The Diagnostic Void: A Systemic Failure of Commercial Precision Agriculture Platforms to Integrate Science-Based Soil Health Indicators

The Diagnostic Void: A Systemic Failure of Commercial Precision Agriculture Platforms to Integrate Science-Based Soil Health Indicators | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Diagnostic Void: A Systemic Failure of Commercial Precision Agriculture Platforms to Integrate Science-Based Soil Health Indicators Hanna Radziuk, Marcin Świtoniak This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7541765/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Soil degradation is a primary threat to agricultural sustainability, yet addressing it is hampered by high spatial variability of soil properties. While Precision Agriculture (PA) platforms promise tools for site-specific management, their capacity to diagnose the root causes of poor soil health, such as physical degradation, remains critically unassessed. This study confronts this issue by empirically testing the dominant „symptom-based” paradigm of commercial PA. We conducted a multi-phase assessment of 34 platforms, including an in-depth user-experience (UX) test of nine major platforms, using a unique, field-verified dataset of soil physical health indicators (aggregate stability, erodibility K-factor) from a representative hummocky moraine landscape in Poland. Our findings reveal a profound „diagnostic void” in the commercial PA ecosystem. While we confirmed a strong statistical correlation between soil degradation indicators (the „cause”) and vegetation indices like NDVI (the „symptom”), we discovered that none of the tested platforms offered built-in tools for soil structure or erosion analysis. This failure was compounded by a convergence of technical (e.g., lack of GeoTIFF support), economic (paywalls), and ecosystem-level barriers that systematically prevent the integration of user-generated scientific soil data. The current generation of PA platforms is fundamentally limited to treating symptoms rather than diagnosing causes, forcing users into a reactive and potentially unsustainable management paradigm. We argue that this diagnostic gap is not a mere technological oversight but a systemic failure driven by market priorities. Bridging this gap requires a fundamental reorientation of the ag-tech sector towards open, interoperable, and analytically robust platforms that empower science-based soil stewardship. Agroecology Geographic Information Systems Environmental Policy Diagnostic void precision agriculture digital agriculture soil degradation sustainability assessment technology adoption barriers Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction Soil degradation is a major threat to agricultural productivity and environmental quality worldwide (Lal, 2004 ; Arias-Navarro et al., 2024 ). In response, the modern concept of soil health – defined as the soil’s capacity to function as a living ecosystem – has become central to sustainable agriculture (Keesstra et al., 2016 ; Lehmann et al., 2020 ). This concept encompasses not only chemical properties but also physical and biological characteristics, such as soil structure and microbiome activity, which are essential for supporting ecosystem services and ensuring long-term productivity (Barrios, 2007 ; Smith et al., 2016 ). Two forms of soil degradation – water erosion and soil compaction – are particularly widespread in Europe, affecting over 60% of agricultural land (Arias-Navarro et al., 2024 ). Water erosion impacts 24% of EU land, occurring at rates that exceed natural soil formation (Panagos et al., 2015 ), whereas compaction affects 23% of agricultural soils, especially in subsoil (plowpan) layer (Horn and Fleige, 2009 ). The susceptibility of soils to these processes is closely linked to their physical properties, particularly aggregate stability and overall structure (Dexter et al., 2008 ; Lehmann et al., 2020 ). A stable soil structure enhances water infiltration, aeration, and root growth, while its degradation increases runoff and erosion risk. Assessing indicators of soil physical health, such as aggregate stability and erodibility (e.g., the K-factor), is therefore essential for effective soil management (Radziuk et al., 2021 ; Sheng et al., 2024 ). This is particularly important in heterogeneous landscapes, such as hummocky moraine terrain, where complex topography and spatial variability of soils create challenges for both monitoring and management (Świtoniak, 2014 ; Arias-Navarro et al., 2024 ). Precision Agriculture (PA) technologies have been promoted as a solution for managing microscale soil variability (Basso et al., 2016 ; Pierce and Nowak, 1999 ). PA platforms provide tools for spatial analysis and variable rate application (VRA), often relying on proxy satellite data, such as the NDVI, to identify low-performing zones (Babos et al., 2024 ; Zagorda and Walczykova, 2018). This raises a central question for science-based soil management: while NDVI can effectively identify symptom - such as reduced crop vigor – are commercial PA platforms equipped to help users diagnose the underlying cause (e.g., poor soil physical health) and integrate this diagnostic insight into management decisions? Failure to do so can lead to inefficient resource use and misguided agronomic interventions, resulting in both economic losses for the farmer and negative environmental impacts. Despite advances in both soil science and PA platforms, a notable gap remains: no study has systematically evaluated whether commercial PA platforms can bridge the symptom-cause divide by incorporating field-verified data on soil structure and erosion (Lykhovyd et al., 2024 ). Closing this gap is crucial for the wider adoption of science-based soil management practices. However, the widespread adoption of these platforms raises a critical and underexplored question: are these tools designed to support holistic, science-based soil management, or do their underlying technical and business models inadvertently promote a superficial, symptom-driven approach? We hypothesize that despite their technological sophistication, current commercial PA platforms face systemic limitations in their ability to support causal diagnosis of soil degradation. This study tests this hypothesis by moving beyond declared features to empirically assess the practical capabilities of these platforms in bridging the critical gap between the remotely-sensed symptom (vegetation vigor) and the field-verified cause (soil physical degradation). To investigate this, we address the following key research questions: What is the nature and strength of the relationship between field-verified soil degradation indicators and remotely-sensed vegetation indices (NDVI)? To what extent do commercial PA platforms provide built-in tools to diagnose the erosion-related causes of the variability observed in NDVI data? What are there the practical possibilities for integrating user-generated, field-verified soil data into these platforms as an alternative pathway for science-based decision-making? 2. Materials and Methods To address the research questions outlined in the introduction, a multi-stage methodology was designed to systematically bridge field-verified soil science with an empirical assessment of commercial digital tools. As illustrated in the schematic overview (Fig. 1), the research process follows a logical sequence: beginning with the quantification of soil degradation in the field, continuing with an analysis of its manifestation in remote sensing data, and concluding with the evaluation of PA platform capabilities to address these issues. The following sections describe each stage of this integrated approach in detail. 2.1 Field and Laboratory Study This part of the study was designed to empirically quantify the extent and spatial variability of soil physical degradation within a representative, agriculturally managed landscape. The field-verified data collected during this phase served as the primary scientific input for the subsequent evaluation of PA platforms. 2.1.1. Study Areas and Soil Sampling The study was conducted in Northern Poland within two separate agricultural fields – Orzechowo (CT) and Sokołowo (ST), located within the Chełmno and Dobrzyń Lakeland respectively (Fig. 2A). Both physical-geographical regions are characterized by a hummocky moraine landscape formed during the Weichselian glaciation (Solon et al., 2018), with a mean annual air temperature of approximately 8°C and average yearly precipitation of 604 mm (Kottek et al., 2006). The two study sites were selected based on their comparable geomorphological settings and parent materials, enabling a robust comparative case study. Due to the variability of the terrain, the soils in both study areas exhibit considerable natural catenal differences in soil-forming processes – most notably, clay illuviation on higher elevations and strong groundwater influence in depressions. A history of intensive agricultural use has led to significant anthropogenic denudation, resulting in a complex mosaic of soils with varying degrees of profile truncation on summits and slopes, and colluvial material accumulation in lower-lying areas (Świtoniak, 2014; Świtoniak et al. 2020; Radziuk and Świtoniak, 2022). This combination of inherent and human-induced spatial variability makes these landscapes an ideal real-world testbed for evaluating the capabilities of precision agriculture (PA) tools. The first site, a 40-ha field near Orzechowo managed under Conventional Tillage (CT) (Fig. 2C), has been the subject of intensive previous research focused on the spatial distribution of soil properties and erosion-induced carbon redistribution (Radziuk and Świtoniak, 2022). The second site, a 47-ha field near Sokołowo managed under Strip-Till (ST) (Fig. 2B) – has also been previously investigated in the context of conservation tillage and its effects on restoring the physical properties of severely eroded soils (Świtoniak et al., 2020). For both fields, crop rotation follows a typical regional pattern, including winter wheat, winter oilseed rape, and spring wheat, with fertilization applied in accordance with standard agronomic recommendations. This well-established research background provides a robust and well-documented foundation for the present study. To investigate soil cover and its properties, a representative catena along a hillslope was selected within each study field. In the CT study field, a comprehensive sampling campaign was performed across four catenae to capture the high spatial variability (16 full soil profiles in 4 catenae and 50 supplementary augering points). In the ST study field, a focused case study was conducted along a single representative catena, yielding 4 soil profiles with similar geomorphological settings to those in the CT field. Soil samples were collected from both fields during autumn 2019. At each profile and augering point, disturbed composite samples were taken from the topsoil (0–30 cm) for the analysis of basic physical and chemical properties. Additionally, at the 16 profiles in the CT field and 4 profiles in the ST field, large undisturbed block samples were carefully extracted from the Ap and subsoil horizons for aggregate stability assessment. The collected point data, combined with geomorphological analysis of a digital elevation model (DEM) and interpretation of high-resolution orthophotomaps, served as the basis for creating the final soil maps in a GIS environment. 2.1.2. Laboratory Analyses and Key Indicators Soil samples collected from both study fields underwent comprehensive laboratory analyses to characterize their physical, chemical, and structural properties. All laboratory analyses were conducted at the Laboratory for Environmental Analysis (Nicolaus Copernicus University in Toruń, Poland), with detailed methodologies described in previously published works (Radziuk and Świtoniak, 2022, 2021; Świtoniak M. et al., 2020). Disturbed samples were used to determine particle size distribution (aerometric method), soil pH (potentiometrically), soil organic carbon (SOC) content (Tiurin method), and calcium carbonate (CaCO3) content (Scheibler method). Bulk density was ascertained using undisturbed core samples (Hillel, 2013; Revut, 1972). Undisturbed block samples from both CT and ST sites were used to assess aggregate stability. Air-dried samples were first dry-sieved to determine the aggregate size distribution and calculate the MWDdry, mm. Subsequently, the water stability of aggregates from the 7–10 mm fraction was quantified by measuring the TAD, s using a static wetting test (Rząsa and Owczrzak, 2004). Soil erodibility was calculated as the K-factor using the Williams' (Williams, 1990) formula from the EPIC model, which is based on soil texture and SOC content (see Supplementary Material, S1, for detailed equations). These two parameters, TAD and K-factor, representing soil structural resilience and erosion susceptibility respectively, formed the core of the scientific dataset (Radziuk and Świtoniak, 2021). 2.1.3. Creation of the Field-Verified Dataset The point-based data from the field and laboratory analyses were processed in a GIS environment. This involved integrating the point data with geomorphological analysis of a high-resolution Digital Elevation Model (DEM) and interpretation of orthophotomaps. The final output of this stage was a comprehensive field-verified dataset containing spatial layers and tabular data in standard formats, including GeoTIFF (for interpolated maps of TAD and K-factor), Shapefile (for delineated soil/erosion polygons and sampling points), and CSV/Excel (for raw point data). This dataset represents the complex, high-resolution „ground truth” of soil physical condition used for the subsequent testing of PA platforms. 2.2. PA Platform Assessment Methodology To critically evaluate the extent to which commercial PA platforms support the integration of advanced soil data – particularly indicators of soil structure and erosion risk – a custom, multi-phase assessment framework was developed for this study. Due to the lack of standardized protocols for evaluating the scientific data handling capabilities of such diverse commercial software, this methodology was designed to systematically bridge a general market overview with empirical, user-centric testing. 2.2.1. Phase 1 & 2: Market Survey, Platform Selection, and Rationale for Empirical Testing The selection of PA platforms for empirical testing was preceded by a systematic, two-phase market survey. The initial phase involved a comprehensive review of the commercial PA market, identifying 34 platforms based on extensive online searches. To systematically assess their declared capabilities, each platform's publicly available information was evaluated, with a focus on functionalities related to soil data management. Each platform was classified according to five distinct levels of soil-related data handling, ranging to the next levels: (a) No relevant functionality. The platform offers no features related to soil data. (b) Basic soil data visualization or management. Limited capabilities, such as map overlays of pH or lab results. (c) Advanced integration and analysis of soil data. Features like multi-layer analytics or analysis of soil texture maps. (d) Indirect assessment of soil structure or erosion risk. Use of proxy data (e.g., topography) that correlate with soil physical health. (e) Dedicated tools for erosion modeling. Explicit tools for erosion risk, such as K-factor calculators or risk assessment modules. The detailed results of this market survey and functional categorization are provided in the Supplementary Material (Figure S1, Tab. S2). This survey revealed two key findings that shaped the subsequent research. First, a critical gap became immediately evident: out of the 34 platforms, none offered dedicated, built-in tools for the assessment of soil structure or for quantitative erosion risk modeling. Second, a targeted literature review using Scopus and Web of Science confirmed a significant scarcity of independent academic evaluations for most of these commercial tools (see Supplementary Material, Tab. S1). This lack of both declared functionalities and academic scrutiny provided the critical justification for the empirical testing in Phase 3. Based on this preliminary analysis, a diverse set of nine platforms was purposively selected for in-depth testing. The selection was designed to be representative of the market and was stratified to include platforms from across the functionality levels and different business models, including: hardware-centric ecosystems (e.g., John Deere Operations Center, AgFiniti, AGMRI), major corporate solutions (e.g., Climate FieldView, Granular), specialized data analytics platforms (e.g., GeoPard Agriculture), and widely accessible freemium platforms (e.g., OneSoil, SatAgro). One additional platform ( AgriCircle ) was assessed through direct contact due to limited access. 2.2.2. Phase 3: Empirical User Experience (UX) Testing Testing of all nine selected platforms was conducted using a comprehensive, field-verified dataset comprising raster layers (GeoTIFF) for aggregate stability and erodibility, vector maps (Shapefiles) for soil types and erosion zones, and tabular data (CSV, Excel) for laboratory results. All data were prepared in standard formats commonly used in GIS and agronomic research. The testing protocol was designed to directly address the study's research questions by focusing on the key tasks required to bridge the „symptom-cause” divide. Instead of a comprehensive evaluation of all platform features, the protocol specifically targeted the functionalities essential for integrating and utilizing scientific soil health data. The core tested tasks included: Data Import: Systematically attempting to import the field-verified dataset (TAD and K-factor maps in GeoTIFF, soil zone polygons in Shapefile) to assess technical compatibility and data integration barriers. Diagnostic Capability: Exploring the platforms for any built-in tools capable of analyzing soil physical properties or modeling erosion risk. Science-Based VRA Creation: Evaluating the possibility of creating VRA maps based directly on the imported scientific soil data, as an alternative to proxy-based methods. All observations – including successful and failed operations, error messages, and user friction points related to these specific tasks – were meticulously recorded. These insights formed the empirical foundation for identifying the barriers that prevent the practical application of scientific soil knowledge in commercial PA platforms. 2.3. Remote Sensing Data Acquisition and Analysis To determine the nature and strength of the relationship between field-verified soil degradation indicators and vegetation vigor, this study employed the NDVI derived from satellite imagery. These data was acquired and analyzed through a two-step, complementary approach. First, for a qualitative, cross-platform comparison, NDVI imagery was visually assessed. The specific dates for this visual analysis were selected based on the simultaneous availability of cloud-free imagery across the selected platforms (OneSoil, SatAgro, and GeoPard Agriculture) that provided accessible map visualization in their free/trial versions. This resulted in the selection of June 21st, 2022 for the CT field and June 6th, 2023 for the ST field. While OneSoil and SatAgro provided standard NDVI layers, GeoPard Agriculture offered the Enhanced Vegetation Index 2 (EVI2) as its primary indicator of vegetation vigor; therefore, EVI2 was used for this platform. This approach, while necessary for a direct comparison of how different systems present data, highlights a practical limitation of relying solely on commercial platform outputs, as data availability is often inconsistent. Second, to enable a robust quantitative analysis, a separate approach was employed. As the tested platforms did not offer a functionality to export point-specific NDVI values for user-defined coordinates, precise values were extracted for the 66 sampling points using Google Earth Engine (GEE). This process utilized the original Sentinel-2 Level-2A (atmospherically corrected) imagery for both of the dates identified above (21.06.2022 and 06.06.2023) in the platforms, ensuring consistency between the visual and statistical analyses. To confirm the stability of the observed relationships, this extraction was repeated for additional dates representing different crop growth stages. The relationship between these point-specific NDVI values and the field-verified soil data was then quantified using non-parametric statistical methods (Spearman's rank correlation and Kruskal-Wallis test with a subsequent Dunn's post-hoc test), chosen to accommodate the data's characteristics. 3. Results This section presents the empirical results of the study, beginning with a detailed characterization of soil structural properties and erosion risk in the investigated agricultural fields (Section 3.1 ). Subsequently, it presents the relationship between remote sensing data (NDVI) and field-verified soil properties (Section 3.2 ) and concludes with an analysis of the empirical observations of the possibilities of commercial PA platforms to address these issues (Section 3.3 ). 