Typical indicators are productivity (Alam, Humphreys, Sarkar, 2016), water use efficiency (Pascual, Villar, & Rufat, 2016;Wei et al., 2016;Zhao et al., 2010), benefit-cost ratio (Alam et al., 2017;Moreau et al., 2012;Rehman, Farrukh Saleem, Safdar, Hussain, & Akhtar, 2011), nutrient use efficiency (nitrogen: Buckley, Wall, Moran, O'Neill, & Murphy, 2016;Gu, Ju, Chang, Ge, & Chang, 2017;and phosphorous: Gerber, Uwizeye, Schulte, Opio, & de Boer, 2014;Zhang et al., 2016), and energy use efficiency (Arodudu, Helming, Voinov, & Wiggering, 2017;Hill, Nelson, Tilman, Polasky, & Tiffany, 2006). Efficiency assessments with landscape or global implications are relevant to policy makers. Indicators include GHG emissions per yield (Khakbazan et al., 2017;Steyn, Franke, van der Waals, & Haverkort, 2016)  In most cases, increased efficiency aligns with targets set by society or policy targets (e.g., European Commission, 2011; General Assembly, 2015). However, being too efficient is also possible. Reducing the labor required to produce an amount of biomass may have negative implications for job provision; nutrient use efficiencies greater than 100% are indicative of nutrient mining by harvest (Scholz & Wellmer, 2015) and maximizing the share of biomass harvested relative to net primary production conflicts with biodiversity targets (Mudgal et al., 2012).  (Boshrabadi, Villano, & Fleming, 2008;Li, Feng, You, & Fan, 2013). A more common, nonparametric approach is data envelopment analysis (DEA; Table 1 Boshrabadi et al., 2008;Li et al., 2013 Note. Separate indicators are determined by the fraction of benefit divided by resource. Overview of studies published between 2008 and 2017 based on the Web of Science Core Collection by applying the search terms "efficiency" (title) and "agriculture," "indicator" (topic). Commonly applied indicators/methods are marked in red. Potschin, 2009;Potschin-Young et al., 2017;TEEB, 2010) (Grossman, 2015;Meylan et al., 2017;Schulte et al., 2014). It is important to clarify these analytical shortcomings to connect the insights and results from a large body of soil function literature to the ecosystem services concept.
Third, despite its very efficient approach to transferring environmental issues to science and society by streamlining social-ecological complexity to one dimension, that is, services for humans, the ecosystem services concept applied to soil management is limited because it does not consider other targets set by society or other valuation perspectives, such as resource use efficiency, equity, or health (Sandifer, Sutton-Grier, & Ward, 2015;Schröter et al., 2017). However, linkages to other valuation systems are possible, and the first studies have been conducted to discover these linkages, for example, as outlined by Bouma (2014) and Keesstra et al. (2016). Our framework addresses this conceptual gap by combining the ecosystem service approach with the resource use efficiency approach (Figure 1) that again can jointly inform governance of soil management and be considered in the design of particular policy instruments. More research is required to fully capture the link between valuation perspectives and governance as driver of soil management such as pioneered in a study by Öhlund, Hammer, and Björklund (2017). Note. Colors highlight services that we consider to be strongly affected by agricultural soil management: primarily via changes in soil functions (red), primarily nonsoil mediated (blue), and both via management and changes in soil functions (purple).

| RESPONSES AND GOVERNANCE FOR SUSTAINABLE SOIL MANAGEMENT
Multiple drivers influence the soil management decisions of farmers, and soil management decisions are also indirectly influenced via the decisions of other actors, particularly consumers. Their decision-making occurs in the context of governance structures, including obligatory, incentive-based and awareness-raising instruments (e.g., fertilizer regulation, agri-environmental measures and common agricultural policy subsidies, and persuasive campaigns). Of all the drivers of soil management, governance structures can be changed through policy, and thus, these instruments can be used to counteract or reinforce drivers in accordance with goals set by society. In this sense, governance is both a driver (current governance structures) and a means of response. Any governance structure is the result of societal processes of negotiation and deliberation among different actor and lobby groups (e.g., farmers, environmentalists, consumer groups, and industries).
The information generated through impact assessment is crucial to crafting appropriate responses; impact assessment informs the (re-)design of governance instruments and institutions that influence other drivers of soil management in such a way as to redirect pressures in line with goals set by society (Figure 1). Currently, soil-related governance is heavily understudied (Juerges & Hansjürgens, 2018); in the European Union, many different policies that influence soils exist, but because of failure to establish a Soil Framework Directive, they are not necessarily consistent with each other and only partially tackle the needs of sustainable soil management (Glaesner, Helming, & de Vries, 2014;Kutter et al., 2011;Paleari, 2017;Vrebos et al., 2017).
The challenge for sustainable soil governance is thus to align the use of soils with targets set by society, building on the results of impact assessments of different soil managements. In addition to the proper understanding of the sequence from soil management ➔soil processes ➔soil functions ➔human well-being (see Sections 5 and 6), governance needs to account for relevant future developments and drivers of soil management (Section 3) and the way that relevant actors behave and make their decisions-the latter is important for anticipating how actors, especially farmers, will respond to new governance structures and institutions. Monetary aspects of soil management are crucial for farmers acting as economic entities, albeit not the only factor. While a large amount of literature exists that deals with individual specific factors influencing farmers' decisionmaking (e.g., Lamarque, Meyfroidt, Nettier, & Lavorel, 2014;Lienhoop & Brouwer, 2015;Senger, Borges, & Machado, 2017), to be effective, governance has to be informed by the full suite of these factors. Much can be learned from the emerging field of behavioral environmental economics (Gsottbauer & van den Bergh, 2011). Furthermore, designing governance structures and institutions for sustainable soil management occurs in an existing institutional setting that needs to be adapted to new political goals. Future governance structures aimed at sustainable soil management should build upon the profound knowledge of existing structures and institutions in the status quo.
Based on the results of impact assessment on how well targets set by society are met under current management conditions, tailor-made governance instruments can then be identified and applied to steer soil management in favor of soil functioning.

