Next Article in Journal
Quantitative Morphological Resolution of Preservation–Renewal Conflicts for “Shanghai-Style Jiangnan” Villages, China
Previous Article in Journal
Land Use Classification in Rare Earth Mining Areas Based on Multi-Source Remote Sensing and Feature Optimization
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Land
Volume 15
Issue 5
10.3390/land15050794
land-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Land Degradation Assessment in an Olive Orchard Using Different Soil Erosion Estimation Methods

by
Christos Pantazis
1,2,3 and
Panagiotis T. Nastos
1,4,*
1
Department of Geology and Geoenvironment, National and Kapodistrian University of Athens, 15784 Athens, Greece
2
Research Centre of Atmospheric Physics and Climatology, Academy of Athens, 10680 Athens, Greece
3
Navarino Environmental Observatory (N.E.O.), 24001 Messenia, Greece
4
Biomedical Research Foundation of the Academy of Athens, 11527 Athens, Greece
*
Author to whom correspondence should be addressed.
Land 2026, 15(5), 794; https://doi.org/10.3390/land15050794
Submission received: 1 March 2026 / Revised: 8 April 2026 / Accepted: 5 May 2026 / Published: 8 May 2026
(This article belongs to the Section Land – Observation and Monitoring)
Browse Figures
Versions Notes

Abstract

Land degradation caused by soil erosion is a major challenge in Mediterranean sloping agroecosystems, where extreme weather events and conventional land management practices accelerate soil loss and threaten long-term sustainability. This study evaluates and compares three complementary approaches to estimate soil erosion in an olive orchard in Messenia, Greece. Field-based runoff plots provided direct measurements of sediment yield, drone-based Light Detection and Ranging (LiDAR) surveys enabled soil surface change detection through the Difference of Digital Elevation Models (DoD) method, and the Revised Universal Soil Loss Equation (RUSLE) was applied to model erosion risk using site-specific parameters. Results indicate that field measurements and RUSLE estimates are broadly consistent, particularly when the model is calibrated with empirical data, offering reliable insights into soil loss dynamics. In contrast, the LiDAR-DoD analysis identified patterns of soil surface displacement, which reflected spatial variation in surface change across the olive orchard. Overall, the integration of field monitoring, remote sensing, and modeling highlights the strengths and limitations of each method and demonstrates the value of multi-method approaches for improving erosion assessment and supporting sustainable land management in vulnerable Mediterranean landscapes.