3.1. Field-Verified Soil Degradation and Erodibility The field investigations confirmed that the soil cover in the studied hummocky moraine landscape is highly complex. Years of agricultural use and erosion have created a mosaic of different soil types across the fields, from completely eroded soils on hilltops to accumulated, fertile soils in depressions. This inherent spatial variability represents the core management challenge in this region (Świtoniak, 2014 ; Świtoniak et al., 2016 ). The soils in the study areas (both the 40 ha CT field and the 47 ha ST field) were broadly classified into four distinct units (A, B, C, D) based on the degree of anthropogenic denudation and their position along a hillslope catena (Fig. 3 ). This classification, initially developed for this type of landscape (Świtoniak et al., 2016 ), reflects the spatial arrangement of soil types resulting from erosion and accumulation processes. A detailed comparison of the key properties for these soil categories under both tillage systems is presented in Table 1 . Soil A and B are typically found on upper slope segments and summits, where soil material has been removed due to denudation processes, leading to soil profile truncation. Soils A: Regosols – These represent completely eroded soils, primarily found on hummock summits and upper convex slopes. According to WRB (IUSS, 2015a), they are classified as Eutric Regosols (Protocalcic). Their profiles are weakly developed, often lacking diagnostic horizons, and are built directly from glacial till, resulting in an ACkp-Ck sequence. The plough layer often contains admixtures of calcareous parent material (Ck horizon), contributing to a typically low organic carbon content (C-SOC of 0.62% and 0.92% for CT and ST fields, respectively) and, particularly in the CT field, a high carbonate content (7.16%) (Table 1 ). These soils are characterized by the lowest structural stability, as evidenced by the lowest TAD values (208.8 s and 385.2 s for CT and ST, respectively) and the highest K-factor values (0.033 and 0.032 t·ha·h·ha–1·MJ–1·mm–1), indicating the greatest susceptibility to erosion (Table 1 ). Soils B: Eroded clay-illuvial soils – These represent strongly eroded Luvisols, typically located on upper and middle slope segments with noticeable illuvial material (Bt horizon) exposed within the plough layer (ABtp horizon). They are generally classified as Haplic Luvisols (Protocalcic) (IUSS, 2015a). These soils often exhibit a higher clay content in the plough layer (up to 18%) and low SOC content. As detailed in Table 1 , their structural stability (TAD of 374.6 s and 797.8 s for CT and ST respectively) and erodibility (K-factor of 0.031 and 0.027 t·ha·h·ha–1·MJ–1·mm–1) are intermediate between the completely eroded Regosols (Soils A) and the less disturbed soils of the lower slope segments. Table 1 Selected soil properties, aggregate stability indicators (TAD, MWDdry), and erodibility (K-factor) in the arable layer under conventional tillage and strip-till systems Soil group Soils A Soils B Soils C Soils D WRB name Eutric Regosol (Protocalcic) Haplic Luvisol (Protocalcic) Albic Luvisol / Mollic Gleysol Endogleyic Phaeozem / Mollic Gleysol Soil horizon sequence ACkp-Ck ABtp-Bt-Ck Ap-E-Bt-Ckg/ Ap-A-Eg-Btkl Ap-A- A2-Ab-Ckl Conventional tillage (field Orzechowo, 4 catenae) Sand (%) 59.3 ± 4.0 56.3 ± 4.3 66.0 ± 4.1 60.0 ± 3.9 Silt (%) 24.8 ± 3.3 25.8 ± 3.2 24.0 ± 2.6 31.0 ± 2.9 Clay (%) 16.8 ± 2.1 18.0 ± 2.2 10.0 ± 2.2 9.0 ± 1.6 pHKCl 7.48 ± 0.14 7.31 ± 0.17 6.38 ± 0.52 6.77 ± 1.14 C-SOC (%) 0.62 ± 0.09 0.77 ± 0.11 0.93 ± 0.16 2.35 ± 1.34 CaCO3 (%) 7.16 ± 1.3 1.35 ± 0.45 0.17 ± 0.16 0.23 ± 0.16 Bulk density (kg*m-3) 1.64 1.54 1.62 1.56 K (t·ha·h·ha–1·MJ–1·mm–1) 0.033 0.031 0.030 0.028 MWDdry (mm) 8.76 10.79 7.44 8.16 TAD (sec.) 208.8 ± 81.9 374.6 ± 31.2 616.2 ± 28.8 771.8 ± 30.3 Strip-till (field Sokołowo, 1 catena) Sand (%) 59 72 70 64 Silt (%) 27 20 21 31 Clay (%) 14 8 9 5 pHKCl 7.46 7.18 7.03 6.27 C-SOC (%) 0.92 1.43 1.27 2.39 CaCO3 (%) 3.3 0.6 0.5 0.1 Bulk density (kg*m-3) 1.35 1.28 1.28 1.47 K (t·ha·h·ha–1·MJ–1·mm–1) 0.032 0.027 0.026 0.027 MWDdry (mm) 8.14 6.14 7.45 2.45 TAD (sec.) 385.2 ± 66.2 797.8 ± 81.3 489.6 ± 27.5 900.0 ± 0 Next two soil units (C and D) are typically found on lower slope segments, footslopes, and in depressions, exhibiting less severe truncation (C) or accumulation of colluvial material (D). Soils C : These types represent slightly eroded or undisturbed Luvisols, primarily found in the lower parts of slopes. They are characterized by a full or nearly full sequence of genetic horizons (e.g., Ap-E-Bt-Ckl, or Ap-A-Eg-2Bt-2Ckg), classifying as Albic Luvisols or Mollic Gleysols (IUSS, 2015a). These soils represent the largest portion of the CT study area (58%), forming a „background” for the more degraded soils. They exhibit significantly improved structural stability (TAD of 616.2 s for CT) and lower erodibility (K-factor of 0.030 t·ha·h·ha–1·MJ–1·mm–1 for CT) compared to the truncated soils of the upper slopes, which corresponds with their higher C-SOC content (Table 1 ). Soils D: colluvial soils , formed by the accumulation of eroded material in depressions and at footslopes. Their morphology often indicates significant influence of groundwater gleying. They are classified as Endogleyic Phaeozems or Mollic Gleysols (IUSS, 2015a). As shown in Table 1 , these soils are characterized by the highest organic carbon content (C-SOC of 2.35% and 2.39% for CT and ST respectively) and, consequently, the most stable soil structure, with the highest TAD values (771.8 s and 900.0 s) and the lowest erodibility (K-factor of 0.028 and 0.027 t·ha·h·ha–1·MJ–1·mm–1 for CT and ST fields respectively). These findings document the complex, field-scale reality of soil degradation and form the „ground truth” against which platform capabilities are evaluated. 3.2. Remote Sensing of a Symptom: Linking NDVI to Soil Degradation Patterns While direct, field-based assessment provides the most accurate characterization of soil physical properties, it is labor-intensive and costly to perform at a large scale. In contrast, PA heavily relies on remote sensing data, particularly vegetation indices, as a practical and cost-effective means to assess in-field variability. Therefore, the next logical step in our investigation was to determine whether a quantifiable relationship exists between the field-verified „cause” (soil degradation) and the remotely-sensed „symptom” (vegetation vigor). To investigate this relationship, a quantitative statistical analysis was conducted based on 66 point samples from the CT field, where precise NDVI values were extracted from Sentinel-2 imagery via GEE (see Supplementary Material, Tab. S3 and Fig. S2). A non-parametric Spearman's rank correlation (Table 2 ) revealed a significant positive relationship between NDVI and aggregate stability (TAD, s) (r2 = 0.56, p < 0.01) and a statistically significant, but weak, negative relationship with the soil erodibility K-factor, t·ha·h·ha–1·MJ–1·mm–1 (r2 = -0.35, p < 0.05). This indicates that soils with higher structural stability and lower erosion susceptibility support more vigorous vegetation. Crucially, a strong positive correlation was found between the NDVI values from 2022 and 2023 (r2 = 0.72, p < 0.01), indicating a high temporal stability of the observed spatial patterns. Table 2 Spearman's rank correlation coefficients (rs) between NDVI and selected soil degradation indicators for the field under CT system (in bold statistically significant (p < 0.05) data) K, t·ha·h·ha –1 ·MJ –1 ·mm –1 Soil unit TAD, s NDVI 21.06.2022 NDVI 06.06.2023 K, t·ha·h·ha –1 ·MJ –1 ·mm –1 -0.64 -0.58 -0.28 -0.35 Soil unit -0.64 0.86 0.53 0.49 TAD, s -0.58 0.86 0.56 0.59 NDVI 19.06.2022 -0.28 0.53 0.56 0.72 NDVI 06.06.2023 -0.35 0.49 0.59 0.72 Furthermore, a Kruskal-Wallis test showed highly significant differences in median NDVI values among the four delineated soil units (H = 99.12, p < 0.001). A subsequent Dunn's post-hoc analysis (Table 3 ) confirmed that the most eroded soils (Group A) exhibited significantly lower NDVI values than the undisturbed and colluvial soils (Groups C and D), a relationship further illustrated by the scatter plot in Fig. 4 . Table 3 Results of the Dunn's post-hoc test for pairwise comparisons of median NDVI (21.06.2022) values between soil groups Soils_A Soils_B Soils_C Soils_D Soils_A 0.1424 0.01141 5.65E-05 Soils_B 0.1424 0.3783 0.01047 Soils_C 0.01141 0.3783 0.04594 Soils_D 5.65E-05 0.01047 0.04594 A qualitative visual assessment of several commercial PA platforms confirmed that they also detect similar spatial patterns of in-field variability, albeit with significant differences in visualization style, as shown for the CT field in Fig. 5 . The consistent, strong visual and statistical correlation observed leads to a crucial insight. The high temporal stability of the NDVI patterns between 2022 and 2023, despite a change in the cultivated crop, suggests that the legacy of erosion involves more profound, lasting changes than just structural degradation. This indicates that the spatial variability of crop performance is primarily governed by fundamental, persistent site characteristics – such as irreversible alterations in soil texture, depth, and water-holding capacity – rather than by the specific needs of the crop in a given year. These results empirically demonstrate that remotely-sensed vegetation indices are effective at identifying the complex, persistent symptoms of poor soil physical health. This finding provides the context for the subsequent evaluation of the diagnostic capabilities of PA platforms. 3.3. Empirical Assessment of Practical Capabilities of Commercial PA Platforms for Soil Health Management Having established the strong, quantifiable link between the underlying cause (soil degradation) and its remotely-sensed symptom (NDVI), the final and central part of the investigation assessed the practical capabilities of the nine selected commercial PA platforms to facilitate a science-based, causal diagnosis. The key findings from the empirical user experience tests are summarized in a comparative matrix (Table 4 ) and detailed below. Data Integration and Export Capabilities : The ability to import and export standard scientific data formats was found to be a primary and widespread limitation. Raster (GeoTIFF) Import. A direct and functional tool for importing GeoTIFF files, the standard format for continuous soil maps (e.g., TAD, K-factor), was not found in the tested versions of AgFiniti , AGMRI , OneSoil , and JDOC . While platforms like Climate FieldView and AgriCircle declared support, this functionality was not verifiable without paid access. For SatAgro and GeoPard , raster import was either explicitly a premium feature or proved problematic during testing. Vector (Shapefile) and Tabular (CSV) Import. While Shapefile import for simple field boundaries was commonly supported (e.g., JDOC , Granular ), attempts to import thematic polygons with complex attributes frequently resulted in critical errors. Platforms like AGMRI and Granular failed to process standard archives, returning cryptic messages („Missing date for soil data activity”, „File not readable”) (see Supplementary Material, Fig. S4B). Other platforms, like GeoPard , exhibited issues with non-numeric attributes (see Supplementary Material, Figure S4A). Furthermore, some platforms (e.g. , OneSoil ) demonstrated semantic limitations by misinterpreting soil zone polygons as separate fields (see Supplementary Material, Fig. S4C). Support for simple tabular CSV data was also inconsistent and often unavailable. Data Export. The crucial functionality of exporting analyzed data in interoperable formats (e.g., Shapefile) was severely restricted, often absent in free versions ( OneSoil ) or available only as a premium feature ( SatAgro , FieldView ). GeoPard was a notable exception, offering robust export functionality in its trial version. Functional Capabilities and VRA Pathways : Availability of Diagnostic Tools. A near-universal finding was the absence of dedicated, built-in tools for soil structure or erosion risk analysis. None of the tested platforms offered functionalities to calculate or interpret key physical indicators like TAD or K-factor. In the case of AGMRI , the entire „Analytics” section was marked as „Coming Soon” highlighting a functional deficit. Even advanced platforms like GeoPard , while offering a powerful scripting environment (Python), lacked ready-to-use modules, placing the responsibility of additional algorithm development on the user (see Supplementary Material, Fig. S5C). The „black box” nature of some available tools was also problematic, with platforms offering simplified, non-scientific soil classifications (see Supplementary Material, Fig. S5A) or generating unreliable data outputs (see Supplementary Material, Fig. S5B). Science-Based VRA. The ability to create VRA maps based directly on user-imported scientific soil data was systematically restricted and consistently treated as a premium feature locked behind a paywall ( AgFiniti , FieldView , GeoPard ). Table 4 Summary of Key Functionalities and Observed Outcomes for Tested PA Platforms Platform Data Integration Functional Capability Accessibility & Business Model Raster (GeoTIFF) Import Thematic Shapefile Import Data Export (Interoperable) Soil Structure / Erosion Analytics VRA from Custom Soil Data Open Access / Functional Free Trial Ecosystem / Business Model Barrier Compatibility with other PA Platforms AgFiniti (Ag Leader) X* P PL X PL P High ✓ (JDOC, Climate FieldView, New Holland) AGMRI (IntelinAir) X P X CO X X High ✓ (JDOC, Climate FieldView, AgFiniti) AgriCircle DL DL DL X DL X High n.d. Climate FieldView (Bayer) DL/P P PL X PL P Medium ✓(JDOC) GeoPard Agriculture P ✓ (Boundary) ✓ (Shapefile) X PL ✓ Low ✓ (Very Strong, 2-way with JDOC) Granular (Corteva) P ✓ (Boundary) X X X ✓/P High ✓ (AgFinity, JDOC, PP Panorama, FieldAlytics, CNH) JDOC (John Deere) X ✓ (Boundary) PL X X X Extreme ✓( about 100 third-party software partners) OneSoil X ✓ (Boundary, poligons) X X X ✓/P Low ✓(JDOC) SatAgro PL PL PL X PL ✓/P Medium ✓ (JDOC, Trimble Connected Farm) *✓ - Supported and functional in the tested version; X - Absent or not found in the tested version; P - Problematic, limited, or non-intuitive functionality; PL - Functionality exists but is behind a paywall; DL - Declared by the vendor but not empirically verified in the test; CO - Coming Soon (feature announced but not available); ✓ (Boundary) - Supported only for simple field boundaries; Barrier Level (Low/Medium/High/Extreme): A qualitative assessment of the severity of ecosystem/business model barriers to open access and data use. Platform Access and Ecosystem Integration : Fundamental access was often impeded by business model constraints. JDOC and AgFiniti demonstrated a strong reliance on a dealer-centric network, requiring formal business affiliation. AgStudio (Corteva) offered no public sign-up option, while AGMRI 's model presented significant geographical and organizational hurdles. The results also highlight a trend towards interoperability, with platforms like JDOC connecting to a wide network of nearly 100 third-party software partners (see Supplementary Material, S2 for a full list). This extensive ecosystem highlights the platform's focus on operational data integration yet a review of these partners reveals a scarcity of solutions focused on soil physical health diagnostics. 4. DISCUSSION The results of this study provide a multi-layered view of the significant gap between the scientific understanding of soil physical health and the current capabilities of commercial PA software. In this section, we interpret these findings, beginning with the relationship between the degradation cause and its remotely-sensed symptom, then moving to a detailed analysis of the „diagnostic void” in PA platforms and the systemic logic that underpins it. 4.1. Establishing the Ground Truth: The Signature of Soil Degradation on Crop Vigor The findings of this study provide a clear, field-based picture of the complex nature of soil physical degradation in intensively cultivated hummocky moraine landscapes. The direct comparison between conventional (CT) and strip-till (ST) systems, although based on a limited number of catenae, provided valuable insights into the potential effects of tillage on soil physical health. Our findings indicate a clear trend toward improved soil structure under the ST system, particularly for the most degraded soil units. The strongly eroded Luvisols (Soils B), for example, exhibited a more than twofold improvement in aggregate water stability (TAD) under ST compared to the same soil unit under CT (Table 1 ). This suggests that while these soils are significantly degraded, they possess a high potential for structural regeneration when managed under conservation-oriented practices. Soil erodibility (K-factor) also showed a tendency toward lower values under the ST system, which is consistent with its slightly higher SOC content. These observations align with the established body of literature supporting the role of reduced tillage intensity as a key practice for enhancing soil structure and mitigating erodibility (Blake and Hartge, 1986; Lal, 2004 ). Our results from both the conventional (CT) and strip-till (ST) sites demonstrate that long-term, erosion-driven processes have created a distinct soil mosaic, where indicators of soil health – such as aggregate stability (TAD) and erodibility (K-factor) – exhibit pronounced spatial variability (Table 1 , Fig. 3 , Fig. 5 A). This inherent heterogeneity, representing the „cause” of varying soil productivity, poses a fundamental challenge for site-specific management. Crucially, our quantitative analysis established a statistically significant correlation between these field-measured soil properties and remotely-sensed vegetation vigor, as captured by NDVI (Tables 2 , 3 , Fig. 5 (B-D). The visual comparison further confirmed this strong correspondence: areas characterized by lower aggregate stability and higher soil erodibility consistently exhibited lower NDVI values, particularly during early crop growth stages (Fig.s 5 (B-D), Supplementary Material, Fig. S6). This confirms that NDVI, a readily available tool in most PA platforms, can serve as a reliable proxy for the spatial patterns of soil degradation. However, it is essential to recognize that NDVI reflects the symptom (i.e., reduced vegetation vigor) rather than the underlying causes of limited productivity, which may include not only poor soil structure and high erodibility but also related limitations in water and nutrient availability. The observed persistence of these spatial NDVI patterns, even in post-harvest imagery, further suggests that they are strongly governed by stable, inherent site conditions – primarily soil properties shaped by topography and long-term erosion processes – rather than by short-term crop management factors. This fundamental distinction between a detectable symptom and its complex underlying cause frames the central challenge investigated in this study: are commercial PA platforms capable of bridging this diagnostic gap? 4.2. Uncovering the „Diagnostic Void”: A Systemic Failure of Commercial Ag-Tech The empirical investigation of commercial PA platforms revealed a profound disconnect between the sophisticated diagnostic capabilities needed for science-based soil management and the actual functionalities available to users. However, it is important to first acknowledge the positive trends within this dynamic market. Many platforms are increasingly focused on creating interconnected ecosystems. A prime example is the JDOC , which acts as a central hub connecting to nearly 100 third-party software partners (see Supplementary Material, S2). This trend towards interoperability for operational data is a significant step forward, allowing for a more integrated flow of information across different brands and services. Despite this progress in connectivity, our analysis focused on a more fundamental question: do these platforms, individually or as an ecosystem, provide the necessary tools for causal soil health diagnosis? Our comprehensive analysis of 34 distinct PA platforms revealed a striking absence of dedicated tools for diagnosing the erosion-related causes underlying NDVI variability patterns. Despite the platforms' advanced capabilities for vegetation monitoring and spatial analysis, none offered built-in functionalities for assessing key soil structure parameters such as aggregate stability (TAD) or quantitative erosion risk modeling through established metrics like the K-factor (Table 4 ). This gap persists even across platforms with declared focus on sustainability or advanced analytics, suggesting that the absence of these tools is not merely an oversight but reflects underlying systemic priorities within the commercial PA ecosystem. The contrast between platforms' advanced vegetation monitoring capabilities and their rudimentary soil diagnostic tools is particularly striking. While platforms like GeoPard Agriculture demonstrate powerful multi-layer geospatial analysis capabilities for vegetation indices, they lack dedicated modules for soil structure or erosion modeling. This forces users toward a „bring your own algorithm” approach, requiring custom script development (e.g., in Python) for basic soil analysis functions. Such requirements place an extreme knowledge requirements on the average practitioner and effectively exclude those without programming expertise from conducting advanced soil diagnostics, despite the availability of suitable scientific methodologies (see Supplementary Material, Fig. S5C). This functional limitation becomes particularly problematic when viewed against the backdrop of our field findings. The persistence of erosion-related soil degradation patterns even under conservation management systems (Fig. 5 , Supplementary material, Fig. S6) demonstrates that vegetation indices alone provide insufficient diagnostic information for sustainable management decisions. The platforms' reliance on easily accessible proxy data, particularly NDVI, creates a management paradigm focused on treating symptoms rather than addressing the underlying pedological causes of field variability. This symptom-based approach may inadvertently guide users toward agronomically unsound decisions, such as applying additional nitrogen inputs to areas where poor soil structure or low water-holding capacity are the primary limiting factors. Furthermore, even the presentation of basic proxy data like NDVI is fraught with usability and transparency issues. Our qualitative comparison revealed a significant lack of standardization in how platforms visualize the same underlying satellite data. Users are faced with disparate outputs, making direct, cross-platform comparison difficult and potentially misleading ( Fig. 5 ). These outputs range from high-contrast, pixelated maps of „ Contrast NDVI” ( OneSoil , Fig. 5 B ), to the differently processed and classified NDVI map from SatAgro ( Fig. 5 D ), and the smoothed, low-contrast surfaces of the EVI2, which GeoPard Agriculture offers as its primary vigor indicator ( Fig. 5 C ). Such „black box algorithm” problem is even more acute when platforms offer „inferred” data, which our verification showed can be unreliable ( AgriCircle ), eroding user trust. The practical consequence of this „diagnostic void” is most evident in the VRA modules (see Supplementary Material, Fig. S7). The risk associated with simplified, proxy-based VRA strategies is amplified by the opaque and contradictory logic offered to the user. Our empirical testing of the SatAgro platform provides a stark illustration. The platform requires the user to generate a VRA prescription by choosing between just two opposing quantitative strategies: „Directly proportional” or „Inversely proportional”. This forces the user into a simplistic, binary choice without any diagnostic support to determine if a low-NDVI zone is caused by a temporary nutrient deficiency or a permanent physical limitation. This critical diagnostic decision between two fundamentally different agronomic philosophies is left entirely to the user, who is given contradictory options based on the same superficial data, creating a scenario of high risk for both farm economics and environmental health. This symptom-based approach may inadvertently guide users toward agronomically unsound decisions, such as applying additional nitrogen inputs to areas where poor soil structure or low water-holding capacity are the primary limiting factors. For the farmer, such a misdiagnosis leads to direct economic losses through wasted inputs and unrealized yield potential. Simultaneously, it poses significant environmental risks, including nutrient runoff which can contribute to the eutrophication of surface waters – a critical issue in lake-rich landscapes like the one studied. Beyond technical limitations, our investigation revealed a deeper, conceptual constraint rooted in the underlying design logic of most commercial PA platforms. These systems are not built primarily as flexible scientific analysis tools, but as integrated farm management environments. As a result, accessing even basic analytical functions often requires full operational context – such as crop type, machinery data, or management plans – effectively constraining the user to an „operation-first” workflow. This design logic poses a significant barrier to hypothesis-driven evaluation and impedes the integration of scientific soil data by users who are not operating within a full-season management framework. 