Modeling of soil compaction impact on key soil functions: results

Towards climate-smart sustainable management of agricultural soils SoilCompaC Mapping and alleviating soil compaction in a climate change context Deliverable 3.3 Modeling of soil compaction impact on key soil functions: results Due date of deliverable: M57 (October 2024) Actual submission date: M58 (November 2024) Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 2 GENERAL DATA Grant Agreement: 862695 Project acronym: SoilCompaC Project title: Mapping and alleviating soil compaction in a climate change context Project website: SoilCompaC Start date of the project: November 1st, 2021 Project duration: 36 months Project coordinator: Mathieu Lamandé DELIVERABLE NUMBER: D3.3 DELIVERABLE TITLE: Modeling soil compaction impact on key soil functions: results DELIVERABLE TYPE: Report DISSEMINATION LEVEL: Int = Internal Working Document WORK PACKAGE N: WP3 WORK PACKAGE TITLE: Effect of soil compaction on key soil functions in a changing climate DELIVERABLE LEADER: AGS AUTHORS: Alejandro Romero-Ruiz, Thomas Keller, Lorena Chagas Torres, Michael Kuhwald, Mathieu Lamandé SOILCOMPACCONTRIBUTORS: ACKNOWLEDGEMENTS: Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 3 About this deliverable This deliverable is a manuscript in preparation not in its final version. Authors suggest looking for the accepted paper in spring 2025 for a final version. Title Modelling long-term effects of soil compaction on crop yield, soil organic carbon stocks, and nitrogen losses from soil Authors A. Romero-Ruiz, L.C. Torres, M. Lamandé, M. Kuhwald, T. Keller 1 Introduction Soil compaction is a regarded as a major environmental hazard globally, affecting soils in all ecosystems and climates (Chamen et al., 2015; Hamza & Anderson, 2005; Nawaz et al., 2013). Almost three decades ago, Oldeman (1992) estimated that 68 Mha of agricultural lands are compacted worldwide, and two decades ago Steinfeld et al. (2006) estimated that 20% of the world’s grassland are degraded by overgrazing and compaction. Soil compaction is caused by construction, military, agricultural and forestry operations involving heavy vehicles and due to grazing animals trampling on the soil under mechanically vulnerable conditions. The size and weight of vehicles and the intensity of grazing have continued to increase in recent years in response to food, and infrastructure demands of a growing population (maybe add some references). This, combined with the slow recovery rates of soil structure following compaction, suggests that soil compaction levels and extent have increased (Keller and Or, 2022). Compaction induces changes in soil pore structure, often implying reduction and disruption of soil macroporosity that lead to significant reductions in soil water and gas transport properties (Berisso et al., 2012) and increases soil mechanical resistance to root penetration (Colombi et al., 2018), thereby impacting soil physical and biological processes such as water infiltration, oxygen diffusion, root growth (Rabot et al., 2018), and carbon and nutrient cycling (Tubeileh et al., 2003). This combination of factors typically produces negative Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 4 impacts on soil processes that are central for soil productivity, water regulation and purification, and climate regulation. The reduction in soil water and gas transport properties (e.g., hydraulic conductivity, air permeability, gas diffusivity) may promote anaerobic conditions within the soil, enhancing nitrous oxide emissions (Pulido-Moncada, 2023). In addition, root penetration is hampered due to an increase in mechanical resistance of the soil, which limits nutrient and water uptake in the soil profile, which may promote nutrient leaching. Such limiting plant growing conditions produce a decrease in productivity, manifested as decreases in plants biomass above and below ground (roots, leaves and shoot), which can further result in a reduction of carbon inputs into the soil (Pulido-Moncada et al., 2022; Soares et al., 2015). Considering the processes mentioned above, the main risks associated with compaction include: (1) crop yield decline, (2) losses of soil organic carbon, and (3) nitrogen losses through increased nitrous oxide emissions and nitrate leaching. Despite acknowledging these risks and general knowledge of the physical, chemical and biological mechanisms responsible for them, understanding how different processes interact and quantifying the long-term effects on soil functions and the environmental and economic consequences of soil compaction remain challenging. This is associated to (I) a lack of solid data on the spatial and temporal extent of harmful soil compaction (Håkansson & Lipiec, 2000), (II) the technical difficulties of measuring and quantifying long-term impacts of soil compaction in a systematic way (Keller et al., 2021) and, (III) the complexity in understanding how the different compaction-impacted soil processes interact and what the relative contributions of the compaction level, soil texture, crop, and climate are. Mechanistic modelling of soil compaction and its impacts on soil physical properties may be combined with existing agroescosystem models for deriving new insights to understand the impacts of compaction on soil (dis)services in a systematic way. In this work, we aim at systematically assessing the long-term impact of soil compaction on crop yield, soil organic carbon stocks, nitrous oxide emissions, and nitrate leaching. For this, we established the following objectives: Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 5 1. To report literature evidence on key aspects and trends of how compaction impacts crop yield, soil organic carbon stocks, nitrous oxide emissions, and nitrate leaching. 2. To establish a modelling framework that systematically incorporates soil compaction impacts in soil structure in an agroecosystem modelling approach. 3. To compare the model simulations with literature-inferred trends. 4. To dissect modelling results to seek a better process-based understanding of the impacts of soil compaction on soil (dis)services across diverse soil textures, climatic conditions and recovery rates. 