1. Introduction

Land degradation is a major threat to soil fertility [1], agricultural productivity [2], water quality [3], and ecosystem stability worldwide [4], with soil erosion by water representing one of its most important forms. It is especially critical in Mediterranean environments, where steep topography, intense seasonal rainfall, sparse vegetation cover during part of the year, and long-term agricultural use create favorable conditions for runoff generation and soil detachment [5,6,7]. In these landscapes, erosion is not only a geomorphological process but also a major environmental and socio-economic issue because it reduces the productive capacity of soils and contributes to the progressive degradation of cultivated land [8,9]. Olive-growing areas are particularly relevant in this context, as they represent one of the dominant land uses in many Mediterranean regions and are often located on sloping terrain that is vulnerable to erosion [7,10].
The Mediterranean climate is characterized by a strong seasonal contrast, with dry summers and rainfall concentrated mainly in autumn and winter [11]. Under these conditions, the erosive impact of rainfall is often intensified by the limited protective cover of the soil surface at the beginning of the rainy season. In many olive orchards, especially those managed conventionally, the soil between tree rows may remain bare or only partly covered for long periods, increasing its exposure to raindrop impact and overland flow [12]. Repeated tillage, low organic matter content, and inadequate conservation practices may further increase soil susceptibility to erosion [13,14]. As a result, Mediterranean olive groves have frequently been identified as high-risk areas for soil degradation, sediment redistribution, and long-term decline in soil quality [15].
Because soil erosion is controlled by the interaction of multiple factors, including rainfall, soil properties, slope, vegetation cover, and land management [16], its assessment remains methodologically challenging [17]. No single approach can fully capture all dimensions of the process. Direct field measurements provide valuable information on actual soil loss under real conditions, but they are often limited to small plots and specific time periods [18]. Empirical models, on the other hand, make it possible to estimate erosion over larger areas and to explore its spatial distribution, although their reliability depends strongly on the quality of the input data and on local validation [19]. In recent years, high-resolution topographic techniques such as Light Detection and Ranging (LiDAR) and photogrammetry have added a new dimension to erosion studies by allowing the detection of subtle changes in surface elevation and local redistribution patterns [20]. These different approaches are therefore complementary rather than interchangeable.
The Revised Universal Soil Loss Equation (RUSLE) is an empirical model used to estimate average annual soil loss caused by water erosion based on rainfall erosivity, soil erodibility, topographic characteristics, land cover, and conservation practices [21]. Due to its broad applicability and relatively simple parameterization, RUSLE has become one of the most widely used approaches for soil erosion estimation [22], making it particularly suitable for comparison with field-based measurements in the present study. Its widespread use is related to its relatively simple structure, modest data requirements compared with more physically based models, and compatibility with Geographical Information Systems (GIS)-based spatial analysis [23]. For this reason, RUSLE has been used extensively in watershed studies, agricultural landscapes, and erosion risk mapping applications. However, its use also requires caution. Since RUSLE is an empirical model, its predictions depend heavily on the representativeness and spatial accuracy of the input factors, and its application is more robust when supported by local information and field observations. Validation is therefore essential if model outputs are to be interpreted with confidence [24].
At the same time, field experiments remain one of the most direct ways to quantify real soil loss from a defined area. Measurements obtained from monitored plots can provide useful evidence of temporal variability and event-driven erosion, particularly in cultivated systems where soil loss may occur in short pulses associated with rainfall episodes [18]. Such data are particularly valuable because they offer an empirical basis against which model results can be compared. Nevertheless, field measurements alone cannot describe the spatial distribution of erosion beyond the monitored plot, nor can they easily reveal how sediment is redistributed across the surface [25]. This limitation has encouraged the integration of field observations with spatial models and terrain-based techniques.
In this context, LiDAR-derived digital elevation models (DEMs) and DEM of Difference (DoD) analysis provide an additional tool for investigating soil surface dynamics at very fine scales. By comparing elevation models acquired at different times, DoD analysis can identify areas of local surface lowering and raising, thereby helping to describe short-term terrain change and soil redistribution [26]. While this method does not necessarily measure soil loss directly, it is particularly useful for visualizing surface behavior and the movement of material within small experimental plots. As such, it can complement both direct field measurements and model-based predictions by adding information on local microtopographic change.
The present study was developed within this broader methodological context and aimed to assess soil erosion in a Mediterranean olive-growing sub-catchment in southwestern Greece using a combination of approaches. More specifically, the study integrates (i) direct field measurements of soil loss from an experimental subplot, (ii) spatial estimation of soil erosion using the RUSLE model, and (iii) LiDAR- and DoD-based analysis of short-term surface elevation changes. The decision to focus on a small, well-defined sub-catchment was made to improve the spatial reliability of the RUSLE application and to allow direct comparison between model outputs and field observations at the location of the experimental plot. This design also makes it possible to examine how different methods perform when applied to the same erosion-prone environment.
The study, therefore, aims not only to estimate soil erosion but also to compare three different assessment approaches within the same sloping Mediterranean olive orchard, where erosion is strongly event-driven. The results indicate that runoff plots provide direct measurements of soil loss at the event scale, RUSLE yields satisfactory estimates when supported by local calibration data, and LiDAR-DoD contributes primarily to the spatial interpretation of surface change and sediment redistribution. The main contribution of the study thus lies not in the individual application of these methods, but in their combined evaluation within the same agroecosystem, highlighting how their complementarity can support a more complete understanding of erosion processes.