4.2.1 . Barriers to Scientific Data Integration The second dimension of the diagnostic void concerns the systematic exclusion of pathways for integrating user-generated, field-verified soil data. Our empirical testing revealed that the inability to import standard raster formats like GeoTIFF – the lingua franca for spatial soil science – represents a critical obstacle across most platforms. This technical limitation is particularly problematic given the abundance of high-resolution soil data available through modern sensing technologies and the growing body of research demonstrating the integration of vegetation indices with field-verified soil data using machine learning models (Bantchina et al., 2024 ; Cai et al., 2019 ; Vedeneeva and Koshelev, 2023 ). These technical barriers are compounded by educational and cognitive obstacles. Even when data upload is technically possible, the complexity of interfaces, cryptic error messages (observed in AGMRI, Granular ), and restrictive data structure requirements effectively block the integration of field-verified soil information (see Supplementary Material, Fig. S4). This confirms that the knowledge barrier identified in earlier studies persists, with data entry, conversion, and visualization in farm software remaining too complex for most users, particularly when working with scientific soil data (Knapic, 2022 ; Padhiary et al., 2024 ; Robinson et al., 2023 ). The economic dimension of this barrier is equally significant. Commercial models increasingly rely on subscription-based access, creating economic lock-ins that hinder experimentation with scientific data tools. Platforms like GeoPard Agriculture link critical analytical features to monthly payments, making long-term data sovereignty dependent on continued financial engagement (see Supplementary Material, Fig. S3). This pricing structure particularly affects small and medium-sized farms, which are often priced out of access to advanced PA tools or cloud-based analytics services (Knapic, 2022 ; Padhiary et al., 2024 ; Pandeya et al., 2025 ). 4.2.2. The Paradox of Technological Sophistication The existence of this diagnostic void presents a fundamental paradox within the commercial PA ecosystem. While platforms demonstrate remarkable sophistication in exchanging complex, proprietary machine data across extensive partner networks, they remain unable to integrate standardized scientific data formats. This „GeoTIFF paradox” highlights that the technical challenges are not insurmountable but rather reflect deliberate design priorities that favor operational data exchange over scientific data integration. The paradox becomes even more pronounced when contrasted with the vibrant research landscape in soil health assessment. The scientific community actively develops advanced solutions using in-situ and proximal soil sensing (Loria et al., 2024 ), remote sensing applications (Arab et al., 2024 ; Meinen and Robinson, 2020 ), and machine learning approaches for mapping key soil indicators (Bantchina et al., 2024 ; Baseca et al., 2019 ; Jana et al., 2024 ; Medeiros et al., 2025 ). The discrepancy between available scientific knowledge and platform capabilities suggests that the challenge lies not in technical feasibility but in the translation and deployment of this knowledge into commercially viable and user-accessible tools. However, the very existence of these extensive ecosystems, such as the one surrounding JDOC , also represents a significant opportunity. While our study identified a systemic „blind spot” regarding soil physical health tools, this interconnected structure could facilitate rapid, widespread adoption of new solutions. If even one partner within this network were to develop and validate a robust module for erosion risk assessment or soil structure analysis, it could potentially be made available to the entire user base of the ecosystem. This suggests that bridging the science-practice gap may not require every platform to build these tools from scratch, but could instead be achieved through the strategic development of specialized, interoperable applications within these existing commercial networks. 4.2.3. Implications for Science-Based Management The „diagnostic void” identified in commercial PA platforms carries significant implications for the implementation of science-based soil management strategies. By limiting users to vegetation indices as primary diagnostic tools, platforms may inadvertently perpetuate management approaches that fail to address the root causes of soil degradation. This limitation is particularly problematic in landscapes shaped by historical erosion, where the legacy effects of soil degradation involve profound, often persistent changes to soil profiles that are not fully mitigated by short- to medium-term improvements in management practices (Berhe, 2019 ; Lal, 2004 ). The absence of diagnostic capabilities and restricted data integration pathways effectively forces users into a reactive, symptom-based management paradigm rather than enabling proactive, science-based approaches to soil health. This limitation undermines the potential for PA technologies to contribute meaningfully to sustainable agriculture and may even exacerbate existing environmental challenges if users are guided toward inappropriate management responses based on incomplete diagnostic information (Carolan, 2020 ). The current state of commercial PA platforms thus represents a critical bottleneck in the translation of scientific soil knowledge into practical management applications. Addressing this diagnostic void will require fundamental changes in platform design priorities, business models, and user interface development to bridge the gap between scientific capabilities and commercial implementation. 4.3. Environmental and Economic Implications of the Diagnostic Void The identified diagnostic void is not merely an academic concern; it carries significant risks for both farm profitability and environmental health. By limiting users to symptom-based diagnostics (i.e., NDVI maps), platforms inadvertently encourage management decisions that can be both economically inefficient and ecologically harmful. For instance, a farmer observing a low-vigor zone may be guided to apply more nitrogen fertilizer, following a „fix-the-symptom” logic. However, as our field data shows, such zones are often limited by poor soil structure and low water-holding capacity, not nutrient deficiency. In this scenario, the applied fertilizer is not utilized by the crop, leading to : Direct economic loss for the farmer through wasted inputs and unrealized yield potential. Increased environmental pollution, as excess nitrogen leaches into groundwater or contributes to greenhouse gas emissions (N₂O), a particularly pressing issue in nitrogen-vulnerable zones like our study area. Perpetuation of soil degradation, as the root cause of the problem remains unaddressed, potentially worsening with each season. This symptom-driven paradigm, hardwired into the current PA ecosystem, thus undermines the core promise of precision agriculture – to enhance efficiency and sustainability. It creates a high-risk environment where farmers are given sophisticated tools to make potentially wrong decisions with precision, locking them into a cycle of reactive management that fails to build long-term soil health and resilience. 4.4. Limitations, Implications, and Future Outlook This study has several limitations that frame the context of its findings and highlight critical directions for future research. The comparative analysis of tillage systems was based on an unbalanced sampling design, meaning conclusions regarding the spatial effects of strip-tillage should be considered preliminary. Similarly, our evaluation of PA platforms, based on trial versions, represents a „snapshot” of a highly dynamic market. While we argue that the identified systemic barriers are fundamental, their specific manifestations may evolve. These limitations underscore a broader challenge: the scarcity of independent, systematic evaluations of commercial PA tools, which creates an information vacuum for end-users (Table 4 ). The implications of our findings are significant. The identified „diagnostic void” forces users into a reactive, symptom-based management paradigm, which is economically inefficient and environmentally risky, particularly in landscapes with a strong „legacy of erosion” (Berhe, 2019 ; Lal, 2004 ). This undermines the core promise of PA to deliver truly precise, science-based soil management. Bridging this gap requires a concerted effort. Based on our findings, we propose several key recommendations for the future: Enhancing Data and Model Integration : Platform developers must prioritize robust import tools for scientific formats (GeoTIFF, complex Shapefiles) and begin embedding validated scientific models (e.g., for K-factor calculation) directly into their workflows, moving beyond the „bring your own algorithm” approach. Exploring Open and Collaborative Models : The potential of open-source tools (QGIS, R, Python) as a transparent and flexible alternative should be further explored, despite their current expertise requirements (Zagórda and Walczykova, 2018). Furthermore, closer academia-industry collaboration is crucial. PA companies should facilitate independent validation through APIs or research accounts (as exemplified by GeoPard ), while researchers must work to make their models „platform-ready”. Investigating Ecosystem Interoperability : A key area for future research lies in the practical reality of ecosystem interoperability. Our study highlights the potential of large partner networks like that of JDOC . Future studies should empirically investigate the end-to-end data workflow within these ecosystems to determine if they offer genuine, holistic decision support or create a more fragmented digital landscape. Developing User-Centric and Inclusive Tools : Finally, the focus must shift towards designing tools that empower users to understand the causes of in-field variability, not just the symptoms. The direct benefit for farmers would be the ability to make more targeted and cost-effective management decisions – for example, allocating resources for soil amendments in degraded zones instead of over-applying fertilizers that the crop cannot utilize. This is particularly crucial for small and medium-sized farms, which are disproportionately affected by both soil degradation and the barriers of current PA tools (Knapic, 2022 ; Padhiary et al., 2024 ). By creating more inclusive and genuinely diagnostic systems that demonstrate a clear return on investment through improved input efficiency and long-term soil productivity, a true market demand for sustainable soil management can be fostered. Addressing these challenges is essential if digital agriculture is to move beyond its current limitations and genuinely contribute to a more sustainable and resilient food system. 5. Conclusions This study provided a unique, field-verified assessment of the gap between the scientific understanding of soil physical degradation and the practical capabilities of commercial Precision Agriculture (PA) platforms. Conducted within a representative hummocky moraine landscape, the study leads to the following key conclusions: There is a strong, statistically significant relationship between field-verified indicators of soil degradation and remotely sensed vegetation indices (NDVI). This confirms that NDVI, while a powerful proxy for crop performance, is fundamentally a symptom indicator. Used in isolation – as is common practice in commercial PA platforms – it cannot diagnose the underlying causes of poor soil health, making it a powerful yet potentially misleading tool for science-based management. Our empirical investigation revealed a systemic „diagnostic void” across the commercial PA ecosystem. Current platforms are fundamentally limited in their ability to support causal diagnosis of soil degradation, characterized by a near-total absence of built-in tools for assessing soil structure or erosion risk. This void is reinforced by a wall of technical, usability, and economic barriers that prevent the integration of user-generated scientific soil data. This „diagnostic void” is not an accidental oversight but a direct consequence of a market failure where business priorities for short-term returns and proprietary data ecosystems trump the need for genuine, science-based soil stewardship. This focus perpetuates a symptom-driven management paradigm that is not just ineffective, but potentially economically wasteful and environmentally hazardous, especially in sensitive, erosion-prone landscapes. In conclusion, realizing the true potential of digital agriculture requires a fundamental reorientation of the entire PA sector. Moving forward, the focus must shift from simply treating visible symptoms to empowering users to manage the complex causes of soil degradation. This demands a move towards open standards, genuine interoperability, and the integration of validated scientific models. Without this paradigm shift, precision agriculture risks becoming a tool for precisely managing our way into continued soil degradation, rather than a solution for building a resilient and sustainable agricultural future. Abbreviations CSV Comma Separated Values CT Conventional Tillage DEM Digital Elevation Model EPIC Erosion-Productivity Impact Calculator EVI2 Enhanced Vegetation Index 2 FMIS Farm Management Information System FMIS/ERP Farm Management Information System / Enterprise Resource Planning GEE Google Earth Engine GIS Geographic Information System HCP Hardware-Centric Platform JDOC John Deere Operations Center K-factor Soil Erodibility Factor MWDdry Mean Weight Diameter (dry) NDVI Normalized Difference Vegetation Index PA Precision Agriculture ROI Return on Investment RUSLE Revised Universal Soil Loss Equation SM Supplementary Material SOC Soil Organic Carbon ST Strip-Till TAD Time of Aggregate Destruction USLE Universal Soil Loss Equation VRA Variable Rate Application WRB World Reference Base for Soil Resources References Ait Issad, H., Aoudjit, R., Rodrigues, J.J.P.C., 2019. A comprehensive review of Data Mining techniques in smart agriculture. Eng. Agric. Environ. Food 12, 511–525. https://doi.org/10.1016/j.eaef.2019.11.003 Arab, S., Easson, G., Ghaffari, Z., 2024. Integration of Sentinel-1A Radar and SMAP Radiometer for Soil Moisture Retrieval over Vegetated Areas. Sensors 24, 2217. https://doi.org/10.3390/s24072217 Arias-Navarro, C., Baritz, R., Jones, A., European Commission, European Environment Agency (Eds.), 2024. The state of soils in Europe: fully evidenced, spatially organised assessment of the pressures driving soil degradation. Publications Office, Luxembourg. https://doi.org/10.2760/5897030 Arsenoaia, V.-N., Rațu, R.-N., Veleșcu, I.-D., Țenu, I., 2023. Field tests of a John Deere harvester for the purpose of production maps achievement. Babos, D.V., Tadini, A.M., De Morais, C.P., Barreto, B.B., Carvalho, M.A.R., Bernardi, A.C.C., Oliveira, P.P.A., Pezzopane, J.R.M., Milori, D.M.B.P., Martin-Neto, L., 2024. Laser-induced breakdown spectroscopy (LIBS) as an analytical tool in precision agriculture: Evaluation of spatial variability of soil fertility in integrated agricultural production systems. CATENA 239, 107914. https://doi.org/10.1016/j.catena.2024.107914 Bantchina, B.B., Qaswar, M., Arslan, S., Ulusoy, Y., Gündoğdu, K.S., Tekin, Y., Mouazen, A.M., 2024. Corn yield prediction in site-specific management zones using proximal soil sensing, remote sensing, and machine learning approach. Comput. Electron. Agric. 225, 109329. https://doi.org/10.1016/j.compag.2024.109329 Barrios, E., 2007. Soil biota, ecosystem services and land productivity. Ecol. Econ., Special Section – Ecosystem Services and Agriculture 64, 269–285. https://doi.org/10.1016/j.ecolecon.2007.03.004 Baseca, C.C., Sendra, S., Lloret, J., Tomas, J., 2019. A smart decision system for digital farming. Agronomy 9. https://doi.org/10.3390/agronomy9050216 Basir, Md.S., Buckmaster, D., Raturi, A., Zhang, Y., 2024. From pen and paper to digital precision: a comprehensive review of on-farm recordkeeping. Precis. Agric. 25, 2643–2682. https://doi.org/10.1007/s11119-024-10172-7 Basso, B., Dumont, B., Cammarano, D., Pezzuolo, A., Marinello, F., Sartori, L., 2016. Environmental and economic benefits of variable rate nitrogen fertilization in a nitrate vulnerable zone. Sci. Total Environ. 545–546, 227–235. https://doi.org/10.1016/j.scitotenv.2015.12.104 Berhe, A.A., 2019. Chapter 3 – Drivers of soil change, in: Busse, M., Giardina, C.P., Morris, D.M., Page-Dumroese, D.S. (Eds.), Developments in Soil Science, Global Change and Forest Soils. Elsevier, pp. 27–42. https://doi.org/10.1016/B978-0-444-63998-1.00003-3 Borrero, J.D., Mariscal, J., 2022. A Case Study of a Digital Data Platform for the Agricultural Sector: A Valuable Decision Support System for Small Farmers. Agriculture 12, 767. https://doi.org/10.3390/agriculture12060767 Cai, Y., Guan, K., Lobell, D., Potgieter, A.B., Wang, S., Peng, J., Xu, T., Asseng, S., Zhang, Y., You, L., Peng, B., 2019. Integrating satellite and climate data to predict wheat yield in Australia using machine learning approaches. Agric. For. Meteorol. 274, 144–159. https://doi.org/10.1016/j.agrformet.2019.03.010 Carolan, M., 2020. Acting like an algorithm: digital farming platforms and the trajectories they (need not) lock-in. Agric. Hum. Values 37, 1041–1053. https://doi.org/10.1007/s10460-020-10032-w Dexter, A.R., Richard, G., Arrouays, D., Czyż, E.A., Jolivet, C., Duval, O., 2008. Complexed organic matter controls soil physical properties. Geoderma 144, 620–627. https://doi.org/10.1016/j.geoderma.2008.01.022 Escribà-Gelonch, M., Liang, S., van Schalkwyk, P., Fisk, I., Long, N.V.D., Hessel, V., 2024. Digital Twins in Agriculture: Orchestration and Applications. J. Agric. Food Chem. 72, 10737–10752. https://doi.org/10.1021/acs.jafc.4c01934 Hillel, D., 2013. Introduction to Soil Physics. Academic Press. Horn, R., Fleige, H., 2009. Risk assessment of subsoil compaction for arable soils in Northwest Germany at farm scale. Soil Tillage Res., Soil Management for Sustainability 102, 201–208. https://doi.org/10.1016/j.still.2008.07.015 Huber, F., Markward, D., 2021. AgriCircle: Innovating Agricultural Ecosystems, in: Gassmann, O., Ferrandina, F. (Eds.), Connected Business: Create Value in a Networked Economy. Springer International Publishing, Cham, pp. 325–331. https://doi.org/10.1007/978-3-030-76897-3_21 Institute for Environment and Sustainability (Joint Research Centre), Montanarella, L., Rusco, E., Tóth, G., 2008. Threats to soil quality in Europe. Publications Office of the European Union. Jacob, T., n.d. Evaluating Precision Agriculture Using Open Software Platforms. Jana, S., Chatterjee, D., Pal, N., Pal, K., Roy, K., Bashak, S., 2024. AI in Soil Health Monitoring: A Data-Driven Approach. Int. J. Res. Appl. Sci. Eng. Technol. 12, 1327–1335. https://doi.org/10.22214/ijraset.2024.64871 Keesstra, S.D., Bouma, J., Wallinga, J., Tittonell, P., Smith, P., Cerdà, A., Montanarella, L., Quinton, J.N., Pachepsky, Y., van der Putten, W.H., Bardgett, R.D., Moolenaar, S., Mol, G., Jansen, B., Fresco, L.O., 2016. The significance of soils and soil science towards realization of the United Nations Sustainable Development Goals. SOIL 2, 111–128. https://doi.org/10.5194/soil-2-111-2016 Knapic, R., 2022. Who’s Running the Farm? How a Sample of Kansas Farmers View Their Usage of Agricultural Operating Systems (M.A.). Kottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World Map of the Köppen-Geiger climate classification updated. Meteorol. Z. 259–263. https://doi.org/10.1127/0941-2948/2006/0130 Lal, R., 2004. Soil Carbon Sequestration Impacts on Global Climate Change and Food Security. Science 304, 1623–1627. https://doi.org/10.1126/science.1097396 Lehmann, J., Bossio, D.A., Kögel-Knabner, I., Rillig, M.C., 2020. The concept and future prospects of soil health. Nat. Rev. Earth Environ. 1, 544–553. https://doi.org/10.1038/s43017-020-0080-8 Loria, N., Lal, R., Chandra, R., 2024. Handheld In Situ Methods for Soil Organic Carbon Assessment. Sustainability 16, 5592. https://doi.org/10.3390/su16135592 Lykhovyd, P., Vozhehova, R., Averchev, O., 2024. Using Remote Sensing Normalised Difference Vegetation Index to Recognise Irrigated Croplands via Agroland Classifier Application. Visnyk V N Karazin Kharkiv Natl. Univ. Ser. Geol. Geogr. Ecol. 2024, 223–233. https://doi.org/10.26565/2410-7360-2024-61-18 Medeiros, B.M., Cândido, B., Jimenez, P.A.J., Avanzi, J.C., Silva, M.L.N., 2025. UAV-Based Soil Water Erosion Monitoring: Current Status and Trends. Drones 9, 305. https://doi.org/10.3390/drones9040305 Meinen, B.U., Robinson, D.T., 2020. Mapping erosion and deposition in an agricultural landscape: Optimization of UAV image acquisition schemes for SfM-MVS. Remote Sens. Environ. 