2. Evidence of soil compaction effects on crop yield, soil organic carbon stocks, and nitrogen losses from soil: A data compilation In order to compile empirical data from the literature and identify trends in the effects of soil compaction on crop yield, soil organic carbon stocks, nitrous oxide emissions, and nitrate leaching for comparison with modelling results, we conducted four different searches in WebofScience for publications that contained the term “soil compaction” combined with either the terms “yield”, “carbon stocks”, “nitrous oxide emissions”, and “nitrate leaching”. We selected studies comparing these variables in field conditions, where compaction was induced by agricultural operations (wheeling or trampling), and both in cropland or grassland (but not in forests). All the selected studies measured at least one of these variables of interest for both one (or various) compaction treatment(s) and a non-compacted control treatment. The combination of the elements: (i) compaction-induced damage, (ii) soil texture and (iii) climate is in general different for all studies. This may increase variability in the studied trends, yet, the number of studies is limited which makes it unfeasible to divide them into groups according to soils, climate and compaction level A summary of the compiled studies can be found in Tables S1-S4. The number of studies that address the impact of compaction in the variables of interest is limited due to technical experimental difficulties of designing, maintaining and monitoring field experiments focused on soil structure dynamics and the related elevated economic costs (Keller et al., 2018; Vanino et al., 2023). Such difficulties are further exacerbated for Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 6 long-term experiments, limiting our ability to understand long-term impacts of soil compaction in the variables of interest. Because of these limitations and the resulting gap in literature evidence, we opted for grouping all gathered studies for a given variable despite presenting different compaction levels (produced by different numbers of wheeling events, vehicle weights, and livestock densities, and for different soil textures and moisture conditions), soil textures, climates, length of experiments, etc. To report this in a meaningful way, we plotted the variables of interest from all selected studies in relative terms (compacted/non-compacted), and as a function of time after the compaction event. Our literature search comprises 15, 6, 16 and 5 studies for crop yield, soil organic carbon stocks, nitrous oxide emissions, and nitrate leaching, respectively. From these studies we found 7, 81, 41, and 20 number of paired comparisons (compacted vs non-compacted) for crop yield, soil organic carbon stocks, nitrous oxide emissions, and nitrate leaching, respectively. Most studies correspond to temperate regions and loamy soils as summarised in Table S4. Most studies reported relative crop yields lower than 1, ranging from 0.58 to 1.1, after compaction. They generally measure yields on the season following compaction and are therefore presented here as being measured one year after compaction (Figure 1a). Relative soil organic carbon stocks are reported as a function of years after compaction. Compaction does not have a consistent effect on soil organic carbon stocks, the relative values (i.e., ratio of compacted to uncompacted treatment) range from 0.7 to 1.4 (Figure 1b). The relative nitrous oxide emissions are typically higher than 1, indicating an increase in N2O emissions due to compaction (Figure 1c). They are reported as cumulative emissions during a given period of time after compaction. Emissions increase due to compaction, yet the increases are highly variable and do not seem to follow a trend as a function of the duration of the monitoring period after compaction (Figure 1c). The relative nitrate leaching is reported as cumulative nitrate leaching as a function of time after compaction. Data are typically reported for one-year long periods after compaction and do no present any clear trend, with values being both higher and lower than 1 (Figure 1d). Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 7 Figure 1 Literature data comparing impacts of soil compaction on (a) crop yield , (b) soil organic carbon stocks, (c) nitrous oxide emissions, and (d) nitrate leaching. The figure reports relative changes (compacted vs non-compacted), the control line (compaction/no compaction = 1; black line), and the mean value of the data (magenta line) and the range (grey, standard deviation of the mean of the relative values). Note the different scales on the x-axes are in years after compaction for crop yield and soil organic carbon stocks, and in months after compaction for nitrous oxide emissions and nitrate leaching. 3 Soil structure-based modelling framework for … In this work, we present the model “CompaC” , which is an expansion of the model presented in Romero-Ruiz et. al., (2024) used to evaluate grazing-induced compaction. The CompaC modelling framework simulates wheeling induced soil compaction effects on soil processes related to crop growth, Cand N-cycling by coupling (i) a soil compaction model (Romero-Ruiz et al., 2023) and (ii) an agroecosystem model (The Rothamsted Landscape Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 8 Model, RLM; Coleman et. al., 2018). By systematically inducing information about soil compaction-induced changes in soil physical properties, this framework allows quantifying impacts of compaction on soil states and (dis)services (Figure 1). Here, we describe the elements of these models that are most relevant for the present study. For further details we refer to Romero-Ruiz et al., (2023) and Coleman et al., (2018). Figure 2 Schematic representation on the modelling approach used in this study. A soil compaction model is first used to systematically model changes in soil structure induced by a compaction event. The modelled changes in soil structure are then infused in an agroecosystem model. By doing this, the modelling framework is able to simulate compaction-induced changes in soil state variable and their impacts on soil processes and functions. Model inputs and outputs are presented in orange and green blocks, respectively. 3.1 Soil compaction model Integrating modelled impacts of compaction in an agroecosystem model, which typically have a simplified representation of soil structure, requires a matching simplified description of soil Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 9 compaction impacts on soil structure. The soil compaction model by (Romero-Ruiz et al., 2023) was developed for this purpose. In this work, this model was used to systematically calculate the corresponding temporal dynamics (different compaction events and recovery at different rates) of three key soil properties impacted by compaction: soil bulk density (𝑑𝑑), macroporosity (𝑤𝑤𝑚𝑚𝑚𝑚𝑚𝑚) and saturated hydraulic conductivity (𝐾𝐾𝑠𝑠𝑚𝑚𝑠𝑠). A compaction event produces an irreversible deformation, 𝜖𝜖𝑣𝑣, which is modelled using information about the initial (prior to compaction) strain 𝜖𝜖0, the axial load and duration of stress application, and the soil rheological properties as (Ghezzehei & Or, 2001): ϵ𝑣𝑣(𝑡𝑡)=�ϵ𝐵𝐵 2𝑆𝑆𝑠𝑠𝑚𝑚(𝑡𝑡)𝑁𝑁ν �1−𝑐𝑐𝑐𝑐𝑐𝑐(ω𝑡𝑡)�+ϵ0 2�1 2,(1) where t is the time, 𝜔𝜔 is the angular frequency, ϵ𝐵𝐵 comprises information of the soil rheological properties and the characteristics of the compaction event (e.g., stress induced by a vehicle or animal), 𝑆𝑆𝑠𝑠𝑚𝑚=θ𝑠𝑠𝑚𝑚/ϕ𝑠𝑠𝑚𝑚 is the degree of water saturation in the soil matrix, where θsm is the water content in the soil matrix, and 𝑁𝑁ν is an empirical exponent. The compaction induced strain is then used to calculate 𝑑𝑑, 𝑤𝑤𝑚𝑚𝑚𝑚𝑚𝑚 and 𝐾𝐾𝑠𝑠𝑚𝑚𝑠𝑠. Further details on the soil compaction model, including a discussion of its limitations, can be found in (Romero-Ruiz et al., 2023). The soil compaction model computes recovery of soil properties as a function of time following a compaction event .This is done by simulating a recovery of the viscous strain with an empirical exponential function given by: ϵ𝑣𝑣=ϵ0−(ϵ0−ϵ𝑖𝑖)𝑒𝑒−𝑑𝑑𝑟𝑟/λ𝑡𝑡𝑟𝑟,(2) where ϵ𝑖𝑖 is the soil strain, representing the strain resulting after the compaction event, dr is the number of days after the compaction event, and λ𝑠𝑠𝑡𝑡 is the recovery rate. 3.2. The agroecosystem model The RLM discretizes the soil into three layers with model-defined interfaces at 0.23 m and 0.46 m (final layer ends at 1 m). This is done for simplicity to represent a topsoil, a plow pan layer and a subsoil, as explained by Coleman et. al., (2018). In the following, we describe how water flow, nitrogen dynamics, crop growth and carbon cycling are modelled in this framework. 