2. Materials and Methods

The study was carried out in a sub-catchment (0.221 km2) of the Xerias River watershed (about 60 km2), near the coastal area of Navarino gulf in Pylos-Nestor Municipality of Messenia, southwestern Greece (Figure 1). Although the broader Xerias watershed covers a much larger area, the present analysis focused on a smaller section that was selected according to local topography which defines a clear and practical boundary for modelling purposes. This smaller unit was considered appropriate for the application of the RUSLE model, since it includes the experimental plots used in field study and represents the terrain conditions most relevant to the objectives of this work.
The landscape is typical of Mediterranean environments, with gentle to moderately sloping terrain, small drainage pathways, and agricultural land mixed with patches of natural vegetation. The dominant land use in the area is olive cultivation, accompanied by shrubland and scattered woodland vegetation. Soils are mainly calcareous Mediterranean soils, generally ranging from loam to clay loam [27], and they have a moderate to relatively high susceptibility to erosion, especially where vegetation cover is limited. In addition, soil samples from the experimental field plot were analyzed in the beginning of the experiment for selected chemical and physical properties, including organic matter content [28] and texture [29]. Soil texture was classified according to the USDA soil texture triangle [30].
The climate of the region is Mediterranean, with precipitation concentrated mainly in the wet season from autumn to spring, while summers are hot and dry [31,32]. Average annual precipitation during the 2016–2025 reference period was 806 mm, according to the local weather stations network. Under these conditions, soil loss is strongly influenced by seasonal rainfall intensity, slope, land cover, and management practices [14]. The selected sub-catchment is a representative and well-defined area for investigating erosion processes at the local scale.

2.1. Field Monitoring of Runoff and Sediment

To anchor the modelling in empirical evidence under local orchard conditions, we used runoff and sediment data from a field experiment during October 2023–March 2025. More specifically, the experiment employed bounded runoff plots that route surface runoff and associated sediment to a collector and storage tanks, enabling event-scale quantification of runoff volume and sediment yield. Such plots are widely used in Mediterranean olive systems to measure gross soil loss and compare management practices, although known scale and border effects should be considered when interpreting results [7].
A randomized block design was used in the experiment, consisting of nine plots (10 m × 10 m) established on a representative hillslope within an olive grove. Three soil management treatments were evaluated, each with three replicates: herbicide treatment, cover cropping, and mowing of natural vegetation. In the present study, we report and analyze data from one plot corresponding to the herbicide treatment (one of the nine plots). This plot was selected because it was the only replicate within the herbicide treatment equipped with a surface runoff collection system, due to the high cost and complexity of the installation, which limited its implementation to one plot per treatment. In addition, the herbicide-treated plot belongs to the olive grove land-cover class considered in the assignment of the cover-management factor (C), while it was essential for the DoD analysis because reduced vegetation improved LiDAR surface detection. The plot was located on a slope segment with a mean gradient of ~16% and was equipped at its downslope boundary with (a) a collection channel/trough spanning the plot width and (b) a sealed storage tank positioned downslope to retain runoff and sediment from each rainfall event (Figure 2). This configuration follows standard plot-collection principles for erosion experiments, including ensuring free discharge into the tank and sufficient storage capacity for infrequent high-magnitude storms. Although the experimental plot area differs from the standard erosion plot dimensions defined by [33], the selected plot size was considered appropriate for preserving the representativeness of the local olive orchard system, including tree architecture and within-plot heterogeneity.
After each rainfall event that generated measurable surface runoff and sediment transport to the collection tank, the total runoff volume was quantified using a collection tank with a storage capacity of 1 t. The sediment-laden water was then processed to determine sediment yield for each event. Deposited material was first re-suspended and then either sub-sampled or fully recovered, depending on the volume collected. The sediment was subsequently oven-dried to constant weight and weighed, following standard plot-scale water erosion protocols [34]. Event soil loss was calculated as dry sediment mass divided by plot area and reported in t ha−1 event−1, with seasonal totals obtained by summing events over the monitoring period (typically October-April in Mediterranean climates). These measurements were used to (i) document the magnitude and temporal variability of erosion under the herbicide-managed orchard condition and (ii) provide an external check on whether GIS-based RUSLE predictions fall within a plausible local range at the hillslope scale. Technical problems in field data collection of the first year caused gaps in the measurement series, and some intermediate erosive events may therefore not have been captured.