239, 111666. https://doi.org/10.1016/j.rse.2020.111666 Padhiary, M., Roy, P., Dey, P., Sahu, B., 2024. Harnessing AI for Automated Decision- Making in Farm Machinery and Operations: Optimizing Agriculture. pp. 249–282. https://doi.org/10.4018/979-8-3693-6230-3.ch008 Paleari, S., 2017. Is the European Union protecting soil? A critical analysis of Community environmental policy and law. Land Use Policy 64, 163–173. https://doi.org/10.1016/j.landusepol.2017.02.007 Panagos, P., Borrelli, P., Poesen, J., Ballabio, C., Lugato, E., Meusburger, K., Montanarella, L., Alewell, C., 2015. The new assessment of soil loss by water erosion in Europe. Environ. Sci. Policy 54, 438–447. https://doi.org/10.1016/j.envsci.2015.08.012 Pandeya, S., Gyawali, B.R., Upadhaya, S., 2025. Factors Influencing Precision Agriculture Technology Adoption Among Small-Scale Farmers in Kentucky and Their Implications for Policy and Practice. Agriculture 15, 177. https://doi.org/10.3390/agriculture15020177 Pham, X., Stack, M., 2018. How data analytics is transforming agriculture. Bus. Horiz. 61, 125–133. https://doi.org/10.1016/j.bushor.2017.09.011 Pierce, F.J., Nowak, P., 1999. Aspects of Precision Agriculture. Adv. Agron. 67, 1–85. https://doi.org/10.1016/S0065-2113(08)60513-1 Rabot, E., Wiesmeier, M., Schlüter, S., Vogel, H.-J., 2018. Soil structure as an indicator of soil functions: A review. Geoderma 314, 122–137. https://doi.org/10.1016/j.geoderma.2017.11.009 Radziuk, H., Świtoniak, M., 2022. The Effect of Erosional Transformation of Soil Cover on the Stability of Soil Aggregates within Young Hummocky Moraine Landscapes in Northern Poland. Agronomy 12, 2595. https://doi.org/10.3390/agronomy12112595 Radziuk, H., Świtoniak, M., 2021. Soil erodibility factor (K) in soils under varying stages of truncation. Soil Sci. Annu. https://doi.org/10.37501/soilsa/134621 Radziuk, H., Świtoniak, M., Nowak, M., 2021. Microscale spatial variation of soil erodibility factor (K) in a young hummocky moraine landscape in Northern Poland. Bull. Geogr. Phys. Geogr. Ser. 21, 5–16. https://doi.org/10.2478/bgeo-2021-0005 Revut, Ivan, 1972. Soil physics. Kolos, Leningrad. Robinson, S.V.J., Schwinghamer, T., Cárcamo, H., Galpern, P., 2023. Precision agricultural data and ecosystem services: Can we put the pieces together? Ecol. Solut. Evid. 4, e12271. https://doi.org/10.1002/2688-8319.12271 Rząsa, S., Owczrzak, W., 2004. Struktura gleb mineralnych -. Wydawnictwo Akademii Rolniczej im. Augusta Cieszkowskiego, Poznań. Sheng, Y., Sun, Z., Lu, S., Omasa, K., 2024. Ratio of physical model parameters can retrieve aggregate size from different types of soil in cultivated regions. Soil Tillage Res. 244, 106262. https://doi.org/10.1016/j.still.2024.106262 Six, J., Bossuyt, H., Degryze, S., Denef, K., 2004. A history of research on the link between (micro)aggregates, soil biota, and soil organic matter dynamics. Soil Tillage Res., Advances in Soil Structure Research 79, 7–31. https://doi.org/10.1016/j.still.2004.03.008 Smith, P., House, J.I., Bustamante, M., Sobocká, J., Harper, R., Pan, G., West, P.C., Clark, J.M., Adhya, T., Rumpel, C., Paustian, K., Kuikman, P., Cotrufo, M.F., Elliott, J.A., McDowell, R., Griffiths, R.I., Asakawa, S., Bondeau, A., Jain, A.K., Meersmans, J., Pugh, T.A.M., 2016. Global change pressures on soils from land use and management. Glob. Change Biol. 22, 1008–1028. https://doi.org/10.1111/gcb.13068 Soil Health Improvement | FAO SOILS PORTAL | Food and Agriculture Organization of the United Nations [WWW Document], n.d. URL https://www.fao.org/soils-portal/soil-degradation-restoration/soil-health-improvement/en/ (accessed 6.4.25). Solon J., Borzyszkowski J., Bidіasik M., Richling A., Badora K., Balon J., Brzeziсska-Wуjcik T., Chabudziсski Ј., Dobrowolski R., Grzegorczyk I., Jodіowski M., Kistowski M., Kot R., Lechnio J., Macias A., 2018. Physico-geographical mesoregions of Poland – modified version of J. Kondracki’s regionalisation. Geogr. Pol. 91. https://doi.org/10.7163/GPol.0115 Świtoniak, M., 2014. Use of soil profile truncation to estimate influence of accelerated erosion on soil cover transformation in young morainic landscapes, North-Eastern Poland. CATENA 116, 173–184. https://doi.org/10.1016/j.catena.2013.12.015 Świtoniak, M., Mroczek, P., Bednarek, R., 2016. Luvisols or Cambisols? Micromorphological study of soil truncation in young morainic landscapes – Case study: Brodnica and Chełmno Lake Districts (North Poland). CATENA 137, 583–595. https://doi.org/10.1016/j.catena.2014.09.005 Świtoniak M., Nowak M., Radziuk H., 2020. Low-altitude photogrammetry in studies of soil cover variability of areas transformed by anthropogenic denudation., in: Współczesne Problemy i Kierunki Badawcze w Geografii. Instytut Geografii i Gospodarki Przestrzennej Uniwersytetu Jagiellońskiego, Kraków, p. S. 81-100. Vedeneeva, V.A., Koshelev, A.V., 2023. Analysis of the seasonal dynamics of the NDVI index of potato agrocenosis in the conditions of the dry steppe zone of the Volgograd region. Res. Crops 24, 366–372. https://doi.org/10.31830/2348-7542.2023.ROC-935 Williams, J.R., 1990. The Erosion-Productivity Impact Calculator (EPIC) Model: A Case History. Philos. Trans. Biol. Sci. 329, 421–428. Zagorda, M., Walczykova, M., 2018. The application of various software programs for mapping yields in precision agriculture, in: SzelagSikora, A. (Ed.), CONTEMPORARY RESEARCH TRENDS IN AGRICULTURAL ENGINEERING, BIO Web of Conferences. Presented at the Conference on Contemporary Research Trends in Agricultural Engineering, E D P Sciences, Cedex A, p. 01018. https://doi.org/10.1051/bioconf/20181001018 Additional Declarations The authors declare no competing interests. Supplementary Files SupplementaryArticleRadziukwitoniak2025.pdf Supplementary materials Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7541765","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":510644095,"identity":"8292b694-69de-40d2-b2a1-57b1a9136dfa","order_by":0,"name":"Hanna Radziuk","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYBACAygtx8ADotiAmJ1ILcYILcxEaklsIFqLOfvZYw9+MNxL33DmjAHDh7LDDOaEtFj25KUb9jAU524422PAOOPcYQbLZkIOO5BjJsHDkJC74TzvBmbetsMMBocJaTn/xkzyD0NCugFIy1+itNzIMZMG2pJgcLZ3AzMjcVremEnLGCQYzjxz/sPBnnPpPIT9cj7HTPJNRYI835m0xAc/yqzlzNkbCOiBaIRQB4CYxwCfQrzaR8EoGAWjYBTAAQC3yD6/IoO9YAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-5279-2175","institution":"Kazimierz Wielki University in Bydgoszcz","correspondingAuthor":true,"prefix":"","firstName":"Hanna","middleName":"","lastName":"Radziuk","suffix":""},{"id":510644096,"identity":"cc036776-b184-49ed-903a-fdc1d44369a5","order_by":1,"name":"Marcin Świtoniak","email":"","orcid":"https://orcid.org/0000-0002-9907-7088","institution":"Nicolaus Copernicus University in Toruń","correspondingAuthor":false,"prefix":"","firstName":"Marcin","middleName":"","lastName":"Świtoniak","suffix":""}],"badges":[],"createdAt":"2025-09-05 07:36:54","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7541765/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7541765/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90842820,"identity":"0c0d305d-dd54-4ab1-80df-6905a8475b2a","added_by":"auto","created_at":"2025-09-08 21:16:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":728678,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSchematic overview of the research methodology.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7541765/v1/14f18aa548ca0d040572276b.png"},{"id":90843739,"identity":"4d3cca93-f6ac-4550-ac83-3da0ff33abab","added_by":"auto","created_at":"2025-09-08 21:40:41","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":2881039,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7541765/v1/d24265c369cd26be39e252c6.png"},{"id":90842825,"identity":"5f08f0fe-1686-44c8-aafc-c802db502e12","added_by":"auto","created_at":"2025-09-08 21:16:41","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":529104,"visible":true,"origin":"","legend":"\u003cp\u003eSoil map of study areas under \u0026nbsp;conventional tillage (A) and strip-till (B) systems\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7541765/v1/6d3dfa42657a534bc2f61f79.png"},{"id":90842824,"identity":"cc4990d5-05ea-432b-b46c-15fe8eecdddc","added_by":"auto","created_at":"2025-09-08 21:16:41","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":297248,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eScatter plot showing the relationship between TAD, sec and NDVI for the 66 sampling points in the CT field (June 2022-06-21, Google Earth Engine)\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7541765/v1/7598988ee77591d617fd7058.png"},{"id":90842833,"identity":"8fbeb44b-4370-4228-9a99-f744d9d14623","added_by":"auto","created_at":"2025-09-08 21:16:41","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":2965570,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of vegetation indices from three different PA platforms (5B-D) with a field-verified map of aggregate stability (TAD, sec) (5A) for the CT field, based on Sentinel-2 imagery (10m resolution) for the same date (2022-06-21). Despite the common data source, the platforms exhibit significant differences in the specific index offered and visualization style used, including: contrast NDVI in OneSoil (5B); a smoothed EVI2, which GeoPard (5C) provides as its primary vegetation vigor indicator; and a classified NDVI map in SatAgro (5D)\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7541765/v1/ea2093d14ca50a3564694d21.png"},{"id":90843797,"identity":"95819fde-e812-4ec8-a81a-e55b39ffd72b","added_by":"auto","created_at":"2025-09-08 21:48:47","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":12394060,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7541765/v1/f56f7e55-209a-45be-8778-4bf68d076a6a.pdf"},{"id":90842986,"identity":"6bb70ca1-bf7a-4b4e-be2b-a49063ca4c6e","added_by":"auto","created_at":"2025-09-08 21:24:41","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2373306,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary materials\u003c/p\u003e","description":"","filename":"SupplementaryArticleRadziukwitoniak2025.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7541765/v1/741db558fac1996938399316.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eThe Diagnostic Void: A Systemic Failure of Commercial Precision Agriculture Platforms to Integrate Science-Based Soil Health Indicators\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eSoil degradation is a major threat to agricultural productivity and environmental quality worldwide (Lal, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Arias-Navarro et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). In response, the modern concept of soil health \u0026ndash; defined as the soil\u0026rsquo;s capacity to function as a living ecosystem \u0026ndash; has become central to sustainable agriculture (Keesstra et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Lehmann et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This concept encompasses not only chemical properties but also physical and biological characteristics, such as soil structure and microbiome activity, which are essential for supporting ecosystem services and ensuring long-term productivity (Barrios, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Smith et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eTwo forms of soil degradation \u0026ndash; water erosion and soil compaction \u0026ndash; are particularly widespread in Europe, affecting over 60% of agricultural land (Arias-Navarro et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Water erosion impacts 24% of EU land, occurring at rates that exceed natural soil formation (Panagos et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), whereas compaction affects 23% of agricultural soils, especially in subsoil (plowpan) layer (Horn and Fleige, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). The susceptibility of soils to these processes is closely linked to their physical properties, particularly aggregate stability and overall structure (Dexter et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Lehmann et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). A stable soil structure enhances water infiltration, aeration, and root growth, while its degradation increases runoff and erosion risk.\u003c/p\u003e\u003cp\u003eAssessing indicators of soil physical health, such as aggregate stability and erodibility (e.g., the K-factor), is therefore essential for effective soil management (Radziuk et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Sheng et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). This is particularly important in heterogeneous landscapes, such as hummocky moraine terrain, where complex topography and spatial variability of soils create challenges for both monitoring and management (Świtoniak, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Arias-Navarro et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003ePrecision Agriculture (PA) technologies have been promoted as a solution for managing microscale soil variability (Basso et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Pierce and Nowak, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). PA platforms provide tools for spatial analysis and variable rate application (VRA), often relying on proxy satellite data, such as the NDVI, to identify low-performing zones (Babos et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zagorda and Walczykova, 2018). This raises a central question for science-based soil management: while NDVI can effectively identify symptom - such as reduced crop vigor \u0026ndash; are commercial PA platforms equipped to help users diagnose the underlying cause (e.g., poor soil physical health) and integrate this diagnostic insight into management decisions? \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eFailure to do so can lead to inefficient resource use and misguided agronomic interventions, resulting in both economic losses for the farmer and negative environmental impacts.\u003c/span\u003e\u003c/p\u003e\u003cp\u003eDespite advances in both soil science and PA platforms, a notable gap remains: no study has systematically evaluated whether commercial PA platforms can bridge the symptom-cause divide by incorporating field-verified data on soil structure and erosion (Lykhovyd et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Closing this gap is crucial for the wider adoption of science-based soil management practices.\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eHowever, the widespread adoption of these platforms raises a critical and underexplored question: are these tools designed to support holistic, science-based soil management, or do their underlying technical and business models inadvertently promote a superficial, symptom-driven approach? We hypothesize that despite their technological sophistication, current commercial PA platforms face systemic limitations in their ability to support causal diagnosis of soil degradation. This study tests this hypothesis by moving beyond declared features to empirically assess the practical capabilities of these platforms in bridging the critical gap between the remotely-sensed symptom (vegetation vigor) and the field-verified cause (soil physical degradation).\u003c/span\u003e\u003c/p\u003e\u003cp\u003eTo investigate this, we address the following key research questions:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eWhat is the nature and strength of the relationship between field-verified soil degradation indicators and remotely-sensed vegetation indices (NDVI)?\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eTo what extent do commercial PA platforms provide built-in tools to diagnose the erosion-related causes of the variability observed in NDVI data?\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eWhat are there the practical possibilities for integrating user-generated, field-verified soil data into these platforms as an alternative pathway for science-based decision-making?\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cp\u003eTo address the research questions outlined in the introduction, a multi-stage methodology was designed to systematically bridge field-verified soil science with an empirical assessment of commercial digital tools. As illustrated in the schematic overview (Fig.\u0026nbsp;1), the research process follows a logical sequence: beginning with the quantification of soil degradation in the field, continuing with an analysis of its manifestation in remote sensing data, and concluding with the evaluation of PA platform capabilities to address these issues. The following sections describe each stage of this integrated approach in detail.\u003c/p\u003e\n\u003cdiv id=\"Sec3\"\u003e\n \u003ch2\u003e2.1 Field and Laboratory Study\u003c/h2\u003e\n \u003cp\u003eThis part of the study was designed to empirically quantify the extent and spatial variability of soil physical degradation within a representative, agriculturally managed landscape. The field-verified data collected during this phase served as the primary scientific input for the subsequent evaluation of PA platforms.\u003c/p\u003e\n \u003cdiv id=\"Sec4\"\u003e\n \u003ch2\u003e2.1.1. Study Areas and Soil Sampling\u003c/h2\u003e\n \u003cp\u003eThe study was conducted in Northern Poland within two separate agricultural fields – Orzechowo (CT) and Sokołowo (ST), located within the Chełmno and Dobrzyń Lakeland respectively (Fig.\u0026nbsp;2A). Both physical-geographical regions are characterized by a hummocky moraine landscape formed during the Weichselian glaciation (Solon et al., 2018), with a mean annual air temperature of approximately 8°C and average yearly precipitation of 604 mm (Kottek et al., 2006).\u003c/p\u003e\n \u003cp\u003eThe two study sites were selected based on their comparable geomorphological settings and parent materials, enabling a robust comparative case study. Due to the variability of the terrain, the soils in both study areas exhibit considerable natural catenal differences in soil-forming processes – most notably, clay illuviation on higher elevations and strong groundwater influence in depressions. A history of intensive agricultural use has led to significant anthropogenic denudation, resulting in a complex mosaic of soils with varying degrees of profile truncation on summits and slopes, and colluvial material accumulation in lower-lying areas (Świtoniak, 2014; Świtoniak et al. 2020; Radziuk and Świtoniak, 2022). This combination of inherent and human-induced spatial variability makes these landscapes an ideal real-world testbed for evaluating the capabilities of precision agriculture (PA) tools. The first site, a 40-ha field near Orzechowo managed under Conventional Tillage (CT) (Fig.\u0026nbsp;2C), has been the subject of intensive previous research focused on the spatial distribution of soil properties and erosion-induced carbon redistribution (Radziuk and Świtoniak, 2022). The second site, a 47-ha field near Sokołowo managed under Strip-Till (ST) (Fig.\u0026nbsp;2B) – has also been previously investigated in the context of conservation tillage and its effects on restoring the physical properties of severely eroded soils (Świtoniak et al., 2020). For both fields, crop rotation follows a typical regional pattern, including winter wheat, winter oilseed rape, and spring wheat, with fertilization applied in accordance with standard agronomic recommendations. This well-established research background provides a robust and well-documented foundation for the present study.\u003c/p\u003e\n \u003cp\u003eTo investigate soil cover and its properties, a representative catena along a hillslope was selected within each study field.\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eIn the CT study field, a comprehensive sampling campaign was performed across four catenae to capture the high spatial variability (16 full soil profiles in 4 catenae and 50 supplementary augering points).\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eIn the ST study field, a focused case study was conducted along a single representative catena, yielding 4 soil profiles with similar geomorphological settings to those in the CT field.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n \u003cp\u003eSoil samples were collected from both fields during autumn 2019. At each profile and augering point, disturbed composite samples were taken from the topsoil (0–30 cm) for the analysis of basic physical and chemical properties. Additionally, at the 16 profiles in the CT field and 4 profiles in the ST field, large undisturbed block samples were carefully extracted from the Ap and subsoil horizons for aggregate stability assessment. The collected point data, combined with geomorphological analysis of a digital elevation model (DEM) and interpretation of high-resolution orthophotomaps, served as the basis for creating the final soil maps in a GIS environment.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec5\"\u003e\n \u003ch2\u003e2.1.2. Laboratory Analyses and Key Indicators\u003c/h2\u003e\n \u003cp\u003eSoil samples collected from both study fields underwent comprehensive laboratory analyses to characterize their physical, chemical, and structural properties. All laboratory analyses were conducted at the Laboratory for Environmental Analysis (Nicolaus Copernicus University in Toruń, Poland), with detailed methodologies described in previously published works (Radziuk and Świtoniak, 2022, 2021; Świtoniak M. et al., 2020).\u003c/p\u003e\n \u003cp\u003eDisturbed samples were used to determine particle size distribution (aerometric method), soil pH (potentiometrically), soil organic carbon (SOC) content (Tiurin method), and calcium carbonate (CaCO3) content (Scheibler method). Bulk density was ascertained using undisturbed core samples (Hillel, 2013; Revut, 1972). Undisturbed block samples from both CT and ST sites were used to assess aggregate stability. Air-dried samples were first dry-sieved to determine the aggregate size distribution and calculate the MWDdry, mm. Subsequently, the water stability of aggregates from the 7–10 mm fraction was quantified by measuring the TAD, s using a static wetting test (Rząsa and Owczrzak, 2004).\u003c/p\u003e\n \u003cp\u003eSoil erodibility was calculated as the K-factor using the Williams' (Williams, 1990) formula from the EPIC model, which is based on soil texture and SOC content (see Supplementary Material, S1, for detailed equations). These two parameters, TAD and K-factor, representing soil structural resilience and erosion susceptibility respectively, formed the core of the scientific dataset (Radziuk and Świtoniak, 2021).