3.2.1 Soil water dynamics and water retention Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 16 model was run with compacted inferred properties for all the simulated years. In the base case scenario, we considered 30 levels of compaction obtained by varying the normalized vehicle weight 𝛿𝛿𝑚𝑚 from 0 to 1.5 with 0.05 increment (corresponding to wheel loads from 0 to x, with increments of x). Therefore, the base case scenario comprises 30 simulations for 30 different levels of compaction, each with a different set of modelled bulk densities, macroporosities and saturated hydraulic conductivities. Although it is expected that variations of soil texture, precipitation and temperature regimes impact the level of compaction and result in a different 𝜀𝜀vbase, we considered 𝜀𝜀vbase to be the same in all modelled scenarios (the base case and all other cases described in the following sections). This allows analysing the impact of soil texture, weather variables, recovery rates and climate on crop yield, soil organic carbon stocks, nitrous oxide emissions and nitrate leaching according to the predicted changes in bulk density, macroporosity and saturated hydraulic conductivity. For simplicity, all the simulations presented in this study were done for one crop only (wheat, the most important crop in Europe). 3.4.2 Changes in soil texture: texture-scenarios To investigate the role of soil texture, we considered three modelling scenarios with different textures corresponding to clay contents of 15%, 25% (base case, see previous section) and 35%, covering with this the range of the most dominant clay contents found in the literature (Tables S1-S4) and most dominant textural classes globally (reference to soil grids). The impacts of soil texture in the modelling framework (and outputs) are effectively incorporated by considering clay-content dependent water retention properties based on the data compilation presented by Carsel & Parrish (1988). The van Genuchten model parameters were then obtained by linear regression for θ𝑡𝑡, θ𝑠𝑠, 𝑙𝑙𝑐𝑐𝑔𝑔(α) and 𝑙𝑙𝑐𝑐𝑔𝑔(𝑘𝑘𝑠𝑠) as a function of clay content, and by fitting an exponential function between n and clay content (Figure 4). Note that 𝑘𝑘𝑠𝑠 here refers to the saturated hydraulic conductivity of the soil matrix , which differs from 𝐾𝐾𝑠𝑠𝑚𝑚𝑠𝑠 predicted by the soil compaction model, which is the saturated hydraulic conductivity of the dual-domain soil (considering the soil matrix and macropores) as explained in Section 3.2. The texture simulations scenarios were done using the same weather data and considering the 30 levels of compaction as the base case scenario (see also Table 1). Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 17 Figure 4 Means of the van Genuchten model parameters and Ks as a function of the mean clay content (crosses), as reported by Carsel & Parrish (1988), with their respective regression curves (solid lines). The figure shows the (a) residual water content, (b) saturated water content, the empirical parameters (c) alpha and (d) n, and (e) the saturated hydraulic conductivity of the soil matrix. 3.4.3 Changes in mean annual precipitation: Precipitation-scenarios Disentangling effects of soil compaction in different climates is challenging as precipitation and temperature regimes vary simultaneously for different climates. We opted for running model simulations varying the precipitation and temperature independently. As described in Section 3.3.1, the climate data used in this study for the base case scenarios were obtained from weather projections reported in… and correspond to a mean annual precipitation of approximately 800 mm. We considered two more scenarios with higher or lower mean annual precipitation, namely 400 mm and 1200 mm, which were obtained by multiplying daily precipitation by 0.5 and 1.5, respectively. The rest of the weather variables (e.g., solar radiation) were kept the same. The precipitation scenarios were made using the soil with 25% clay content and with the same compaction scenarios considered in the base case (see Table 1). Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 18 3.4.4 Changes in mean annual temperature: Temperature-scenarios Similar to the previous section, we ran model simulations varying the mean annual temperature independently. In the base case, the mean annual temperature was calculated to be approximately 10 oC. We considered two more with lower or higher mean annual temperatures:, namely 7.5 oC and 12.5 oC. These were obtained by subtracting and adding 2.5 oC to all the daily temperatures, respectively. The rest of the weather variables were kept the same. The temperature scenarios were made using the soil with 25% clay content and with the same compaction scenarios considered as in the base case (see Table 1). 3.4.5 Changes in soil structure recovery rates: Recovery-scenarios To investigate the impacts of recovery of soil structure in modelled outputs, we considered three recovery scenarios (1) no-recovery (base case, corresponding to λ𝑠𝑠𝑡𝑡=𝑚𝑚𝑛𝑛𝑓𝑓 ), (2) full recovery after five years (corresponding to a recovery rate of λ𝑠𝑠𝑡𝑡= 1.6) and (3) full recovery after ten years (corresponding to a recovery rate λ𝑠𝑠𝑡𝑡= 2.3). In the modelled recoveryscenarios, soil structure recovery strictly refers to the recovery of bulk density, macroporosity and saturated hydraulic conductivity. These structural properties are considered to recover at the same recovery rate λ𝑠𝑠𝑡𝑡. Time for “full-recovery” after compaction was defined as time needed until the relative difference between preand postcompaction bulk density was lower than 1%. No re-compaction events (i.e., repeated wheeling events) were considered in any scenarios. Table 1. Summary of all modelling scenarios of this study. All the modelled scenarios comprise 30 compaction models. The x mark the different characteristics of the simulations regarding clay content, mean annual precipitation, mean annual temperature, and time until full recovery. Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 19 Compacti on Clay content Mean yearly precipitation ( mm/y) Mean yearly temperature (oC) Time for full recovery (years) Section 30 levels 1 5 2 5 3 5 80 0 40 0 120 0 10 7.5 12.5 No 5 10 Section 4.1 Base case (BC) x x x x x Clayscenario (4.2) Case 1 x x x x x Case 2 (BC) x x x x x Case 3 x x x x x Precipitatio n -scenario (4.3) Case 1 (BC) x x x x x Case 2 x x X X x Case 3 x x x x x Temperatur e-scenario ( 4.4 Case 1 (BC) x x X X x Case 2 x x X x x Case 3 x x x X x Recoveryscenario ( 4.4 Case 1 (BC) x x X x x Case 2 X x X x x Case 3 x x x x x 4 Results In this section, we present modelling results corresponding to the scenarios presented in Table 1. To be consistent with literature data reported in Figure 1, we simulated the following variables: (1) yearly relative crop yield, (2) yearly relative organic carbon stocks in the topsoil, (3) cumulative nitrous oxide emissions, and (4) cumulative nitrate leaching, all as a function of time after compaction. 4.1 General trends in the base case simulations We first compared simulations corresponding to the base case with the ranges found in the literature. The modelled compaction-induced changes in environmental variables and crop yields agreed with the ranges observed in the literature (see results for 𝛿𝛿𝑚𝑚=1 in Figure 5). Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 20 Crop yields are predicted to decrease with soil compaction for most years after compaction (Figure 5a). The yearly relative change in yield is predicted to fluctuate around a mean value (0.95). A decrease in soil organic carbon stocks due to compaction of about 3% after 50 years was predicted by the model (Figure 5b). The model predicted an increase of cumulative nitrous oxide emissions that remain constant during the first few months after compaction (Figure 5c). The relative increase was about 3.5 times for the base case scenario. The nitrate cumulative leaching was predicted to decrease by 30% by compaction in the base case scenario (Figure 5d). Figure 5 Modelling results comparing relative impacts of soil compaction on (a) crop yield , (b) soil organic carbon stocks, (c) nitrous oxide emissions, and (d) nitrate leaching. These modelling results correspond to soil structure properties calibrated using 𝛿𝛿𝑚𝑚=1. The control line (no compaction = 1), the mean value of the literature data and the shaded area corresponds to the standard deviation range of the literature data. Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 21 4.2 Compaction impacts for different soil textures-scenarios Changes in crop yield and environmental variables were predicted for soil textures characterized by having 15%, 25% and 35% clay content as a function of time for one compaction level (corresponding to 𝛿𝛿𝑚𝑚=1, Figure 6), and as a function of clay content for all compaction levels and for a time period of 80 years following compaction (Figure 7). When considering the same compaction level (𝛿𝛿𝑚𝑚=1), the model predicted similar trends in changes in soil organic carbon stocks for all soil textures (Figure 6b). Crop yields are predicted to decrease with compaction for most years in all clay contents, with the relative yield values fluctuating around a constant value (Figure 6a). The modelled ranges of relative yields were similar for all clay contents (Figure 7a). Soil organic carbon stocks showed a general decreasing trend due to compaction, with some small increases for some periods of time. Similar to crop yields, the ranges of relative carbon stocks were similar for all clay contents (Figure 7b). The trends in relative nitrous oxide emissions were similar for all clay contents, and simulated to be constant for the first few months after compaction (Figure 6c). Relative increase in nitrous oxide emissions was inversely proportional to clay content (Figure 7c). The cumulative nitrate leaching was predicted to decrease with soil compaction for all clay contents in the first few months after compaction (Figure 6d). Yet, the model predicted increases in nitrate leaching for some compaction levels and for some years after compaction (Figure 7d). Adverse soil compaction effects in crop yield and carbon stocks were predicted to be similar (in relative terms) for all clay contents, and to increase for increasing compaction levels (Figures 7a,7b and S2). In contrast, the relative increases in nitrous oxide emissions due to compaction were lower for higher clay contents (Figure 7c), and relative nitrate leaching decreased for increasing clay contents indicating that compaction hindered leaching (Figure 7d, and S2). Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 22 Figure 6 Modelling results of texture-scenarios comparing relative impacts of soil compaction on (a) crop yield , (b) soil organic carbon stocks, (c) nitrous oxide emissions, and (d) nitrate leaching. These modelling results correspond to soil structure properties calibrated using 𝛿𝛿𝑚𝑚=1 and for different clay content. The control line (no compaction = 1) is represented by the black line, the mean value of the literature data are shown by the magenta lines, and the shaded area corresponds to the standard deviation range of the literature data. Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 23 Figure 7 Modelling results of texture-scenarios and literature data comparing relative impacts of soil compaction on (a) crop yield , (b) soil organic carbon stocks, (c) yearly nitrous oxide emissions, and (d) yearly nitrate leaching as a function of clay content. Each modelled curve corresponds to one compaction level in a given year for the first 80 years after compaction. Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 24 Figure 8 Modelling results of texture-scenarios comparing relative impacts of soil compaction on crossplot of (a) crop yield and carbon stocks, (b) yield and yearly nitrate leaching, and (c) yearly nitrous oxide emissions and yearly nitrate leaching. Each modelled point corresponds to all levels of compaction on a given year for the first 80 years after compaction. 4.3 Compaction impacts for different precipitation-scenarios Changes in crop yield and environmental variables were predicted for mean annual precipitation of 400, 800 and 1200 mm/y as a function of time for one compaction level (see results for 𝛿𝛿𝑚𝑚=1, Figure 8), and as a function of mean annual precipitation for all compacted levels and for 80 years after compaction (Figure 9). The yield is predicted to decrease with compaction for all precipitation regimes (Figure 8a), and the relative yield fluctuates around the same value. Yet, fluctuations are higher for lower mean annual precipitation predicting Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 25 also more years when yield is higher for compacted soils than for non-compacted (Figures 8a and 9a). The model predicted a similar trend in relative carbon stocks for all precipitation regimes (Figure 8a). Soil organic carbon stocks generally decreased due to compaction, with some small fluctuations. The relative carbon stocks are predicted to have similar ranges for all mean yearly precipitation regimes (Figure 8b). An increase in nitrous oxide emissions is predicted in all precipitation regimes for the first few months after compaction, and was particularly higher for high precipitation (Figure 8c). Yet, the ranges in relative yearly nitrous oxide emissions were similar for all precipitation regimes (Figure 9c). The cumulative nitrate leaching is predicted to decrease with soil compaction for the precipitation regime of 800 and 1200 mm/y but remain the same as in the non-compacted simulations for the 400 mm/y regimes(Figure 8d). Adverse soil compaction effects in yield and carbon stocks are predicted to be similar for all precipitation regimes, but the yield is more variable for the 400 mm/y regime (Figure 9a, 9b, and S3). The precipitation regime predicted the relative increase in nitrous oxide emissions due to compaction in a have similar ranges (Figure 9c and S3). The nitrate leaching is decreased with increased mean annual precipitation (Figure 9d and S3). Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 32 5.1 Literature data Soil structure changes induced by soil compaction change soil mechanical and hydraulic properties across the soil profile which has a strong effect on soil processes involving water flow, gas transport, root growth and water uptake, nutrient and carbon cycling. These interacting processes are crucial for soil functions such as climate regulation, water filtering, crop production and carbon sequestration. While there is literature evidence of strong effects of soil compaction on carbons stocks (that may increase or decrease), nitrous oxide emissions, yield and nitrate leaching, it is unclear how these variables interact and how they are affected by different climate conditions and soil textures. Most of these data correspond to temperate climates and silt loamy textures (Table S1) and measured in relatively short periods. Driven by the limited number of conditions in which data have been collected, it is difficult to establish relationships with covariates such as soil type and climate. Therefore, the modelling approach presented here represents a valuable tool to understand these complex interactions, despite the absence of experimental for extensive validation. The limited literature evidence currently suggests that (1) crop yield consistently decreases with soil compaction, (2) carbon stocks may decrease or increase due to soil compaction, (3) nitrous oxide emissions increase due to compaction but the relative increase as a function of time is unclear, and (4) nitrate leaching may decrease or increase due to soil compaction. In general, this raises the following questions: • What controls the decrease in yield due to compaction? • Under which conditions are carbon stocks increased or decreased? • What controls the level and trend of increases of nitrous oxide emissions due to compaction? • Under which conditions is nitrate leaching increased or decreased?. Providing integral answer for these questions is very challenging and outside the scope of this work. Yet, the modelling approach developed in this study provides insights that help answering these questions by providing quantitative predictions of the variables of interest in compacted systems that can be analysed as a function of the processes that control them and are (or are not) considered by the model. Furthermore, by modelling various scenarios Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 33 in a controlled way, the model helps clarifies how different soil textures and climate conditions influence agricultural and environmental variables for compacted soils. 5.2 Decreasing yield due to compaction The model seems to do a fair job on predicting decline in yield with compaction that is predominantly reported in literature, although some studies report increases in yield in response to compaction (Håkansson et al., 1987, Liu et al., 2022). The literature data suggests that the compaction-induced increase in yields occur more predominantly for low clay contents, which is replicated by the modelling results (Figure 7). The slight increase in yield with compaction may, in general, be explained by considering that a certain degree of soil compaction can offer better mechanical support for crop roots to proliferate in the soil (Liu et al., 2022). However, our modelling results suggest that compacted soils may occasionally show an increase in yield for certain years (Figure 7a) even when mechanical support for root growth is not considered in the model. Instead, the modelled increases in crop yield originate from fluctuations in annual temperature and precipitation for certain years. This may mask negative effects of compaction in yield for such years, and yet the modelling results suggest that accumulated effect will result in losses in crop yield when several years are considered even for high recovery rates (Figures S2-S5a and 13a). 5.3 Carbon stocks The model predicted a decrease in carbon stocks for all soil textures, climates, recovery rates and compaction levels (Figures 6b-13b). This could only explain part of the literature observations, as the modelling results did not predict increases in soil carbon stocks of the topsoil as observed in the literature. The model predicted decreases in soil carbon stocks as a result of accumulated decreases in carbon inputs by plants to the soil. This was produced by the reductions in hydraulic conductivity, that ultimately produced a persisting increased water filled pore space in the soil (see Figure S1) that further limited the biomass formation through the transpiration reduction factor (Equation 5, Figure S9). The decrease in biomass production rate hence produced a decrease yield, shoot, and root biomass that are later used as carbon inputs (Section 3). The modelled effect was similar for different clay contents Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 34 (Figures 6b and 7b), for slight variations in mean annual precipitation (Figures 8b and 9b), and for small variations of temperature (Figures 10b and 11b). Similarly, the rapid recover of soil hydraulic properties was predicted to maintain soil organic carbon stocks in compacted soils similar as those of non-compacted (Figure 12).This can be attributed to the concurrent rapid recovery of the water dynamics (Figure S1f). Despite not capturing all scenarios found in the literature, the model may offer some insights about the elements of compacted soils that are responsible for the observed increased accumulations of carbon in some situations. For example, literature reports (and consequently this study) of carbon stocks typically focus on the topsoil (e.g., Dupla et al., 1993) where carbon may accumulate in compacted soils due to the lateral grow of crop roots promoted by increased penetration resistance of the plow pan resulting from tilling operations (Or et al., 2021). Thanks to this, soil organic carbon stocks may be higher in the topsoil for compacted soils but reduced across the soil profile or viceversa (see Figure S6). Our modelling framework does not consider such lateral (or in general ‘less resistance’) root and hence it is unable to predict such depth dependent dynamics as shown in Figure S5. In addition, the shoot to root ratio biomass is assumed to be constant. This assumption may not valid in the case of compacted soils where plants may (have to) invest more into roots under adverse conditions such as compaction, and leaving the below-ground carbon input is potentially unaffected. Similarly, roots may excrete more mucilage and may result in increased slaughtering of root-cells in compacted soils in order to decrease friction, which could lead to increase soil carbon inputs. Mineralization of soil organic carbon is another process that may be hindered by compaction due to the reduced aeration by decreased gas transport properties (Asady and Smucky, 1989). This reduction may further benefit preserving soil organic carbon pools in compacted soils in comparison with non-compacted soils, and compaction may further create physical protection for soil carbon in the soil (Marques Monroe et al, 2021). Finally, certain managements practices, that are not considered in this model, such as the addition of carbon amendments and tillage operations may impact the dynamics of carbon stock differently in compacted and non-compacted soils. 