2.2. RUSLE Framework and GIS Implementation

Spatial modelling was conducted using the RUSLE multiplicative structure:
A = R × K × LS × C × P
where A is average annual soil loss (long-term mean) from rill and interrill processes, R is rainfall erosivity, K is soil erodibility, LS is the combined slope length and slope steepness factor, C is the cover-management factor, and P is the support practice factor [33].
All these factors affect soil loss [21], and the layers were prepared as gridded rasters at a consistent 5 m spatial resolution, aligned to a common projection and extent. The rasters were then combined using cell-by-cell multiplication to generate a continuous soil loss surface across the study sub-catchment. The GIS workflow followed standard map-algebra practice: (i) preprocess primary inputs (rainfall data, soil properties, DEM, and land cover), (ii) derive R, K, LS, C, and P layers with consistent units and scaling, and (iii) compute A from their product [35] (Figure 3).
Because RUSLE predicts long-term average annual soil loss rather than event sediment delivery at the catchment outlet, outputs were interpreted as gross erosion potential (sheet and rill) at the hillslope scale. Measured plot erosion therefore provides a valuable consistency check, but it is not expected to match model output exactly, particularly when monitoring is limited in duration and when local deposition or rill initiation thresholds affect sediment transfer [22].

2.2.1. Rainfall Erosivity Factor

Rainfall erosivity represents the erosive power of rainfall and its capacity to generate runoff capable of detaching and transporting soil. According to [33], erosivity is derived from rainfall intensity metrics (e.g., EI30), but such high-frequency records are often unavailable for many Mediterranean monitoring networks. As an alternative, erosivity can be approximated from precipitation totals aggregated at monthly or annual scales [36].
In this study, R was derived from the modified Fournier framework described by [37] which documents [38] Fournier’s original precipitation concentration index to improve correlation with erosivity. For each station, daily rainfall records were aggregated to monthly totals and annual totals. The modified Fournier index (MFI) was computed as:
M F I = Σ 1 12 P i 2 P
where Pi is precipitation for month i and P is annual precipitation.
Following the relationship reported by [39], erosivity was estimated as:
R = 0.264   M F I 1.50
where 0.264 is an empirically derived coefficient obtained from the statistical relationship between MFI and rainfall erosivity (R), ensuring consistency with standard R-factor units.
This precipitation-based erosivity approach is explicitly recommended for application within homogeneous climatic settings, because the F-R relationship can vary across rainfall regimes [36]. Accordingly, TEMES network (Figure 4) of 4 rain gauges was selected to represent the same regional climatic domain as the target sub-catchment, and multi-year averaging (2016–2025) was performed to suppress single-year anomalies and better approximate long-term erosive forcing.
Station-level R-factor values were interpolated across the catchment to produce a continuous rainfall erosivity raster at 5 m spatial resolution, consistent with DEM resolution. Interpolation was performed with the IDW (Inverse Distance Weighted) method appropriate to station density and topographic variability, and the resulting raster was clipped to the study-area boundary. The resulting R map was treated as time-invariant for the modelling period, consistent with the long-term average nature of RUSLE.