\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec6\"\u003e\n \u003ch2\u003e2.1.3. Creation of the Field-Verified Dataset\u003c/h2\u003e\n \u003cp\u003eThe point-based data from the field and laboratory analyses were processed in a GIS environment. This involved integrating the point data with geomorphological analysis of a high-resolution Digital Elevation Model (DEM) and interpretation of orthophotomaps. The final output of this stage was a comprehensive field-verified dataset containing spatial layers and tabular data in standard formats, including GeoTIFF (for interpolated maps of TAD and K-factor), Shapefile (for delineated soil/erosion polygons and sampling points), and CSV/Excel (for raw point data). This dataset represents the complex, high-resolution „ground truth” of soil physical condition used for the subsequent testing of PA platforms.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec7\"\u003e\n \u003ch2\u003e2.2. PA Platform Assessment Methodology\u003c/h2\u003e\n \u003cp\u003eTo critically evaluate the extent to which commercial PA platforms support the integration of advanced soil data – particularly indicators of soil structure and erosion risk – a custom, multi-phase assessment framework was developed for this study. Due to the lack of standardized protocols for evaluating the scientific data handling capabilities of such diverse commercial software, this methodology was designed to systematically bridge a general market overview with empirical, user-centric testing.\u003c/p\u003e\n \u003cdiv id=\"Sec8\"\u003e\n \u003ch2\u003e2.2.1. Phase 1 \u0026amp; 2: Market Survey, Platform Selection, and Rationale for Empirical Testing\u003c/h2\u003e\n \u003cp\u003eThe selection of PA platforms for empirical testing was preceded by a systematic, two-phase market survey. The initial phase involved a comprehensive review of the commercial PA market, identifying 34 platforms based on extensive online searches. To systematically assess their declared capabilities, each platform's publicly available information was evaluated, with a focus on functionalities related to soil data management. Each platform was classified according to five distinct levels of soil-related data handling, ranging to the next levels:\u003c/p\u003e\n \u003cp\u003e(a) No relevant functionality. The platform offers no features related to soil data.\u003cbr\u003e(b) Basic soil data visualization or management. Limited capabilities, such as map overlays of pH or lab results.\u003cbr\u003e(c) Advanced integration and analysis of soil data. Features like multi-layer analytics or analysis of soil texture maps.\u003cbr\u003e(d) Indirect assessment of soil structure or erosion risk. Use of proxy data (e.g., topography) that correlate with soil physical health.\u003cbr\u003e(e) Dedicated tools for erosion modeling. Explicit tools for erosion risk, such as K-factor calculators or risk assessment modules.\u003cbr\u003e\u003c/p\u003e\n \u003cp\u003eThe detailed results of this market survey and functional categorization are provided in the Supplementary Material (Figure S1, Tab. S2).\u003c/p\u003e\n \u003cp\u003eThis survey revealed two key findings that shaped the subsequent research. First, a critical gap became immediately evident: out of the 34 platforms, none offered dedicated, built-in tools for the assessment of soil structure or for quantitative erosion risk modeling. Second, a targeted literature review using Scopus and Web of Science confirmed a significant scarcity of independent academic evaluations for most of these commercial tools (see Supplementary Material, Tab. S1). This lack of both declared functionalities and academic scrutiny provided the critical justification for the empirical testing in Phase 3.\u003c/p\u003e\n \u003cp\u003eBased on this preliminary analysis, a diverse set of nine platforms was purposively selected for in-depth testing. The selection was designed to be representative of the market and was stratified to include platforms from across the functionality levels and different business models, including: hardware-centric ecosystems (e.g., John Deere Operations Center, AgFiniti, AGMRI), major corporate solutions (e.g., Climate FieldView, Granular), specialized data analytics platforms (e.g., GeoPard Agriculture), and widely accessible freemium platforms (e.g., OneSoil, SatAgro). One additional platform (\u003cem\u003eAgriCircle\u003c/em\u003e) was assessed through direct contact due to limited access.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec9\"\u003e\n \u003ch2\u003e2.2.2. Phase 3: Empirical User Experience (UX) Testing\u003c/h2\u003e\n \u003cp\u003eTesting of all nine selected platforms was conducted using a comprehensive, field-verified dataset comprising raster layers (GeoTIFF) for aggregate stability and erodibility, vector maps (Shapefiles) for soil types and erosion zones, and tabular data (CSV, Excel) for laboratory results. All data were prepared in standard formats commonly used in GIS and agronomic research.\u003c/p\u003e\n \u003cp\u003eThe testing protocol was designed to directly address the study's research questions by focusing on the key tasks required to bridge the „symptom-cause” divide. Instead of a comprehensive evaluation of all platform features, the protocol specifically targeted the functionalities essential for integrating and utilizing scientific soil health data. The core tested tasks included:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eData Import: Systematically attempting to import the field-verified dataset (TAD and K-factor maps in GeoTIFF, soil zone polygons in Shapefile) to assess technical compatibility and data integration barriers.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eDiagnostic Capability: Exploring the platforms for any built-in tools capable of analyzing soil physical properties or modeling erosion risk.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eScience-Based VRA Creation: Evaluating the possibility of creating VRA maps based directly on the imported scientific soil data, as an alternative to proxy-based methods.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n \u003cp\u003eAll observations – including successful and failed operations, error messages, and user friction points related to these specific tasks – were meticulously recorded. These insights formed the empirical foundation for identifying the barriers that prevent the practical application of scientific soil knowledge in commercial PA platforms.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec10\"\u003e\n \u003ch2\u003e2.3. Remote Sensing Data Acquisition and Analysis\u003c/h2\u003e\n \u003cp\u003eTo determine the nature and strength of the relationship between field-verified soil degradation indicators and vegetation vigor, this study employed the NDVI derived from satellite imagery. These data was acquired and analyzed through a two-step, complementary approach.\u003c/p\u003e\n \u003cp\u003eFirst, for a qualitative, cross-platform comparison, NDVI imagery was visually assessed. The specific dates for this visual analysis were selected based on the simultaneous availability of cloud-free imagery across the selected platforms (OneSoil, SatAgro, and GeoPard Agriculture) that provided accessible map visualization in their free/trial versions. This resulted in the selection of June 21st, 2022 for the CT field and June 6th, 2023 for the ST field. While OneSoil and SatAgro provided standard NDVI layers, GeoPard Agriculture offered the Enhanced Vegetation Index 2 (EVI2) as its primary indicator of vegetation vigor; therefore, EVI2 was used for this platform. This approach, while necessary for a direct comparison of how different systems present data, highlights a practical limitation of relying solely on commercial platform outputs, as data availability is often inconsistent.\u003c/p\u003e\n \u003cp\u003eSecond, to enable a robust quantitative analysis, a separate approach was employed. As the tested platforms did not offer a functionality to export point-specific NDVI values for user-defined coordinates, precise values were extracted for the 66 sampling points using Google Earth Engine (GEE). This process utilized the original Sentinel-2 Level-2A (atmospherically corrected) imagery for both of the dates identified above (21.06.2022 and 06.06.2023) in the platforms, ensuring consistency between the visual and statistical analyses. To confirm the stability of the observed relationships, this extraction was repeated for additional dates representing different crop growth stages. The relationship between these point-specific NDVI values and the field-verified soil data was then quantified using non-parametric statistical methods (Spearman's rank correlation and Kruskal-Wallis test with a subsequent Dunn's post-hoc test), chosen to accommodate the data's characteristics.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"3. Results","content":"\u003cp\u003eThis section presents the empirical results of the study, beginning with a detailed characterization of soil structural properties and erosion risk in the investigated agricultural fields (Section \u003cspan refid=\"Sec12\" class=\"InternalRef\"\u003e3.1\u003c/span\u003e). Subsequently, it presents the relationship between remote sensing data (NDVI) and field-verified soil properties (Section \u003cspan refid=\"Sec13\" class=\"InternalRef\"\u003e3.2\u003c/span\u003e) and concludes with an analysis of the empirical observations of the possibilities of commercial PA platforms to address these issues (Section \u003cspan refid=\"Sec14\" class=\"InternalRef\"\u003e3.3\u003c/span\u003e).\u003c/p\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e3.1. Field-Verified Soil Degradation and Erodibility\u003c/h2\u003e\u003cp\u003eThe field investigations confirmed that the soil cover in the studied hummocky moraine landscape is highly complex. Years of agricultural use and erosion have created a mosaic of different soil types across the fields, from completely eroded soils on hilltops to accumulated, fertile soils in depressions. This inherent spatial variability represents the core management challenge in this region (Świtoniak, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Świtoniak et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe soils in the study areas (both the 40 ha CT field and the 47 ha ST field) were broadly classified into four distinct units (A, B, C, D) based on the degree of anthropogenic denudation and their position along a hillslope catena (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003e). This classification, initially developed for this type of landscape (Świtoniak et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), reflects the spatial arrangement of soil types resulting from erosion and accumulation processes. A detailed comparison of the key properties for these soil categories under both tillage systems is presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eSoil A and B are typically found on upper slope segments and summits, where soil material has been removed due to denudation processes, leading to soil profile truncation.\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eSoils A: Regosols\u003c/b\u003e \u0026ndash; These represent completely eroded soils, primarily found on hummock summits and upper convex slopes. According to WRB (IUSS, 2015a), they are classified as Eutric Regosols (Protocalcic). Their profiles are weakly developed, often lacking diagnostic horizons, and are built directly from glacial till, resulting in an ACkp-Ck sequence. The plough layer often contains admixtures of calcareous parent material (Ck horizon), contributing to a typically low organic carbon content (C-SOC of 0.62% and 0.92% for CT and ST fields, respectively) and, particularly in the CT field, a high carbonate content (7.16%) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). These soils are characterized by the lowest structural stability, as evidenced by the lowest TAD values (208.8 s and 385.2 s for CT and ST, respectively) and the highest K-factor values (0.033 and 0.032 t\u0026middot;ha\u0026middot;h\u0026middot;ha\u0026ndash;1\u0026middot;MJ\u0026ndash;1\u0026middot;mm\u0026ndash;1), indicating the greatest susceptibility to erosion (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eSoils B: Eroded clay-illuvial soils\u003c/b\u003e \u0026ndash; These represent strongly eroded Luvisols, typically located on upper and middle slope segments with noticeable illuvial material (Bt horizon) exposed within the plough layer (ABtp horizon). They are generally classified as Haplic Luvisols (Protocalcic) (IUSS, 2015a). These soils often exhibit a higher clay content in the plough layer (up to 18%) and low SOC content. As detailed in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, their structural stability (TAD of 374.6 s and 797.8 s for CT and ST respectively) and erodibility (K-factor of 0.031 and 0.027 t\u0026middot;ha\u0026middot;h\u0026middot;ha\u0026ndash;1\u0026middot;MJ\u0026ndash;1\u0026middot;mm\u0026ndash;1) are intermediate between the completely eroded Regosols (Soils A) and the less disturbed soils of the lower slope segments.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSelected soil properties, aggregate stability indicators (TAD, MWDdry), and erodibility (K-factor) in the arable layer under conventional tillage and strip-till systems\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSoil group\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cem\u003eSoils A\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cem\u003eSoils B\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eSoils C\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003eSoils D\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWRB name\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEutric Regosol (Protocalcic)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHaplic Luvisol (Protocalcic)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAlbic Luvisol /\u003c/p\u003e\u003cp\u003eMollic Gleysol\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eEndogleyic Phaeozem /\u003c/p\u003e\u003cp\u003eMollic Gleysol\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSoil horizon sequence\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eACkp-Ck\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eABtp-Bt-Ck\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAp-E-Bt-Ckg/\u003c/p\u003e\u003cp\u003eAp-A-Eg-Btkl\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAp-A- A2-Ab-Ckl\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\u003cp\u003eConventional tillage (field Orzechowo, 4 catenae)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSand (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e59.3\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;4.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e56.3\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;4.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e66.0\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;4.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e60.0\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;3.9\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSilt (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e24.8\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;3.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e25.8\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;3.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e24.0\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;2.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e31.0\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;2.9\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClay (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e16.8\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;2.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e18.0\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;2.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e10.0\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;2.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e9.0\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1.6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003epHKCl\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.48\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7.31\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e6.38\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e6.77\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1.14\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eC-SOC (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.62\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.77\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.93\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2.35\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1.34\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCaCO3 (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.16\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.35\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.17\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.23\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.16\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBulk density (kg*m-3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.56\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eK (t\u0026middot;ha\u0026middot;h\u0026middot;ha\u0026ndash;1\u0026middot;MJ\u0026ndash;1\u0026middot;mm\u0026ndash;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.033\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.031\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.030\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.028\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMWDdry (mm)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e7.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e8.16\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTAD (sec.)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e208.8\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;81.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e374.6\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;31.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e616.2\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;28.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e771.8\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;30.3\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\u003cp\u003eStrip-till (field Sokołowo, 1 catena)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSand (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e72\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e64\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSilt (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e31\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClay (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003epHKCl\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e7.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e6.27\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eC-SOC (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2.39\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCaCO3 (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBulk density (kg*m-3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.47\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eK (t\u0026middot;ha\u0026middot;h\u0026middot;ha\u0026ndash;1\u0026middot;MJ\u0026ndash;1\u0026middot;mm\u0026ndash;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.027\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.026\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.027\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMWDdry (mm)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e7.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2.45\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTAD (sec.)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e385.2\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;66.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e797.8\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;81.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e489.6\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;27.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e900.0\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eNext two soil units (C and D) are typically found on lower slope segments, footslopes, and in depressions, exhibiting less severe truncation (C) or accumulation of colluvial material (D).\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eSoils C\u003c/b\u003e: These types represent slightly eroded or undisturbed Luvisols, primarily found in the lower parts of slopes. They are characterized by a full or nearly full sequence of genetic horizons (e.g., Ap-E-Bt-Ckl, or Ap-A-Eg-2Bt-2Ckg), classifying as Albic Luvisols or Mollic Gleysols (IUSS, 2015a). These soils represent the largest portion of the CT study area (58%), forming a \u0026bdquo;background\u0026rdquo; for the more degraded soils. They exhibit significantly improved structural stability (TAD of 616.2 s for CT) and lower erodibility (K-factor of 0.030 t\u0026middot;ha\u0026middot;h\u0026middot;ha\u0026ndash;1\u0026middot;MJ\u0026ndash;1\u0026middot;mm\u0026ndash;1 for CT) compared to the truncated soils of the upper slopes, which corresponds with their higher C-SOC content (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eSoils D: colluvial soils\u003c/b\u003e, formed by the accumulation of eroded material in depressions and at footslopes. Their morphology often indicates significant influence of groundwater gleying. They are classified as Endogleyic Phaeozems or Mollic Gleysols (IUSS, 2015a). As shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, these soils are characterized by the highest organic carbon content (C-SOC of 2.35% and 2.39% for CT and ST respectively) and, consequently, the most stable soil structure, with the highest TAD values (771.8 s and 900.0 s) and the lowest erodibility (K-factor of 0.028 and 0.027 t\u0026middot;ha\u0026middot;h\u0026middot;ha\u0026ndash;1\u0026middot;MJ\u0026ndash;1\u0026middot;mm\u0026ndash;1 for CT and ST fields respectively).\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eThese findings document the complex, field-scale reality of soil degradation and form the \u0026bdquo;ground truth\u0026rdquo; against which platform capabilities are evaluated.