| POTENTIAL AND LIMITATIONS
The assessment framework constructed here embeds the ecological sphere of soil process/soil function interactions within the socioeconomic spheres of soil management and its driving forces on the one side, and targets set by society on the other side, with governance linking the two sides. Its application enables a comparative analysis of alternative decision options and thereby supports decision-making towards sustainable implementation of bioeconomy strategies from a production perspective. However, the application of the framework is subject to an improved understanding of the causal chain relationships in each of its steps, including feedback loops, spatio-temporal variations, data needs, and modeling approaches. The particular asset is to use the framework for the design and setup of application studies, to identify the required expertise from different disciplinary backgrounds, and to facilitate their interdisciplinary cooperation. To deal with data limitations, applications would involve a combination of quantitative analysis and qualitative, expert based assessments, for example, through the use of Bayesian belief networks (Gonzalez, Luque, Poggio, Smith, & Gimona, 2016). Stakeholder engagement in the application would improve the adoption of assessment results by land managers and policy (Bünemann et al., 2018).
We constructed the assessment perspective by combining the concepts of ecosystem services with that of resource use efficiency to operationalize the term sustainable intensification in the context of bioeconomy strategies. Sustainable intensification aims for increased production while avoiding environmental damages (BMBF, 2010;BMEL, 2014). However, the term can also be understood as increasing the sum of benefits from ecosystem services (Buckwell et al., 2014;Garnett et al., 2013). This result may be achieved with constant or even lower production, especially in countries that already exhibit low yield gaps, for example, through higher shares of organic farming. Similar concepts are subsumed under the terms ecological intensification (Tittonell, 2014), agroecology (Duru et al., 2015, sustainable extensification (van Grinsven, Erisman, de Vries, & Westhoek, 2015), or smart intensification (Govers, Merckx, van Wesemael, & Van Oost, 2017). Additionally, consumption-side inefficiencies are at least as relevant as production-side inefficiencies. For example, reduction of wastage in food systems may significantly lower pressure on resources, including soils (Alexander et al., 2017). Sustainable intensification in the wider sense does therefore include the adoption of practices along the entire value chain of the food system (Rockström et al., 2017). Although these aspects have not been included in our framework, an extension in this direction is possible. Higher resource use efficiencies are often seen as a win-win solution for increasing economic competitiveness and reducing environmental burdens.
However, most studies fail to include take-back or rebound effects.
Rebound effects occur when higher resource use efficiencies lower production costs and product prices, resulting in changing patterns of production and consumption. This offsets part of the original resource saving and may in extreme cases lead to higher resource use than before. For example, higher irrigation efficiency can lower the price of irrigation, leading to a more widespread adoption of irrigation technologies and higher total water consumption (Pfeiffer & Lin, 2014). While more research is required to allow for quantifications of this effect, it must be considered in impact assessments. A strong collaboration of soil scientists with economists would be most beneficial in this respect.
Although the targets set by society that form the basis of our impact assessment framework, that is, raising efficiencies and maintaining or enhancing environmental services, are well defined, there is no consensus on how to handle trade-offs. Normative decisions are required to derive comprehensive evaluations, and they should ideally be made in cooperative processes involving all relevant stakeholder groups. Participatory elements can be linked to the framework presented here to structure such a process (Bond, Morrison-Saunders, Gunn, Pope, & Retief, 2015). A different study context, for example, in developing countries, can also require different assessment perspectives and targets set by society that are not considered here, including risks to human health or equity issues between societal groups and/or generations. These issues would involve different indicator systems, but they could be implemented within the same assessment framework.

| CONCLUSIONS
Increasing demand for production from soils carries threats to soil functions and increases risk of land degradation. New agricultural management methods provide opportunities and threats for aligning production with soil function maintenance. Assessments are necessary to avoid threats and optimize benefits. The comprehensive framework presented here is well suited for this task. It helps to identify drivers for the adoption of different management methods and pressures that require an impact assessment. It connects management to soil functions. In addition, it offers a systematic approach to connecting changes in soil functions with targets set by society, which in this case were sustainable solutions for bioeconomy strategies. We have integrated the two most common assessment concepts for this context in countries with low yield gap agricultures, namely, ecosystem services and resource use efficiency. The framework can also be adapted to other assessment concepts. Although efficiency gains from improved management are usually enabled through soil processes and soil functions, this link is rarely made explicit in scientific studies.
Efforts must be made to better understand the contribution of soil functions to resource use efficiencies. From bioeconomy perspective, resource use efficiency is highly relevant for decision-making. However, it does not necessarily lead to maintaining soil functions. Regarding ecosystem services, there is still no conceptual consensus among scientists on the links from soil management and functions to ecosystem services. More research is required to improve systematic assessments of the effects of soil functions on ecosystem services, systematically assess leakage and rebound effects, consider trade-offs between targets set by society and include other targets, such as human health or equity. All this requires interdisciplinary cooperation.
The proposed assessment framework helps to structure this interdisciplinary research.