2.2.2. Soil Erodibility Factor

Soil erodibility (K) describes how easily soil can be detached and carried away by rainfall and runoff under standard reference conditions. It mainly depends on soil texture, organic matter, structure, and how quickly water can move through the soil [35]. K is usually estimated either from standard-plot measurements or from equations that relate K to measured soil properties (pedotransfer methods).
In this study, the required soil inputs (soil texture, organic matter, soil-structure class, and permeability class) were obtained from the OPEKEPE open soil datasets (Ministry of Rural Development and Food). These spatial layers were used to parameterize the K-factor across the study sub-catchment, ensuring consistent coverage and attribute definitions across land-cover units. To confirm that the mapped properties were representative locally, the OPEKEPE values were validated using field observations and laboratory analyses from soil samples collected at the experimental plot. Agreement was then checked for texture class, organic matter levels, and the assigned structure and permeability classes.
The erodibility factor was computed using the widely cited algebraic approximation of the original USLE nomograph developed by [33,35] and reported in metric-converted form (including the 0.1317 unit conversion multiplier [22]) in authoritative documentation. The equation applied was:
K = 0.1317 × ( 2.1 × M − 1.14 × 10 − 4 × ( 12 − a ) + 3.25 ( b − 2 ) + 2.5 ( c − 3 ) ) / 100
where M is the texture term M = (msilt + mvfs) (100 − mc), with msilt = percent silt, mvfs = percent very fine sand, and mc = percent clay; a is organic matter (%); b is the soil-structure class code (b = 1 for very fine granular, b = 2 for fine granular, b = 3 for medium or coarse granular and b = 4 for blocky, platy or massive); and c is the permeability class code (c = 1 for very rapid and c = 6 for very slow). Six classes are distinguished [40]: (1) rapid, (2) moderate to rapid, (3) moderate, (4) moderate to slow, (5) slow and (6) very slow. The numerical coefficients in the equation are empirically derived from the original Wischmeier and Smith nomograph [33] and represent the relative influence of soil texture, organic matter, structure, and permeability on soil erodibility, while 0.1317 is a unit conversion factor to SI units.
A catchment-wide K raster at 5 m resolution was generated by linking computed K values to mapped soil units. The approach ensured that K varied spatially according to dominant soil property patterns while maintaining physical plausibility (e.g., higher K in fine-textured, low-OM soils). This aligns with the conceptual interpretation of K as a soil-property-controlled modifier of erosion susceptibility in RUSLE-type formulations.

2.2.3. Topographic Factor: Slope Length and Steepness

Topography strongly affects erosion because it controls how runoff concentrates and how much energy the flowing water has. In GIS, slope-length effects are often represented using upslope contributing area (the area draining into each cell). This lets the LS factor reflect flow convergence and divergence on complex terrain instead of assuming a uniform hillslope. Accordingly, the combined LS factor was computed using a unit stream power/contributing-area formulation linked to [41].
L S = ( A s / 22.13 ) m × ( s i n β / 0.0896 ) n
where As is the specific catchment area (upslope contributing area per unit contour width), β is the slope angle, m = 0.4 and n = 1.3. The standard slope length of the RUSLE plot is 22.13 m, and 0.0896 corresponds to the sine of the standard 9% slope; both values are used to normalize slope length and steepness in the LS factor calculation.
A hydrologically corrected DEM was used to compute flow direction, flow accumulation, and slope. The contributing-area term As was derived by combining flow accumulation with grid-cell size (i.e., converting accumulated cell counts to contributing area per unit width), while slope was computed in degrees and converted as required for trigonometric functions. This approach captures how runoff concentrates in converging areas of the terrain, which is a key advantage formulation compared with methods that represent slope length only [42].

2.2.4. Cover-Management Factor

Vegetation and management reduce soil loss by shielding the soil from raindrop impact, increasing hydraulic roughness, and improving infiltration. In RUSLE, this is captured by the cover-management factor C.
Land cover in the study sub-catchment was mapped into a small number of dominant classes that reflect the local landscape (olive groves, sclerophyllous vegetation, and mixed agricultural land with significant natural vegetation). Because the sub-catchment is small and well known, the classification was produced using local knowledge and detailed interpretation. Values for each class (Table 1) were assigned from published literature values reported by [43].