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e3.2. Remote Sensing of a Symptom: Linking NDVI to Soil Degradation Patterns\u003c/h2\u003e\u003cp\u003eWhile direct, field-based assessment provides the most accurate characterization of soil physical properties, it is labor-intensive and costly to perform at a large scale. In contrast, PA heavily relies on remote sensing data, particularly vegetation indices, as a practical and cost-effective means to assess in-field variability. Therefore, the next logical step in our investigation was to determine whether a quantifiable relationship exists between the field-verified \u0026bdquo;cause\u0026rdquo; (soil degradation) and the remotely-sensed \u0026bdquo;symptom\u0026rdquo; (vegetation vigor).\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eTo investigate this relationship, a quantitative statistical analysis was conducted based on 66 point samples from the CT field, where precise NDVI values were extracted from Sentinel-2 imagery via\u003c/span\u003e GEE (see Supplementary Material, Tab. S3 and Fig. S2). A non-parametric Spearman's rank correlation (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) revealed a significant positive relationship between NDVI and aggregate stability (TAD, s) (r2\u0026thinsp;=\u0026thinsp;0.56, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01) and a statistically significant, but weak, negative relationship with the soil erodibility K-factor, t\u0026middot;ha\u0026middot;h\u0026middot;ha\u0026ndash;1\u0026middot;MJ\u0026ndash;1\u0026middot;mm\u0026ndash;1 (r2 = -0.35, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). This indicates that soils with higher structural stability and lower erosion susceptibility support more vigorous vegetation. Crucially, a strong positive correlation was found between the NDVI values from 2022 and 2023 (r2\u0026thinsp;=\u0026thinsp;0.72, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), indicating a high temporal stability of the observed spatial patterns.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSpearman's rank correlation coefficients (rs) between NDVI and selected soil degradation indicators for the field under CT system (in bold statistically significant (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) data)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eK, t\u0026middot;ha\u0026middot;h\u0026middot;ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e\u0026middot;MJ\u003csup\u003e\u0026ndash;1\u003c/sup\u003e\u0026middot;mm\u003csup\u003e\u0026ndash;1\u003c/sup\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSoil unit\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTAD, s\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eNDVI 21.06.2022\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNDVI 06.06.2023\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eK, t\u0026middot;ha\u0026middot;h\u0026middot;ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e\u0026middot;MJ\u003csup\u003e\u0026ndash;1\u003c/sup\u003e\u0026middot;mm\u003csup\u003e\u0026ndash;1\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e-0.64\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e-0.58\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e-0.28\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e-0.35\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSoil unit\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e-0.64\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.86\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.53\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.49\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTAD, s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e-0.58\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.86\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.56\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.59\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNDVI 19.06.2022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e-0.28\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.53\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.56\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.72\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNDVI 06.06.2023\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e-0.35\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.49\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.59\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.72\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eFurthermore, a Kruskal-Wallis test showed highly significant differences in median NDVI values among the four delineated soil units (H\u0026thinsp;=\u0026thinsp;99.12, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). A subsequent Dunn's post-hoc analysis (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) confirmed that the most eroded soils (Group A) exhibited significantly lower NDVI values than the undisturbed and colluvial soils (Groups C and D), a relationship further illustrated by the scatter plot in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003e .\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of the Dunn's post-hoc test for pairwise comparisons of median NDVI (21.06.2022) values between soil groups\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSoils_A\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSoils_B\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eSoils_C\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSoils_D\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSoils_A\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.1424\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.01141\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e5.65E-05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSoils_B\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.1424\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.3783\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.01047\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSoils_C\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e0.01141\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.3783\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.04594\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSoils_D\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e5.65E-05\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e0.01047\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.04594\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eA qualitative visual assessment of several commercial PA platforms confirmed that they also detect similar spatial patterns of in-field variability, albeit with significant differences in visualization style, as shown for the CT field in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eThe consistent, strong visual and statistical correlation observed leads to a crucial insight. The high temporal stability of the NDVI patterns between 2022 and 2023, despite a change in the cultivated crop, suggests that the legacy of erosion involves more profound, lasting changes than just structural degradation. This indicates that the spatial variability of crop performance is primarily governed by fundamental, persistent site characteristics \u0026ndash; such as irreversible alterations in soil texture, depth, and water-holding capacity \u0026ndash; rather than by the specific needs of the crop in a given year. These results empirically demonstrate that remotely-sensed vegetation indices are effective at identifying the complex, persistent symptoms of poor soil physical health. This finding provides the context for the subsequent evaluation of the diagnostic capabilities of PA platforms.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e3.3. Empirical Assessment of Practical Capabilities of Commercial PA Platforms for Soil Health Management\u003c/h2\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eHaving established the strong, quantifiable link between the underlying\u003c/span\u003e cause \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e(soil degradation) and its remotely-sensed\u003c/span\u003e symptom \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e(NDVI), the final and central part of the investigation assessed the practical capabilities of the nine selected commercial PA platforms to facilitate a science-based, causal diagnosis.\u003c/span\u003e\u003c/p\u003e\u003cp\u003eThe key findings from the empirical user experience tests are summarized in a comparative matrix (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) and detailed below.\u003c/p\u003e\u003cp\u003e\u003cb\u003eData Integration and Export Capabilities\u003c/b\u003e:\u003c/p\u003e\u003cp\u003eThe ability to import and export standard scientific data formats was found to be a primary and widespread limitation.\u003c/p\u003e\u003cp\u003e\u003cb\u003eRaster (GeoTIFF) Import.\u003c/b\u003e A direct and functional tool for importing GeoTIFF files, the standard format for continuous soil maps (e.g., TAD, K-factor), was not found in the tested versions of \u003cem\u003eAgFiniti\u003c/em\u003e, \u003cem\u003eAGMRI\u003c/em\u003e, \u003cem\u003eOneSoil\u003c/em\u003e, and \u003cem\u003eJDOC\u003c/em\u003e. While platforms like \u003cem\u003eClimate FieldView\u003c/em\u003e and \u003cem\u003eAgriCircle\u003c/em\u003e declared support, this functionality was not verifiable without paid access. For \u003cem\u003eSatAgro\u003c/em\u003e and \u003cem\u003eGeoPard\u003c/em\u003e, raster import was either explicitly a premium feature or proved problematic during testing.\u003c/p\u003e\u003cp\u003e\u003cb\u003eVector (Shapefile) and Tabular (CSV) Import.\u003c/b\u003e While Shapefile import for simple field boundaries was commonly supported (e.g., \u003cem\u003eJDOC\u003c/em\u003e, \u003cem\u003eGranular\u003c/em\u003e), attempts to import thematic polygons with complex attributes frequently resulted in critical errors. Platforms like \u003cem\u003eAGMRI\u003c/em\u003e and \u003cem\u003eGranular\u003c/em\u003e failed to process standard archives, returning cryptic messages (\u0026bdquo;Missing date for soil data activity\u0026rdquo;, \u0026bdquo;File not readable\u0026rdquo;) \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e(see Supplementary Material, Fig. S4B). Other platforms, like\u003c/span\u003e \u003cspan type=\"ItalicSmallCaps\" class=\"ItalicSmallCaps\" name=\"Emphasis\"\u003eGeoPard\u003c/span\u003e, \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eexhibited issues with non-numeric attributes (see Supplementary Material, Figure S4A). Furthermore, some platforms (e.g.\u003c/span\u003e, \u003cspan type=\"ItalicSmallCaps\" class=\"ItalicSmallCaps\" name=\"Emphasis\"\u003eOneSoil\u003c/span\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e) demonstrated semantic limitations by misinterpreting soil zone polygons as separate fields (see Supplementary Material, Fig. S4C). Support for simple tabular CSV data was also inconsistent and often unavailable.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eData Export.\u003c/b\u003e The crucial functionality of exporting analyzed data in interoperable formats (e.g., Shapefile) was severely restricted, often absent in free versions (\u003cem\u003eOneSoil\u003c/em\u003e) or available only as a premium feature (\u003cem\u003eSatAgro\u003c/em\u003e, \u003cem\u003eFieldView\u003c/em\u003e). \u003cem\u003eGeoPard\u003c/em\u003e was a notable exception, offering robust export functionality in its trial version.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eFunctional Capabilities and VRA Pathways\u003c/b\u003e:\u003c/p\u003e\u003cp\u003e\u003cb\u003eAvailability of Diagnostic Tools.\u003c/b\u003e A near-universal finding was the absence of dedicated, built-in tools for soil structure or erosion risk analysis. None of the tested platforms offered functionalities to calculate or interpret key physical indicators like TAD or K-factor. In the case of \u003cem\u003eAGMRI\u003c/em\u003e, the entire \u0026bdquo;Analytics\u0026rdquo; section was marked as \u0026bdquo;Coming Soon\u0026rdquo; highlighting a functional deficit. Even advanced platforms like \u003cem\u003eGeoPard\u003c/em\u003e, while offering a powerful scripting environment (Python), lacked ready-to-use modules, placing \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ethe responsibility\u003c/span\u003e of additional algorithm development \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eon the user (see Supplementary Material, Fig. S5C). The \u0026bdquo;black box\u0026rdquo; nature of some available tools was also problematic, with platforms offering simplified, non-scientific soil classifications (see Supplementary Material, Fig. S5A) or generating unreliable data outputs (see Supplementary Material, Fig. S5B).\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eScience-Based VRA.\u003c/b\u003e The ability to create VRA maps based directly on user-imported scientific soil data was systematically restricted and consistently treated as a premium feature locked behind a paywall (\u003cem\u003eAgFiniti\u003c/em\u003e, \u003cem\u003eFieldView\u003c/em\u003e, \u003cem\u003eGeoPard\u003c/em\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSummary of Key Functionalities and Observed Outcomes for Tested PA Platforms\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"9\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003ePlatform\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eData Integration\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003eFunctional Capability\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c9\" namest=\"c7\"\u003e\u003cp\u003eAccessibility \u0026amp; Business Model\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRaster (GeoTIFF) Import\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eThematic Shapefile Import\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eData Export (Interoperable)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSoil Structure / Erosion Analytics\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eVRA from Custom Soil Data\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eOpen Access / Functional Free Trial\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003eEcosystem / Business Model Barrier\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e\u003cp\u003eCompatibility with other PA Platforms\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eAgFiniti (Ag Leader)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eX*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e✓ (JDOC, Climate FieldView, New Holland)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eAGMRI (IntelinAir)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eCO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e✓ (JDOC, Climate FieldView, AgFiniti)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eAgriCircle\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eDL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003en.d.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eClimate FieldView (Bayer)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDL/P\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eMedium\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e✓(JDOC)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eGeoPard Agriculture\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e✓ (Boundary)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e✓ (Shapefile)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e✓\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eLow\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e✓ (Very Strong, 2-way with JDOC)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eGranular (Corteva)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e✓ (Boundary)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e✓/P\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e✓ (AgFinity, JDOC, PP Panorama, FieldAlytics, CNH)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eJDOC (John Deere)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e✓ (Boundary)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eExtreme\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e✓(\u0026nbsp;about 100 third-party software partners)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eOneSoil\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e✓ (Boundary, poligons)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e✓/P\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eLow\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e✓(JDOC)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eSatAgro\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e✓/P\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eMedium\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e✓ (JDOC, Trimble Connected Farm)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e*✓\u003c/b\u003e - Supported and functional in the tested version; X - Absent or not found in the tested version; P - Problematic, limited, or non-intuitive functionality; PL - Functionality exists but is behind a paywall; DL - Declared by the vendor but not empirically verified in the test; CO - Coming Soon (feature announced but not available); \u003cb\u003e✓\u003c/b\u003e (Boundary) - Supported only for simple field boundaries; Barrier Level (Low/Medium/High/Extreme): A qualitative assessment of the severity of ecosystem/business model barriers to open access and data use.\u003c/p\u003e\u003cp\u003e\u003cb\u003ePlatform Access and Ecosystem Integration\u003c/b\u003e:\u003c/p\u003e\u003cp\u003eFundamental access was often impeded by business model constraints. \u003cem\u003eJDOC\u003c/em\u003e and \u003cem\u003eAgFiniti\u003c/em\u003e demonstrated a strong reliance on a dealer-centric network, requiring formal business affiliation. \u003cem\u003eAgStudio (Corteva)\u003c/em\u003e offered no public sign-up option, while \u003cem\u003eAGMRI\u003c/em\u003e's model presented significant geographical and organizational hurdles.\u003c/p\u003e\u003cp\u003eThe results also highlight a trend towards interoperability, with platforms like \u003cem\u003eJDOC\u003c/em\u003e connecting to a wide network of nearly 100 third-party software partners (see Supplementary Material, S2 for a full list). This extensive ecosystem highlights the platform's focus on operational data integration yet a review of these partners reveals a scarcity of solutions focused on soil physical health diagnostics.\u003c/p\u003e\u003c/div\u003e"},{"header":"4. DISCUSSION","content":"\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe results of this study provide a multi-layered view of the significant gap between the scientific understanding of soil physical health and the current capabilities of commercial PA software. In this section, we interpret these findings, beginning with the relationship between the degradation cause and its remotely-sensed symptom, then moving to a detailed analysis of the \u0026bdquo;diagnostic void\u0026rdquo; in PA platforms and the systemic logic that underpins it.\u003c/span\u003e\u003c/p\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e4.1. \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eEstablishing the Ground Truth: The Signature of Soil Degradation on Crop Vigor\u003c/span\u003e\u003c/h2\u003e\u003cp\u003eThe findings of this study provide a clear, field-based picture of the complex nature of soil physical degradation in intensively cultivated hummocky moraine landscapes. The direct comparison between conventional (CT) and strip-till (ST) systems, although based on a limited number of catenae, provided valuable insights into the potential effects of tillage on soil physical health. Our findings indicate a clear trend toward improved soil structure under the ST system, particularly for the most degraded soil units. The strongly eroded Luvisols (Soils B), for example, exhibited a more than twofold improvement in aggregate water stability (TAD) under ST compared to the same soil unit under CT (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). This suggests that while these soils are significantly degraded, they possess a high potential for structural regeneration when managed under conservation-oriented practices. Soil erodibility (K-factor) also showed a tendency toward lower values under the ST system, which is consistent with its slightly higher SOC content. These observations align with the established body of literature supporting the role of reduced tillage intensity as a key practice for enhancing soil structure and mitigating erodibility (Blake and Hartge, 1986; Lal, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2004\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eOur results from both the conventional (CT) and strip-till (ST) sites demonstrate that long-term, erosion-driven processes have created a distinct soil mosaic, where indicators of soil health \u0026ndash; such as aggregate stability (TAD) and erodibility (K-factor) \u0026ndash; exhibit pronounced spatial variability (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). This inherent heterogeneity, representing the \u0026bdquo;cause\u0026rdquo; of varying soil productivity, poses a fundamental challenge for site-specific management.\u003c/p\u003e\u003cp\u003eCrucially, our quantitative analysis established a statistically significant correlation between these field-measured soil properties and remotely-sensed vegetation vigor, as captured by NDVI (Tables\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003e (B-D). The visual comparison further confirmed this strong correspondence: areas characterized by lower aggregate stability and higher soil erodibility consistently exhibited lower NDVI values, particularly during early crop growth stages (Fig.s 5 (B-D), Supplementary Material, Fig. S6). This confirms that NDVI, a readily available tool in most PA platforms, can serve as a reliable proxy for the spatial patterns of soil degradation.\u003c/p\u003e\u003cp\u003eHowever, it is essential to recognize that NDVI reflects the symptom (i.e., reduced vegetation vigor) rather than the underlying causes of limited productivity, which may include not only poor soil structure and high erodibility but also related limitations in water and nutrient availability. The observed persistence of these spatial NDVI patterns, even in post-harvest imagery, further suggests that they are strongly governed by stable, inherent site conditions \u0026ndash; primarily soil properties shaped by topography and long-term erosion processes \u0026ndash; rather than by \u003cem\u003eshort-term\u003c/em\u003e crop management factors. This fundamental distinction between a detectable symptom and its complex underlying cause frames the central challenge investigated in this study: are commercial PA platforms capable of bridging this diagnostic gap?