5.4 Increase in nitrous oxide emissions Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 35 The strong impact of soil compaction on soil water dynamics is believed to be the most dominant factor controlling nitrous oxide emissions from compacted soils . Denitrification by anaerobic microbial activity is naturally very sensitive to compaction-induced increases in water filled pore space due to the simultaneously inflicted reductions of pore space and water and gas transport properties (Pulido-Moncada, 2021) that result in slow water movement and low oxygen diffusion. The modelling framework used in this study is centred in representing soil compaction impacts on soil hydraulic properties and water dynamics. In consequence, the modelling results agreed with most literature reports (Figure 3), predicting an increase in nitrous oxide emission for all soil textures, climates, recovery rates and compaction levels (Figures 6c-13c). Some insights that can therefore be derived from the modelling results. Relative nitrous oxide emissions are inversely related to the clay content (Figures 6), yet, the model predicted accumulated losses of nitrous oxide emissions to be positively correlated to clay content (Figure S9). This is because the rate of change of the denitrification factor that depends in the water filled pore space ( derivative of Equation 14) is higher for lower water content ranges occurring most often when for lower clay content. The slight variations in precipitation and temperature regimes did not have a strong effect in nitrous oxide emissions, and both recovery regimes produce a clear reduction of the relative nitrous oxide emissions (Figure 12c) although there is inevitably accumulated losses irrespective of the recovery rate (Figure 13b). In general, the model seems to successfully predict the ranges and trends compaction-induced increase in nitrous oxide emissions as overwhelmingly reported in literature (Figure 3) and as a function of clay content (Figure 7b). Despite the relatively good agreement between the modelled nitrous oxide emissions and the literature evidence for compacted soils, these results should be carefully taken as not all the processes involved in nitrous oxide emissions are considered by the model, such as those related to the demand of oxygen in soil organic matter decomposition (that similarly creates anaerobic conditions) or the changes in microbial activity ( due to compaction induced changes in community sizes and diversity). 5.5 Nitrate leaching increase/decrease Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 36 Similar to the case of nitrous oxide emissions, the strong impact of soil compaction on soil water dynamics is believed to be the most dominant factor controlling nitrate leaching from compacted soils (Yi et al., 2022). The reduction of both soil water transport properties may prevent leaching, and it is not clear if limited root proliferation volumes due to increase penetration resistances can facilitate or reduce nitrate leaching (Strock et al., 2021). The modelling framework used in this work predicts a predominantly a decrease in nitrate leaching for textures with high clay content (25% and 35%), and irrespective of the climates, recovery rates and compaction levels (Figures 6d-13d). Instead, a similar number of years with increase and decrease in leaching are predicted by the model for the clay content of 15% (Figure 7d), which further lead to a balance between the accumulated leaching predicted for compacted and non-compacted soils (Figure S9d). Although the nitrate leaching was predicted to be generally reduced by compaction, the predicted increases in nitrate leaching for low clay contents are consistent with literature reports (Figure 7d). Unlike all other variables considered here, variable precipitation and temperature regimes have clear effects in nitrate leaching (Figures 9d and 11d). The decrease in yearly nitrate leaching is higher for lower temperatures and higher precipitation regimes (Figures 9d and 11d). Similar to the other variables, the nitrate leaching may increase due to compaction to the climate conditions of a particular year, but the modelling results predicted the accumulated losses to be always higher for non-compacted soils in the long term (Figure S2-S4) even when recovery is considered (Figure 13). The good agreement between modelling results and literature reports (Figure 7d) suggest that even with simplifications the model does capture relatively well soil water flow and the difference in it between compacted and not compacted soils. which is the main process responsible for nitrate leaching. Yet, literature data remains limited and the modelling results must be, again, taken carefully. The model does not capture certain properties, processes and/or scenarios that are impacted by compaction and relevant for nitrate leaching. Interestingly, these processes are also related to those influencing carbon cycling. For example, soil compaction may promote lateral grow of crop roots making nutrients at deeper depths inaccessible and prone to leaching . Also, leaching may be increased by preferential flow occurring in isolated macropores created by mesofauna in compacted soils Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 37 that are inaccessible to soil roots, especially for fertilization events. The model suggests that matrix water flow is the dominant process controlling nitrate leaching and the relative effects of the processes listed above are small. 