2.2.5. Support Practice Factor

The support practice factor (P) reflects the effect of erosion-control practices in reducing soil loss relative to areas without any supporting conservation measure. In this study, the P factor was assigned following the standard procedure of the RUSLE model [33], based on field verification of existing conservation practices in the study area. Field investigation showed that terracing was the only support practice present in the sub-catchment, while the remaining areas had no identified erosion-control practice due to the prevailing landscape and land-use conditions. Therefore, all non-terraced areas were assigned P = 1.0. For terraced areas, P values were assigned according to slope classes for contour-farmed terraced fields following by [33]: 0.60 for slopes of 1–2%, 0.50 for 3–8%, 0.60 for 9–12%, 0.70 for 13–16%, 0.80 for 17–20%, 0.90 for 21–25%, and 1.0 for slopes >25%.

2.3. DEMs of Difference Methodology

DoD methodology was applied to describe the soil movement within the subplots. The LiDAR sensor has the advantage that vegetation cover can be filtered from the image, allowing detection of soil behavior during a long period with precipitation. Repeat UAV-LiDAR surveys were conducted to derive spatially explicit erosion/deposition estimates. Data acquisition used a DJI Matrice 300 RTK platform carrying a DJI Zenmuse L2 LiDAR payload, manufactured by DJI, Shenzhen, China. The UAV-LiDAR dataset was georeferenced using the RTK-enabled direct positioning capabilities of the Matrice 300 RTK; therefore, no dedicated Ground Control Points (GCPs) were established, as this configuration was considered adequate for the accuracy requirements of the study. The Zenmuse L2 integrates LiDAR with a high-accuracy IMU and supports multi-return acquisition (capturing up to five return signals from each laser pulse). In orchard environments, tree crowns and ground vegetation can partially block the soil surface in image-based reconstructions. Multi-return LiDAR increases the likelihood of obtaining ground observations through canopy gaps [44]. In contrast, UAV photogrammetry may fail to reconstruct “dead ground” beneath vegetation because it relies on optical visibility [45].
Two flight surveys carried out on 11 December 2024 and 6 May 2025, and each point cloud was processed to generate a gridded DEM at 0.05 m resolution. Point-cloud reconstruction and terrain modelling were performed in DJI Terra. Ground elevations were not derived simply by selecting the “last return.” Instead, a ground-point classification procedure was applied to distinguish bare-earth points from vegetation and other objects. DJI Terra’s dedicated ground-classification workflow was used, with terrain-dependent parameter settings, to generate the DEM from the classified ground points. This distinction matters because ground filtering/classification is a recognized source of uncertainty in terrain modelling under canopy and on complex slopes.
Topographic change between the two dates was quantified using a DEM-of-Difference (DoD):
DoD = DEM_May2025 − DEM_Dec2024
Following this definition, negative DoD values represent elevation loss (surface lowering/erosion) and positive values represent deposition. The DoD approach is widely used for morphological sediment budgets but requires attention to survey/model uncertainty and the possibility that apparent changes reflect noise rather than geomorphic signal [46]. DoD values (Δz, m) were converted to per-cell volumes using the grid-cell area:
ΔVᵢ = Δzᵢ × A
where A = 0.05 m × 0.05 m = 0.0025 m2.
Summing ΔVᵢ across all cells yielded the net volumetric change (m3) over the observation period. Net volume was then converted to mass using the measured average bulk density from the plot (ρb = 1.1 g cm−3, equivalent to 1.1 t m−3 by unit conversion), giving
M = V × ρb (t)
Soil bulk density used to convert net volumetric change to mass was determined from two undisturbed core samples collected at 0–5 cm in the experimental plot using cylinders of 5 cm height [47]. Bulk density was calculated following [48] and corrected for the coarse fraction (>2 mm), according to:
B D = W c − W g V c
where Wc is the oven-dry mass of soil in the cylinder (g), Wg is the mass of gravel (>2 mm) (g), and Vc is the cylinder volume (cm3).