\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003e4.2. \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eUncovering the \u0026bdquo;Diagnostic Void\u0026rdquo;: A Systemic Failure of Commercial Ag-Tech\u003c/span\u003e\u003c/h2\u003e\u003cp\u003eThe empirical investigation of commercial PA platforms revealed a profound disconnect between the sophisticated diagnostic capabilities needed for science-based soil management and the actual functionalities available to users. However, it is important to first acknowledge the positive trends within this dynamic market. Many platforms are increasingly focused on creating interconnected ecosystems. A prime example is the \u003cem\u003eJDOC\u003c/em\u003e, which acts as a central hub connecting to nearly 100 third-party software partners (see Supplementary Material, S2). This trend towards interoperability for operational data is a significant step forward, allowing for a more integrated flow of information across different brands and services. Despite this progress in connectivity, our analysis focused on a more fundamental question: do these platforms, individually or as an ecosystem, provide the necessary tools for causal soil health diagnosis?\u003c/p\u003e\u003cp\u003eOur comprehensive analysis of 34 distinct PA platforms revealed a striking absence of dedicated tools for diagnosing the erosion-related causes underlying NDVI variability patterns. Despite the platforms' advanced capabilities for vegetation monitoring and spatial analysis, none offered built-in functionalities for assessing key soil structure parameters such as aggregate stability (TAD) or quantitative erosion risk modeling through established metrics like the K-factor (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This gap persists even across platforms with declared focus on sustainability or advanced analytics, suggesting that the absence of these tools is not merely an oversight but reflects underlying systemic priorities within the commercial PA ecosystem.\u003c/p\u003e\u003cp\u003eThe contrast between platforms' advanced vegetation monitoring capabilities and their rudimentary soil diagnostic tools is particularly striking. While platforms like \u003cem\u003eGeoPard Agriculture\u003c/em\u003e demonstrate powerful multi-layer geospatial analysis capabilities for vegetation indices, they lack dedicated modules for soil structure or erosion modeling. This forces users toward a \u0026bdquo;bring your own algorithm\u0026rdquo; approach, requiring custom script development (e.g., in Python) for basic soil analysis functions. Such requirements place an extreme knowledge requirements on the average practitioner and effectively exclude those without programming expertise from conducting advanced soil diagnostics, despite the availability of suitable scientific methodologies (see Supplementary Material, Fig. S5C).\u003c/p\u003e\u003cp\u003eThis functional limitation becomes particularly problematic when viewed against the backdrop of our field findings. The persistence of erosion-related soil degradation patterns even under conservation management systems (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003e, Supplementary material, Fig. S6) demonstrates that vegetation indices alone provide insufficient diagnostic information for sustainable management decisions. The platforms' reliance on easily accessible proxy data, particularly NDVI, creates a management paradigm focused on treating symptoms rather than addressing the underlying pedological causes of field variability. This symptom-based approach may inadvertently guide users toward agronomically unsound decisions, such as applying additional nitrogen inputs to areas where poor soil structure or low water-holding capacity are the primary limiting factors. Furthermore, even the presentation of basic proxy data like NDVI is fraught with usability and transparency issues. \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eOur qualitative comparison revealed a significant lack of standardization in how platforms visualize the same underlying satellite data. Users are faced with disparate outputs, making direct, cross-platform comparison difficult and potentially misleading (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e). These outputs range from high-contrast, pixelated maps of\u003c/span\u003e \u003cspan type=\"BoldSmallCaps\" class=\"BoldSmallCaps\" name=\"Emphasis\"\u003e\u0026bdquo;\u003c/span\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eContrast NDVI\u0026rdquo; (\u003c/span\u003e\u003cspan type=\"ItalicSmallCaps\" class=\"ItalicSmallCaps\" name=\"Emphasis\"\u003eOneSoil\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003eB\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), to the differently processed and classified NDVI map from\u003c/span\u003e \u003cspan type=\"ItalicSmallCaps\" class=\"ItalicSmallCaps\" name=\"Emphasis\"\u003eSatAgro\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e(\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003eD\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e), and the smoothed, low-contrast surfaces of the EVI2, which\u003c/span\u003e \u003cspan type=\"ItalicSmallCaps\" class=\"ItalicSmallCaps\" name=\"Emphasis\"\u003eGeoPard Agriculture\u003c/span\u003e \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eoffers as its primary vigor indicator (\u003c/span\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003eC\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e).\u003c/span\u003e Such \u0026bdquo;black box algorithm\u0026rdquo; problem is even more acute when platforms offer \u0026bdquo;inferred\u0026rdquo; data, which our verification showed can be unreliable (\u003cem\u003eAgriCircle\u003c/em\u003e), eroding user trust.\u003c/p\u003e\u003cp\u003eThe practical consequence of this \u0026bdquo;diagnostic void\u0026rdquo; is most evident in the VRA modules (see Supplementary Material, Fig. S7). The risk associated with simplified, proxy-based VRA strategies is amplified by the opaque and contradictory logic offered to the user. Our empirical testing of the \u003cem\u003eSatAgro\u003c/em\u003e platform provides a stark illustration. The platform requires the user to generate a VRA prescription by choosing between just two opposing quantitative strategies: \u0026bdquo;Directly proportional\u0026rdquo; or \u0026bdquo;Inversely proportional\u0026rdquo;. This forces the user into a simplistic, binary choice without any diagnostic support to determine if a low-NDVI zone is caused by a temporary nutrient deficiency or a permanent physical limitation. This critical diagnostic decision between two fundamentally different agronomic philosophies is left entirely to the user, who is given contradictory options based on the same superficial data, creating a scenario of high risk for both farm economics and environmental health. \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThis symptom-based approach may inadvertently guide users toward agronomically unsound decisions, such as applying additional nitrogen inputs to areas where poor soil structure or low water-holding capacity are the primary limiting factors. For the farmer, such a misdiagnosis leads to direct economic losses through wasted inputs and unrealized yield potential. Simultaneously, it poses significant environmental risks, including nutrient runoff which can contribute to the eutrophication of surface waters \u0026ndash; a critical issue in lake-rich landscapes like the one studied.\u003c/span\u003e\u003c/p\u003e\u003cp\u003eBeyond technical limitations, our investigation revealed a deeper, conceptual constraint rooted in the underlying design logic of most commercial PA platforms. These systems are not built primarily as flexible scientific analysis tools, but as integrated farm management environments. As a result, accessing even basic analytical functions often requires full operational context \u0026ndash; such as crop type, machinery data, or management plans \u0026ndash; effectively constraining the user to an \u0026bdquo;operation-first\u0026rdquo; workflow. This design logic poses a significant barrier to hypothesis-driven evaluation and impedes the integration of scientific soil data by users who are not operating within a full-season management framework.\u003c/p\u003e\u003cdiv id=\"Sec18\" class=\"Section3\"\u003e\u003ch2\u003e\u003cb\u003e4.2.1\u003c/b\u003e. \u003cb\u003eBarriers to Scientific Data Integration\u003c/b\u003e\u003c/h2\u003e\u003cp\u003eThe second dimension of the diagnostic void concerns the systematic exclusion of pathways for integrating user-generated, field-verified soil data. Our empirical testing revealed that the inability to import standard raster formats like GeoTIFF \u0026ndash; the lingua franca for spatial soil science \u0026ndash; represents a critical obstacle across most platforms. This technical limitation is particularly problematic given the abundance of high-resolution soil data available through modern sensing technologies and the growing body of research demonstrating the integration of vegetation indices with field-verified soil data using machine learning models (Bantchina et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Cai et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Vedeneeva and Koshelev, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThese technical barriers are compounded by educational and cognitive obstacles. Even when data upload is technically possible, the complexity of interfaces, cryptic error messages (observed in \u003cem\u003eAGMRI, Granular\u003c/em\u003e), and restrictive data structure requirements effectively block the integration of field-verified soil information (see Supplementary Material, Fig. S4). This confirms that the knowledge barrier identified in earlier studies persists, with data entry, conversion, and visualization in farm software remaining too complex for most users, particularly when working with scientific soil data (Knapic, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Padhiary et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Robinson et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe economic dimension of this barrier is equally significant. Commercial models increasingly rely on subscription-based access, creating economic lock-ins that hinder experimentation with scientific data tools. Platforms like \u003cem\u003eGeoPard Agriculture\u003c/em\u003e link critical analytical features to monthly payments, making long-term data sovereignty dependent on continued financial engagement (see Supplementary Material, Fig. S3). This pricing structure particularly affects small and medium-sized farms, which are often priced out of access to advanced PA tools or cloud-based analytics services (Knapic, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Padhiary et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Pandeya et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section3\"\u003e\u003ch2\u003e4.2.2. The Paradox of Technological Sophistication\u003c/h2\u003e\u003cp\u003eThe existence of this diagnostic void presents a fundamental paradox within the commercial PA ecosystem. While platforms demonstrate remarkable sophistication in exchanging complex, proprietary machine data across extensive partner networks, they remain unable to integrate standardized scientific data formats. This \u0026bdquo;GeoTIFF paradox\u0026rdquo; highlights that the technical challenges are not insurmountable but rather reflect deliberate design priorities that favor operational data exchange over scientific data integration.\u003c/p\u003e\u003cp\u003eThe paradox becomes even more pronounced when contrasted with the vibrant research landscape in soil health assessment. The scientific community actively develops advanced solutions using in-situ and proximal soil sensing (Loria et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), remote sensing applications (Arab et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Meinen and Robinson, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), and machine learning approaches for mapping key soil indicators (Bantchina et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Baseca et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Jana et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Medeiros et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The discrepancy between available scientific knowledge and platform capabilities suggests that the challenge lies not in technical feasibility but in the translation and deployment of this knowledge into commercially viable and user-accessible tools.\u003c/p\u003e\u003cp\u003eHowever, the very existence of these extensive ecosystems, such as the one surrounding \u003cem\u003eJDOC\u003c/em\u003e, also represents a significant opportunity. While our study identified a systemic \u0026bdquo;blind spot\u0026rdquo; regarding soil physical health tools, this interconnected structure could facilitate rapid, widespread adoption of new solutions. If even one partner within this network were to develop and validate a robust module for erosion risk assessment or soil structure analysis, it could potentially be made available to the entire user base of the ecosystem. This suggests that bridging the science-practice gap may not require every platform to build these tools from scratch, but could instead be achieved through the strategic development of specialized, interoperable applications within these existing commercial networks.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section3\"\u003e\u003ch2\u003e4.2.3. Implications for Science-Based Management\u003c/h2\u003e\u003cp\u003eThe \u0026bdquo;diagnostic void\u0026rdquo; identified in commercial PA platforms carries significant implications for the implementation of science-based soil management strategies. By limiting users to vegetation indices as primary diagnostic tools, platforms may inadvertently perpetuate management approaches that fail to address the root causes of soil degradation. This limitation is particularly problematic in landscapes shaped by historical erosion, where the legacy effects of soil degradation involve profound, often persistent changes to soil profiles that are not fully mitigated by short- to medium-term improvements in management practices (Berhe, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Lal, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2004\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe absence of diagnostic capabilities and restricted data integration pathways effectively forces users into a reactive, symptom-based management paradigm rather than enabling proactive, science-based approaches to soil health. This limitation undermines the potential for PA technologies to contribute meaningfully to sustainable agriculture and may even exacerbate existing environmental challenges if users are guided toward inappropriate management responses based on incomplete diagnostic information (Carolan, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe current state of commercial PA platforms thus represents a critical bottleneck in the translation of scientific soil knowledge into practical management applications. Addressing this diagnostic void will require fundamental changes in platform design priorities, business models, and user interface development to bridge the gap between scientific capabilities and commercial implementation.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003e4.3. Environmental and Economic Implications of the Diagnostic Void\u003c/span\u003e\u003c/h2\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThe identified diagnostic void is not merely an academic concern; it carries significant risks for both farm profitability and environmental health. By limiting users to symptom-based diagnostics (i.e., NDVI maps), platforms inadvertently encourage management decisions that can be both economically inefficient and ecologically harmful. For instance, a farmer observing a low-vigor zone may be guided to apply more nitrogen fertilizer, following a \u0026bdquo;fix-the-symptom\u0026rdquo; logic. However, as our field data shows, such zones are often limited by poor soil structure and low water-holding capacity, not nutrient deficiency. In this scenario, the applied fertilizer is not utilized by the crop, leading to\u003c/span\u003e:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eDirect economic loss for the farmer through wasted inputs and unrealized yield potential.\u003c/span\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eIncreased environmental pollution, as excess nitrogen leaches into groundwater or contributes to greenhouse gas emissions (N₂O), a particularly pressing issue in nitrogen-vulnerable zones like our study area.\u003c/span\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003ePerpetuation of soil degradation, as the root cause of the problem remains unaddressed, potentially worsening with each season.\u003c/span\u003e\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThis symptom-driven paradigm, hardwired into the current PA ecosystem, thus undermines the core promise of precision agriculture \u0026ndash; to enhance efficiency and sustainability. It creates a high-risk environment where farmers are given sophisticated tools to make potentially wrong decisions with precision, locking them into a cycle of reactive management that fails to build long-term soil health and resilience.\u003c/span\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003e4.4. Limitations, Implications, and Future Outlook\u003c/h2\u003e\u003cp\u003eThis study has several limitations that frame the context of its findings and highlight critical directions for future research. The comparative analysis of tillage systems was based on an unbalanced sampling design, meaning conclusions regarding the spatial effects of strip-tillage should be considered preliminary. Similarly, our evaluation of PA platforms, based on trial versions, represents a \u0026bdquo;snapshot\u0026rdquo; of a highly dynamic market. While we argue that the identified systemic barriers are fundamental, their specific manifestations may evolve. These limitations underscore a broader challenge: the scarcity of independent, systematic evaluations of commercial PA tools, which creates an information vacuum for end-users (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe implications of our findings are significant. The identified \u0026bdquo;diagnostic void\u0026rdquo; forces users into a reactive, symptom-based management paradigm, which is economically inefficient and environmentally risky, particularly in landscapes with a strong \u0026bdquo;legacy of erosion\u0026rdquo; (Berhe, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Lal, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). This undermines the core promise of PA to deliver truly precise, science-based soil management.\u003c/p\u003e\u003cp\u003eBridging this gap requires a concerted effort. Based on our findings, we propose several key recommendations for the future:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eEnhancing Data and Model Integration\u003c/b\u003e: Platform developers must prioritize robust import tools for scientific formats (GeoTIFF, complex Shapefiles) and begin embedding validated scientific models (e.g., for K-factor calculation) directly into their workflows, moving beyond the \u0026bdquo;bring your own algorithm\u0026rdquo; approach.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eExploring Open and Collaborative Models\u003c/b\u003e: The potential of open-source tools (QGIS, R, Python) as a transparent and flexible alternative should be further explored, despite their current expertise requirements (Zag\u0026oacute;rda and Walczykova, 2018). Furthermore, closer academia-industry collaboration is crucial. PA companies should facilitate independent validation through APIs or research accounts (as exemplified by \u003cem\u003eGeoPard\u003c/em\u003e), while researchers must work to make their models \u0026bdquo;platform-ready\u0026rdquo;.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eInvestigating Ecosystem Interoperability\u003c/b\u003e: A key area for future research lies in the practical reality of ecosystem interoperability. Our study highlights the potential of large partner networks like that of \u003cem\u003eJDOC\u003c/em\u003e. Future studies should empirically investigate the end-to-end data workflow within these ecosystems to determine if they offer genuine, holistic decision support or create a more fragmented digital landscape.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eDeveloping User-Centric and Inclusive Tools\u003c/b\u003e: Finally, the focus must shift towards designing tools that empower users to understand the causes of in-field variability, not just the symptoms. The direct benefit for farmers would be the ability to make more targeted and cost-effective management decisions \u0026ndash; for example, allocating resources for soil amendments in degraded zones instead of over-applying fertilizers that the crop cannot utilize. This is particularly crucial for small and medium-sized farms, which are disproportionately affected by both soil degradation and the barriers of current PA tools (Knapic, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Padhiary et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eBy creating more inclusive and genuinely diagnostic systems that demonstrate a clear return on investment through improved input efficiency and long-term soil productivity, a true market demand for sustainable soil management can be fostered.\u003c/span\u003e\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eAddressing these challenges is essential if digital agriculture is to move beyond its current limitations and genuinely contribute to a more sustainable and resilient food system.\u003c/p\u003e\u003c/div\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eThis study provided a unique, field-verified assessment of the gap between the scientific understanding of soil physical degradation and the practical capabilities of commercial Precision Agriculture (PA) platforms. Conducted within a representative hummocky moraine landscape, the study leads to the following key conclusions:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eThere is a strong, statistically significant relationship between field-verified indicators of soil degradation and remotely sensed vegetation indices (NDVI). This confirms that NDVI, while a powerful proxy for crop performance, is fundamentally a symptom indicator. Used in isolation \u0026ndash; as is common practice in commercial PA platforms \u0026ndash; it cannot diagnose the underlying causes of poor soil health, making it a powerful yet potentially misleading tool for science-based management.