5.3 Limitations and outlook The work presented here has several shortcomings that can be summarized as: (1) limitations on the modelling physical soil deformation due to compaction and how these impacts soil properties, (2) simplifications (or lack of consideration) on how certain soil physical, chemical and biological processes are modelled, and (3) the high uncertainty on the functions used to obtain water retention and transport properties as a function of texture, and (4) the lack of data to validate modelling predictions. This is not exclusive of this modelling work but a limitation of all agroecosystem modelling efforts (Vereecken et al., 2016). Moreover, our work presents an advance on explicitly representing management induced soil structural changes in agroecosystem models which is highly relevant yet rarely included (Jarvis et al, 2024; Romero-Ruiz et al., 2024; Koenig et al, 2023). Overcoming the abovementioned difficulties in a detailed way is very challenging and outside the scope of this work. Instead, the modelling results presented in this work offer a mechanistic explanation on how processes that are currently considered in agroecosystem modelling are impacted by compaction depending on soil texture, climate and recovery rates, and the resulting effect on environmental variables and yield. The model successfully reproduced trends of relative yields, nitrous oxide emissions, and nitrate leaching and partially successfully reproduced trends on relative carbon stocks. Future modelling efforts will be advocated at better representing soil compaction for specific pedoclimatic conditions where detailed soil, weather and management data are available in order to better predict the different environmental variables shown here and to further calibration of the model properties for predictions of absolute values. By doing this, it may be possible to calculate the impacts of compaction on these variables and their net environmental and economic cost, how prevention or remediation practices can help to ameliorate these negative impacts, and what would be the impact of changing climates in these variables. Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 38 6 Conclusions In this work, we used a mechanistic modelling framework to assess the impacts and trends of soil compaction in carbon stocks, nitrous oxide emissions, yield and nitrate leaching for different pedoclimatic conditions. Literature evidence suggests that soil compaction: (1) can either increase or decrease carbon stocks in the soil, (2) produces a constant increase of nitrous oxide emissions, (3) produces a constant decrease in yield and (4) can either increase or decrease nitrate leaching. The modelling results reproduced most of the trends observed in the literature. This suggests that soil water dynamics, a process that is central in this modelling framework and in how soil compaction is represented, maintains a prominent role on soil (dis)services. The retention and transport properties of the soil matrix and the overall total pore space dynamics were central and variations can be attributed to changes in temperature and especially precipitation water inputs. Overall, the model predictions suggest that long-term effects of soil compaction are environmentally negative even if there are some years where the effects are positive. Future work will be focused on using this modelling framework to estimate soil compaction induced impacts on the environment and yield at a country or continental scale, its contribution to climate change, potential economic losses and the potential of prevention or remediation strategies to ameliorate such adverse impacts. REFERENCES Assouline, S., Narkis, K., Gherabli, R., Lefort, P., & Prat, M. (2014). Analysis of the impact of surface layer properties on evaporation from porous systems using column experiments and modified definition of characteristic length. Water Resources Research, 50(5), 3933-3955. Berisso, F. E., Schjnning, P., Keller, T., Lamand, M., Etana, A., De Jonge, L. W., Iversen, B. V., Arvidsson, J., & Forkman, J. (2012). Persistent effects of subsoil compaction on pore size distribution and gas transport in a loamy soil. Soil and Tillage Research, 122, 42-51. Bessou, C., Mary, B., Lonard, J., Roussel, M., Grhan, E., & Gabrielle, B. (2010). Modelling soil compaction impacts on nitrous oxide emissions in arable fields. European Journal of Soil Science, 61(3), 348-363. Carsel, R. F., & Parrish, R. S. (1988). Developing joint probability distributions of soil water retention characteristics. Water Resources Research, 24(5), 755-769. Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 39 Chamen, T. W. C., Moxey, A. P., Towers, W., Balana, B., & Hallett, P. D. (2015). Mitigating arable soil compaction: A review and analysis of available cost and benefit data. Soil and Tillage Research, 146(PA), 10-25. http://dx.doi.org/10.1016/j.still.2014.09.011 Coleman, K., Muhammed, S. E., Milne, A. E., Todman, L. C., Dailey, A. G., Glendining, M. J., & Whitmore, A. P. (2017). The landscape model: A model for exploring trade-offs between agricultural production and the environment. Science of The Total Environment, 609, 1483-1499. Colombi, T., Chagas, L., Walter, A., & Keller, T. (2018). Feedbacks between soil penetration resistance, root architecture and water uptake limit water accessibility and crop growth - A vicious circle. Science of The Total Environment, 626, 1026-1035. Del Grosso, S. J., Mosier, A. R., Parton, W. J., & Ojima, D. S. (2005). DAYCENT model analysis of past and contemporary soil N2O and net greenhouse gas flux for major crops in the USA. Soil and Tillage Research, 83(1), 9-24. Durner, W. (1994). Hydraulic conductivity estimation for soils with heterogeneous pore structure. Water Resources Research, 30(2), 211-223. Ghezzehei, T. A., & Or, D. (2001). Rheological properties of wet soils and clays under steady and oscillatory stresses. Soil Science Society of America Journal, 65(3), 624-637. Håkansson, I., & Lipiec, J. (2000). A review of the usefulness of relative bulk density values in studies of soil structure and compaction. Soil and Tillage Research, 53(2), 71-85. Hamza, M. A., & Anderson, W. K. (2005). Soil compaction in cropping systems: A review of the nature, causes and possible solutions. Soil and Tillage Research, 82(2), 121-145. Keller, T., Colombi, T., Ruiz, S., Manalili, M. P., Rek, J., Stadelmann, V., Wunderli, H., Breitenstein, D., Reiser, R., Oberholzer, H., Schymanski, S., Romero-Ruiz, A., Linde, N., Weisskopf, P., Walter, A., & Or, D. (2017). Long-term Soil Structure Observatory for monitoring post-compaction evolution of soil structure. Vadose Zone Journal, 16(4). Keller, T., Colombi, T., Ruiz, S., Schymanski, S. J., Weisskopf, P., Koestel, J., Sommer, M., Stadelmann, V., Breitenstein, D., Kirchgessner, N., Walter, A., & Or, D. (2021). Soil structure recovery following compaction: Short-term evolution of soil physical properties in a loamy soil. Soil Science Society of America Journal, 85(4), 1002-1020. Keller, T., Dfossez, P., Weisskopf, P., Arvidsson, J., & Richard, G. (2007). SoilFlex: A model for prediction of soil stresses and soil compaction due to agricultural field traffic including a synthesis of analytical approaches. Soil and Tillage Research, 93(2), 391-411. Kuhwald, M., Dörnhöfer, K., Oppelt, N., & Duttmann, R. (2018). Spatially Explicit Soil Compaction Risk Assessment of Arable Soils at Regional Scale: The SaSCiA-Model. Sustainability, 10(5), 1618. https://www.mdpi.com/2071-1050/10/5/1618 Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 40 Laudone, G. M., Matthews, G. P., Bird, N. R. A., Whalley, W. R., Cardenas, L. M., & Gregory, A. S. (2011). A model to predict the effects of soil structure on denitrification and N2O emission. JOURNAL OF HYDROLOGY, 409(1-2), 283-290. Meurer, K., Barron, J., Chenu, C., Coucheney, E., Fielding, M., Hallett, P., Herrmann, A. M., Keller, T., Koestel, J., Larsbo, M., Lewan, E., Or, D., Parsons, D., Parvin, N., Taylor, A., Vereecken, H., & Jarvis, N. (2020). A framework for modelling soil structure dynamics induced by biological activity. Global Change Biology, 26(10), 5382-5403. Nawaz, M. F., Bourri, G., & Trolard, F. (2013). Soil compaction impact and modelling. A review. Agronomy for Sustainable Development, 33(2), 291-309. Oldeman, L. R. (1992). Global extent of soil degradation. In Bi-Annual Report 1991-1992/ISRIC (pp. 1936). ISRIC. Or, D., & Ghezzehei, T. A. (2002). Modeling post-tillage soil structural dynamics: a review. Soil and Tillage Research, 64(1-2), 41-59. Parton, W. J., Mosier, A. R., Ojima, D. S., Valentine, D. W., Schimel, D. S., Weier, K., & Kulmala, A. E. (1996). Generalized model for N2 and N2O production from nitrification and denitrification. Global Biogeochemical Cycles, 10(3), 401-412. Pulido-Moncada, M., Petersen, S. O., & Munkholm, L. J. (2022). Soil compaction raises nitrous oxide emissions in managed agroecosystems. A review. Agronomy for Sustainable Development, 42(3), 1-26. Rabot, E., Wiesmeier, M., Schlter, S., & Vogel, H. (2018). Soil structure as an indicator of soil functions : A review. Geoderma, 314, 122-137. Romero-Ruiz, A. (2021). Geophysical Methods for Field-Scale Characterization of Soil Structure [Theses, {Universit{\'e} de Lausanne}]. https://hal.archives-ouvertes.fr/tel-03207029 Romero-Ruiz, A., Linde, N., Baron, L., Breitenstein, D., Keller, T., & Or, D. (2022). Lasting effects of soil compaction on soil water regime confirmed by geoelectrical monitoring. Water Resources Research, n/a(n/a), e2021WR030696. Romero-Ruiz, A., Monaghan, R., Milne, A., Coleman, K., Cardenas, L., Carmen, S., & Whitmore, A. (2023). Modelling changes in soil structure caused by livestock treading. Geoderma, 431, 116331. https://doi.org/https://doi.org/10.1016/j.geoderma.2023.116331 Semenov, M. A., Jamieson, P. D., & Martre, P. (2007). Deconvoluting nitrogen use efficiency in wheat: A simulation study. European Journal of Agronomy, 26(3), 283-294. https://www.sciencedirect.com/science/article/pii/S1161030106001419 Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 41 Soares, A., Moldrup, P., Vendelboe, A. L., Katuwal, S., Norgaard, T., Delerue-Matos, C., Tuller, M., & de Jonge, L. W. (2015). Effects of soil compaction and organic carbon content on preferential flow in loamy field soils. Soil Science, 180(1), 10-20. Steinfeld, H., Gerber, P., Wassenaar, T. D., Castel, V., Rosales, M., Rosales, M., & de Haan, C. (2006). Livestock's long shadow: environmental issues and options. Food \& Agriculture Org. Tubeileh, A., Groleau-Renaud, V., Plantureux, S., & Guckert, A. (2003). Effect of soil compaction on photosynthesis and carbon partitioning within a maize–soil system. Soil and Tillage Research, 71(2), 151-161. Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 48 Figure S7 Modelling results comparing relative impacts of soil compaction on air filled pore space (WFPS) for different (a) texture, (b) climates, and (c) recovery rates. Modelling results comparing relative impacts of soil compaction on relative WFPS for different (d) texture, (e) climates, and (f) recovery rates. These modelling results correspond to soil structure properties calibrated to the Soil Structure Observatory. The control line indicates no compaction (no compaction = 1) . Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 49 Figure S8 Modelling results comparing relative impacts of soil compaction on the transpiration reduction factor (soil water related) for biomass formation for different (a) texture, (b) climates, and (c) recovery rates. These modelling results correspond to soil structure properties calibrated to the Soil Structure Observatory. The control line indicates no compaction (no compaction = 1) . Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 50 Figure S9 Modelling results comparing absolute impacts of soil compaction on (a) crop yield (at year 20 after initial compaction), (b) soil organic carbon stocks (at year 20 after initial compaction), (c) nitrous oxide emissions (during the first year after initial compaction), and (d) nitrate leaching (during the first year after initial compaction) for different levels of compaction represented as increasing bulk density. These modelling results show differences between different clay contents. Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 51 Table S1. Soil properties that are impacted by soil compaction and accounted for in the modelling approach of this work package. Soil properties Compaction-related implementation in the model Bulk density Through compaction model (Romero-Ruiz et. al., 2023). It depends on the texture and weight of compacting element. Penetration resistance Calculated as a function of the bulk density. Total porosity Calculated as a function of the bulk density. Macroporosity Calculated as a function of the bulk density. K sat Calculated as a function of the bulk density. Water retention (alpha and n) Not needed. Assumed constant for simplicity and based on literature observations. Relative gas diffusion coefficient Calculated as a function of the bulk density and water-filled pore space. Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 52 Table S2. Summary of soil processes impacted by compaction (through the soil properties shown in Table 1), how such impacts are implemented in the current version of the modelling framework, how the implementation can be improved and the outputs of the model that are sensitive to such processes. Soil processes Compactionrelated implementation in the model Potential model improvement Outputs tracked and affected N 2 O e N 2 e N leach C stocks Yield Water flow As affected by the saturated hydraulic conductivity. Accounting for compaction effects in water retention properties, x x x x x Soil evaporation No Adding a compactiondependent soil evaporation model can be included. x x x x x Root water uptake As affected by root growth and water flow. TBD x x x x x Drainage As affected by water retention and transport properties. TBD x x x x x Nitrification Empirical functions that depend on the water filled pore space and temperature. Including mechanistic models. x x x Denitrification Empirical functions that depend on the water filled pore space and temperature. Including mechanistic models. x x x Root growth The rate depends on the penetration resistance. TBD x x Root decay Depends on a decay rate. TBD x Oxygen diffusion Depends on the relative diffusion coefficient. TBD x x x x x Carbon mineralization Rates affected by texture and moisture. TBD x x Deliverable 3.3: Modelling soil compaction impact on key soil functions: results This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 862695 53 Soil structure recovery Empirical function. Making the recovery rates dependent on texture, depth and degree of compaction. x x x x x