\u003c/span\u003e\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eOur empirical investigation revealed a systemic \u0026bdquo;diagnostic void\u0026rdquo; across the commercial PA ecosystem. Current platforms are fundamentally limited in their ability to support causal diagnosis of soil degradation, characterized by a near-total absence of built-in tools for assessing soil structure or erosion risk. This void is reinforced by a wall of technical, usability, and economic barriers that prevent the integration of user-generated scientific soil data.\u003c/span\u003e\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eThis \u0026bdquo;diagnostic void\u0026rdquo; \u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eis not an accidental oversight but a direct consequence of a market failure where business priorities for short-term returns and proprietary data ecosystems trump the need for genuine, science-based soil stewardship. This focus perpetuates a symptom-driven management paradigm that is not just ineffective, but potentially economically wasteful and environmentally hazardous, especially in sensitive, erosion-prone landscapes.\u003c/span\u003e\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"SmallCaps\" class=\"SmallCaps\" name=\"Emphasis\"\u003eIn conclusion, realizing the true potential of digital agriculture requires a fundamental reorientation of the entire PA sector. Moving forward, the focus must shift from simply treating visible symptoms to empowering users to manage the complex causes of soil degradation. This demands a move towards open standards, genuine interoperability, and the integration of validated scientific models. Without this paradigm shift, precision agriculture risks becoming a tool for precisely managing our way into continued soil degradation, rather than a solution for building a resilient and sustainable agricultural future.\u003c/span\u003e\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"572\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eCSV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eComma Separated Values\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eCT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eConventional Tillage\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eDEM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eDigital Elevation Model\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eEPIC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eErosion-Productivity Impact Calculator\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eEVI2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eEnhanced Vegetation Index 2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eFMIS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eFarm Management Information System\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eFMIS/ERP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eFarm Management Information System / Enterprise Resource Planning\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eGEE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eGoogle Earth Engine\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eGIS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eGeographic Information System\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eHCP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eHardware-Centric Platform\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eJDOC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eJohn Deere Operations Center\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eK-factor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eSoil Erodibility Factor\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eMWDdry\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eMean Weight Diameter (dry)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eNDVI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eNormalized Difference Vegetation Index\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003ePA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003ePrecision Agriculture\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eROI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eReturn on Investment\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eRUSLE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eRevised Universal Soil Loss Equation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eSM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eSupplementary Material\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eSOC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eSoil Organic Carbon\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eStrip-Till\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eTAD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eTime of Aggregate Destruction\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eUSLE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eUniversal Soil Loss Equation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eVRA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eVariable Rate Application\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eWRB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 488px;\"\u003e\n \u003cp\u003eWorld Reference Base for Soil Resources\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAit Issad, H., Aoudjit, R., Rodrigues, J.J.P.C., 2019. A comprehensive review of Data Mining techniques in smart agriculture. Eng. Agric. Environ. Food 12, 511–525. https://doi.org/10.1016/j.eaef.2019.11.003\u003c/li\u003e\n \u003cli\u003eArab, S., Easson, G., Ghaffari, Z., 2024. Integration of Sentinel-1A Radar and SMAP Radiometer for Soil Moisture Retrieval over Vegetated Areas. Sensors 24, 2217. https://doi.org/10.3390/s24072217\u003c/li\u003e\n \u003cli\u003eArias-Navarro, C., Baritz, R., Jones, A., European Commission, European Environment Agency (Eds.), 2024. The state of soils in Europe: fully evidenced, spatially organised assessment of the pressures driving soil degradation. Publications Office, Luxembourg. https://doi.org/10.2760/5897030\u003c/li\u003e\n \u003cli\u003eArsenoaia, V.-N., Rațu, R.-N., Veleșcu, I.-D., Țenu, I., 2023. Field tests of a John Deere harvester for the purpose of production maps achievement.\u003c/li\u003e\n \u003cli\u003eBabos, D.V., Tadini, A.M., De Morais, C.P., Barreto, B.B., Carvalho, M.A.R., Bernardi, A.C.C., Oliveira, P.P.A., Pezzopane, J.R.M., Milori, D.M.B.P., Martin-Neto, L., 2024. Laser-induced breakdown spectroscopy (LIBS) as an analytical tool in precision agriculture: Evaluation of spatial variability of soil fertility in integrated agricultural production systems. CATENA 239, 107914. https://doi.org/10.1016/j.catena.2024.107914\u003c/li\u003e\n \u003cli\u003eBantchina, B.B., Qaswar, M., Arslan, S., Ulusoy, Y., Gündoğdu, K.S., Tekin, Y., Mouazen, A.M., 2024. Corn yield prediction in site-specific management zones using proximal soil sensing, remote sensing, and machine learning approach. Comput. Electron. Agric. 225, 109329. https://doi.org/10.1016/j.compag.2024.109329\u003c/li\u003e\n \u003cli\u003eBarrios, E., 2007. Soil biota, ecosystem services and land productivity. Ecol. Econ., Special Section – Ecosystem Services and Agriculture 64, 269–285. https://doi.org/10.1016/j.ecolecon.2007.03.004\u003c/li\u003e\n \u003cli\u003eBaseca, C.C., Sendra, S., Lloret, J., Tomas, J., 2019. A smart decision system for digital farming. Agronomy 9. https://doi.org/10.3390/agronomy9050216\u003c/li\u003e\n \u003cli\u003eBasir, Md.S., Buckmaster, D., Raturi, A., Zhang, Y., 2024. From pen and paper to digital precision: a comprehensive review of on-farm recordkeeping. Precis. Agric. 25, 2643–2682. https://doi.org/10.1007/s11119-024-10172-7\u003c/li\u003e\n \u003cli\u003eBasso, B., Dumont, B., Cammarano, D., Pezzuolo, A., Marinello, F., Sartori, L., 2016. Environmental and economic benefits of variable rate nitrogen fertilization in a nitrate vulnerable zone. Sci. Total Environ. 545–546, 227–235. https://doi.org/10.1016/j.scitotenv.2015.12.104\u003c/li\u003e\n \u003cli\u003eBerhe, A.A., 2019. Chapter 3 – Drivers of soil change, in: Busse, M., Giardina, C.P., Morris, D.M., Page-Dumroese, D.S. (Eds.), Developments in Soil Science, Global Change and Forest Soils. Elsevier, pp. 27–42. https://doi.org/10.1016/B978-0-444-63998-1.00003-3\u003c/li\u003e\n \u003cli\u003eBorrero, J.D., Mariscal, J., 2022. A Case Study of a Digital Data Platform for the Agricultural Sector: A Valuable Decision Support System for Small Farmers. Agriculture 12, 767. https://doi.org/10.3390/agriculture12060767\u003c/li\u003e\n \u003cli\u003eCai, Y., Guan, K., Lobell, D., Potgieter, A.B., Wang, S., Peng, J., Xu, T., Asseng, S., Zhang, Y., You, L., Peng, B., 2019. Integrating satellite and climate data to predict wheat yield in Australia using machine learning approaches. Agric. For. Meteorol. 274, 144–159. https://doi.org/10.1016/j.agrformet.2019.03.010\u003c/li\u003e\n \u003cli\u003eCarolan, M., 2020. Acting like an algorithm: digital farming platforms and the trajectories they (need not) lock-in. Agric. Hum. Values 37, 1041–1053. https://doi.org/10.1007/s10460-020-10032-w\u003c/li\u003e\n \u003cli\u003eDexter, A.R., Richard, G., Arrouays, D., Czyż, E.A., Jolivet, C., Duval, O., 2008. Complexed organic matter controls soil physical properties. Geoderma 144, 620–627. https://doi.org/10.1016/j.geoderma.2008.01.022\u003c/li\u003e\n \u003cli\u003eEscribà-Gelonch, M., Liang, S., van Schalkwyk, P., Fisk, I., Long, N.V.D., Hessel, V., 2024. Digital Twins in Agriculture: Orchestration and Applications. J. Agric. Food Chem. 72, 10737–10752. https://doi.org/10.1021/acs.jafc.4c01934\u003c/li\u003e\n \u003cli\u003eHillel, D., 2013. Introduction to Soil Physics. Academic Press.\u003c/li\u003e\n \u003cli\u003eHorn, R., Fleige, H., 2009. Risk assessment of subsoil compaction for arable soils in Northwest Germany at farm scale. Soil Tillage Res., Soil Management for Sustainability 102, 201–208. https://doi.org/10.1016/j.still.2008.07.015\u003c/li\u003e\n \u003cli\u003eHuber, F., Markward, D., 2021. AgriCircle: Innovating Agricultural Ecosystems, in: Gassmann, O., Ferrandina, F. (Eds.), Connected Business: Create Value in a Networked Economy. Springer International Publishing, Cham, pp. 325–331. https://doi.org/10.1007/978-3-030-76897-3_21\u003c/li\u003e\n \u003cli\u003eInstitute for Environment and Sustainability (Joint Research Centre), Montanarella, L., Rusco, E., Tóth, G., 2008. Threats to soil quality in Europe. Publications Office of the European Union.\u003c/li\u003e\n \u003cli\u003eJacob, T., n.d. Evaluating Precision Agriculture Using Open Software Platforms.\u003c/li\u003e\n \u003cli\u003eJana, S., Chatterjee, D., Pal, N., Pal, K., Roy, K., Bashak, S., 2024. AI in Soil Health Monitoring: A Data-Driven Approach. Int. J. Res. Appl. Sci. Eng. Technol. 12, 1327–1335. https://doi.org/10.22214/ijraset.2024.64871\u003c/li\u003e\n \u003cli\u003eKeesstra, S.D., Bouma, J., Wallinga, J., Tittonell, P., Smith, P., Cerdà, A., Montanarella, L., Quinton, J.N., Pachepsky, Y., van der Putten, W.H., Bardgett, R.D., Moolenaar, S., Mol, G., Jansen, B., Fresco, L.O., 2016. The significance of soils and soil science towards realization of the United Nations Sustainable Development Goals. SOIL 2, 111–128. https://doi.org/10.5194/soil-2-111-2016\u003c/li\u003e\n \u003cli\u003eKnapic, R., 2022. Who’s Running the Farm? How a Sample of Kansas Farmers View Their Usage of Agricultural Operating Systems (M.A.).\u003c/li\u003e\n \u003cli\u003eKottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World Map of the Köppen-Geiger climate classification updated. Meteorol. Z. 259–263. https://doi.org/10.1127/0941-2948/2006/0130\u003c/li\u003e\n \u003cli\u003eLal, R., 2004. Soil Carbon Sequestration Impacts on Global Climate Change and Food Security. Science 304, 1623–1627. https://doi.org/10.1126/science.1097396\u003c/li\u003e\n \u003cli\u003eLehmann, J., Bossio, D.A., Kögel-Knabner, I., Rillig, M.C., 2020. The concept and future prospects of soil health. Nat. Rev. Earth Environ. 1, 544–553. https://doi.org/10.1038/s43017-020-0080-8\u003c/li\u003e\n \u003cli\u003eLoria, N., Lal, R., Chandra, R., 2024. Handheld In Situ Methods for Soil Organic Carbon Assessment. Sustainability 16, 5592. https://doi.org/10.3390/su16135592\u003c/li\u003e\n \u003cli\u003eLykhovyd, P., Vozhehova, R., Averchev, O., 2024. Using Remote Sensing Normalised Difference Vegetation Index to Recognise Irrigated Croplands via Agroland Classifier Application. Visnyk V N Karazin Kharkiv Natl. Univ. Ser. Geol. Geogr. Ecol. 2024, 223–233. https://doi.org/10.26565/2410-7360-2024-61-18\u003c/li\u003e\n \u003cli\u003eMedeiros, B.M., Cândido, B., Jimenez, P.A.J., Avanzi, J.C., Silva, M.L.N., 2025. UAV-Based Soil Water Erosion Monitoring: Current Status and Trends. Drones 9, 305. https://doi.org/10.3390/drones9040305\u003c/li\u003e\n \u003cli\u003eMeinen, B.U., Robinson, D.T., 2020. Mapping erosion and deposition in an agricultural landscape: Optimization of UAV image acquisition schemes for SfM-MVS. Remote Sens. Environ. 239, 111666. https://doi.org/10.1016/j.rse.2020.111666\u003c/li\u003e\n \u003cli\u003ePadhiary, M., Roy, P., Dey, P., Sahu, B., 2024. Harnessing AI for Automated Decision- Making in Farm Machinery and Operations: Optimizing Agriculture. pp. 249–282. https://doi.org/10.4018/979-8-3693-6230-3.ch008\u003c/li\u003e\n \u003cli\u003ePaleari, S., 2017. Is the European Union protecting soil? A critical analysis of Community environmental policy and law. Land Use Policy 64, 163–173. https://doi.org/10.1016/j.landusepol.2017.02.007\u003c/li\u003e\n \u003cli\u003ePanagos, P., Borrelli, P., Poesen, J., Ballabio, C., Lugato, E., Meusburger, K., Montanarella, L., Alewell, C., 2015. The new assessment of soil loss by water erosion in Europe. Environ. Sci. Policy 54, 438–447. https://doi.org/10.1016/j.envsci.2015.08.012\u003c/li\u003e\n \u003cli\u003ePandeya, S., Gyawali, B.R., Upadhaya, S., 2025. Factors Influencing Precision Agriculture Technology Adoption Among Small-Scale Farmers in Kentucky and Their Implications for Policy and Practice. Agriculture 15, 177. https://doi.org/10.3390/agriculture15020177\u003c/li\u003e\n \u003cli\u003ePham, X., Stack, M., 2018. How data analytics is transforming agriculture. Bus. Horiz. 61, 125–133. https://doi.org/10.1016/j.bushor.2017.09.011\u003c/li\u003e\n \u003cli\u003ePierce, F.J., Nowak, P., 1999. Aspects of Precision Agriculture. Adv. Agron. 67, 1–85. https://doi.org/10.1016/S0065-2113(08)60513-1\u003c/li\u003e\n \u003cli\u003eRabot, E., Wiesmeier, M., Schlüter, S., Vogel, H.-J., 2018. Soil structure as an indicator of soil functions: A review. Geoderma 314, 122–137. https://doi.org/10.1016/j.geoderma.2017.11.009\u003c/li\u003e\n \u003cli\u003eRadziuk, H., Świtoniak, M., 2022. The Effect of Erosional Transformation of Soil Cover on the Stability of Soil Aggregates within Young Hummocky Moraine Landscapes in Northern Poland. Agronomy 12, 2595. https://doi.org/10.3390/agronomy12112595\u003c/li\u003e\n \u003cli\u003eRadziuk, H., Świtoniak, M., 2021. Soil erodibility factor (K) in soils under varying stages of truncation. Soil Sci. Annu. https://doi.org/10.37501/soilsa/134621\u003c/li\u003e\n \u003cli\u003eRadziuk, H., Świtoniak, M., Nowak, M., 2021. Microscale spatial variation of soil erodibility factor (K) in a young hummocky moraine landscape in Northern Poland. Bull. Geogr. Phys. Geogr. Ser. 21, 5–16. https://doi.org/10.2478/bgeo-2021-0005\u003c/li\u003e\n \u003cli\u003eRevut, Ivan, 1972. Soil physics. Kolos, Leningrad.\u003c/li\u003e\n \u003cli\u003eRobinson, S.V.J., Schwinghamer, T., Cárcamo, H., Galpern, P., 2023. Precision agricultural data and ecosystem services: Can we put the pieces together? Ecol. Solut. Evid. 4, e12271. https://doi.org/10.1002/2688-8319.12271\u003c/li\u003e\n \u003cli\u003eRząsa, S., Owczrzak, W., 2004. Struktura gleb mineralnych -. Wydawnictwo Akademii Rolniczej im. Augusta Cieszkowskiego, Poznań.\u003c/li\u003e\n \u003cli\u003eSheng, Y., Sun, Z., Lu, S., Omasa, K., 2024. Ratio of physical model parameters can retrieve aggregate size from different types of soil in cultivated regions. Soil Tillage Res. 244, 106262. https://doi.org/10.1016/j.still.2024.106262\u003c/li\u003e\n \u003cli\u003eSix, J., Bossuyt, H., Degryze, S., Denef, K., 2004. A history of research on the link between (micro)aggregates, soil biota, and soil organic matter dynamics. Soil Tillage Res., Advances in Soil Structure Research 79, 7–31. https://doi.org/10.1016/j.still.2004.03.008\u003c/li\u003e\n \u003cli\u003eSmith, P., House, J.I., Bustamante, M., Sobocká, J., Harper, R., Pan, G., West, P.C., Clark, J.M., Adhya, T., Rumpel, C., Paustian, K., Kuikman, P., Cotrufo, M.F., Elliott, J.A., McDowell, R., Griffiths, R.I., Asakawa, S., Bondeau, A., Jain, A.K., Meersmans, J., Pugh, T.A.M., 2016. Global change pressures on soils from land use and management. Glob. Change Biol. 22, 1008–1028. https://doi.org/10.1111/gcb.13068\u003c/li\u003e\n \u003cli\u003eSoil Health Improvement | FAO SOILS PORTAL | Food and Agriculture Organization of the United Nations [WWW Document], n.d. URL https://www.fao.org/soils-portal/soil-degradation-restoration/soil-health-improvement/en/ (accessed 6.4.25).\u003c/li\u003e\n \u003cli\u003eSolon J., Borzyszkowski J., Bidіasik M., Richling A., Badora K., Balon J., Brzeziсska-Wуjcik T., Chabudziсski Ј., Dobrowolski R., Grzegorczyk I., Jodіowski M., Kistowski M., Kot R., Lechnio J., Macias A., 2018. Physico-geographical mesoregions of Poland – modified version of J. Kondracki’s regionalisation. Geogr. Pol. 91. https://doi.org/10.7163/GPol.0115\u003c/li\u003e\n \u003cli\u003eŚwitoniak, M., 2014. Use of soil profile truncation to estimate influence of accelerated erosion on soil cover transformation in young morainic landscapes, North-Eastern Poland. CATENA 116, 173–184. https://doi.org/10.1016/j.catena.2013.12.015\u003c/li\u003e\n \u003cli\u003eŚwitoniak, M., Mroczek, P., Bednarek, R., 2016. Luvisols or Cambisols? Micromorphological study of soil truncation in young morainic landscapes – Case study: Brodnica and Chełmno Lake Districts (North Poland). CATENA 137, 583–595. https://doi.org/10.1016/j.catena.2014.09.005\u003c/li\u003e\n \u003cli\u003eŚwitoniak M., Nowak M., Radziuk H., 2020. Low-altitude photogrammetry in studies of soil cover variability of areas transformed by anthropogenic denudation., in: Współczesne Problemy i Kierunki Badawcze w Geografii. Instytut Geografii i Gospodarki Przestrzennej Uniwersytetu Jagiellońskiego, Kraków, p. S. 81-100.\u003c/li\u003e\n \u003cli\u003eVedeneeva, V.A., Koshelev, A.V., 2023. Analysis of the seasonal dynamics of the NDVI index of potato agrocenosis in the conditions of the dry steppe zone of the Volgograd region. Res. Crops 24, 366–372. https://doi.org/10.31830/2348-7542.2023.ROC-935\u003c/li\u003e\n \u003cli\u003eWilliams, J.R., 1990. The Erosion-Productivity Impact Calculator (EPIC) Model: A Case History. Philos. Trans. Biol. Sci. 329, 421–428.\u003c/li\u003e\n \u003cli\u003eZagorda, M., Walczykova, M., 2018. The application of various software programs for mapping yields in precision agriculture, in: SzelagSikora, A. (Ed.), CONTEMPORARY RESEARCH TRENDS IN AGRICULTURAL ENGINEERING, BIO Web of Conferences. Presented at the Conference on Contemporary Research Trends in Agricultural Engineering, E D P Sciences, Cedex A, p. 01018. https://doi.org/10.1051/bioconf/20181001018\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Kazimierz Wielki University in Bydgoszcz","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Diagnostic void, precision agriculture, digital agriculture, soil degradation, sustainability assessment, technology adoption barriers","lastPublishedDoi":"10.21203/rs.3.rs-7541765/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7541765/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSoil degradation is a primary threat to agricultural sustainability, yet addressing it is hampered by high spatial variability of soil properties. While Precision Agriculture (PA) platforms promise tools for site-specific management, their capacity to diagnose the root causes of poor soil health, such as physical degradation, remains critically unassessed.\u003c/p\u003e\n\u003cp\u003eThis study confronts this issue by empirically testing the dominant „symptom-based” paradigm of commercial PA. We conducted a multi-phase assessment of 34 platforms, including an in-depth user-experience (UX) test of nine major platforms, using a unique, field-verified dataset of soil physical health indicators (aggregate stability, erodibility K-factor) from a representative hummocky moraine landscape in Poland.\u003c/p\u003e\n\u003cp\u003eOur findings reveal a profound „diagnostic void” in the commercial PA ecosystem. While we confirmed a strong statistical correlation between soil degradation indicators (the „cause”) and vegetation indices like NDVI (the „symptom”), we discovered that none of the tested platforms offered built-in tools for soil structure or erosion analysis. This failure was compounded by a convergence of technical (e.g., lack of GeoTIFF support), economic (paywalls), and ecosystem-level barriers that systematically prevent the integration of user-generated scientific soil data.\u003c/p\u003e\n\u003cp\u003eThe current generation of PA platforms is fundamentally limited to treating symptoms rather than diagnosing causes, forcing users into a reactive and potentially unsustainable management paradigm. We argue that this diagnostic gap is not a mere technological oversight but a systemic failure driven by market priorities. Bridging this gap requires a fundamental reorientation of the ag-tech sector towards open, interoperable, and analytically robust platforms that empower science-based soil stewardship.\u003c/p\u003e","manuscriptTitle":"The Diagnostic Void: A Systemic Failure of Commercial Precision Agriculture Platforms to Integrate Science-Based Soil Health Indicators","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-08 21:16:36","doi":"10.21203/rs.3.rs-7541765/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e09701a5-8efa-4e7f-9398-6fa3a0bbcb69","owner":[],"postedDate":"September 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":54352983,"name":"Agroecology"},{"id":54352984,"name":"Geographic Information Systems"},{"id":54352985,"name":"Environmental Policy"}],"tags":[],"updatedAt":"2025-09-08T21:16:37+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-08 21:16:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7541765","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7541765","identity":"rs-7541765","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}