3. Results

3.1. Field-Based Measurements of Soil Loss

The direct measurements collected in the 100 m2 experimental subplot revealed marked temporal variability in soil loss during the two monitoring periods. Soil analysis of the experimental plots showed a clay loam texture, consisting of 32% sand, 32% silt, and 36% clay, with an organic matter content of 1.9%. To better interpret the observed sediment responses, each measurement event was complemented with rainfall descriptors, including total event precipitation and maximum 30 min rainfall intensity (I30), as an event-scale indicator of rainfall erosivity (Figure 5a).
Rainfall during the monitoring period (2023–2025) had a mean annual precipitation of 785 mm, which was 2.6% lower than the reference period (2016–2025) of 806 mm. This indicates that the study period was slightly drier than average but still broadly representative of typical rainfall conditions in the study area. The rainfall record shows that monitored events differed substantially in both total precipitation and intensity. Total precipitation per event ranged approximately from 15 to 95 mm, whereas I30 varied from about 7 to 69 mm h−1. This indicates that the monitored period included events of contrasting erosive potential, from relatively low-intensity storms to short, highly intense rainfall episodes.
Measured soil loss also showed strong event-to-event variability (Figure 5b). The highest values were recorded during a limited number of events, while several other dates produced only minor sediment yield. This pattern confirms that soil loss in the experimental subplot was event-based rather than continuous, with a few rainfall episodes accounting for most of the total observed sediment export.
For the 2023–2024 monitoring period, the cumulative measured soil loss was 0.497 t ha−1 yr−1. In the 2024–2025 monitoring period, which provided a more complete record, the cumulative measured soil loss reached 0.768 t ha−1 yr−1. Overall, the field measurements indicate that sediment exports from the subplot were dominated by discrete erosive events rather than by gradual and uniform soil removal over time.

3.2. RUSLE Model Estimates of Soil Erosion

Rainfall erosivity (R): The R-factor map (Figure 6a) shows that rainfall erosivity is almost uniform across the study sub-catchment. This is expected because the area has a typical Mediterranean climate, with most rainfall occurring during the wet season and with no major spatial differences over such a small area. Slightly elevated R values appear on the more exposed parts of the catchment, while reduced values occur in the lower-lying areas. This pattern suggests that rainfall intensity may be somewhat greater at higher elevations. Although these differences are not large, they are still important because the R factor directly affects the final soil loss values in the RUSLE model. Therefore, even small increases in R can lead to higher predicted erosion, especially where steep slopes or limited vegetation cover are also present. Overall, the map indicates that rainfall provides a relatively consistent erosive force across the sub-catchment, with only minor local variation.
Soil erodibility (K): The K-factor map (Figure 6b) shows very limited spatial variation across the study sub-catchment, with values remaining close to 0.0249 t·ha·h ha−1 MJ−1 mm−1. This relatively uniform pattern is consistent with the available soil survey data from the Ministry of Agriculture [27], which indicates homogeneous soil properties within the study area. The dominant soil material is characterized by 35.3% sand, 30.0% silt, and 34.7% clay, with an average organic matter content of 2.61%. This composition corresponds approximately to a clay loam texture and agrees with the results of our field soil measurements. These properties suggest moderate susceptibility of the soil to detachment by rainfall and runoff. Therefore, although the spatial variability of K is low, soil erodibility still contributes to erosion risk, particularly in areas where it interacts with steeper slopes as indicated in DEM (Figure 6c) and limited vegetation cover.
Topographic factor (LS): The LS-factor map (Figure 6d) shows the strongest spatial variability among all RUSLE factors, with values ranging from 0 to about 43.1. The highest values are concentrated along steeper and longer slope segments, where runoff can accumulate and gain erosive energy. In contrast, low LS values occur in flatter parts of the sub-catchment and in short slope sections, where flow concentration is limited. This pattern indicates that topography is a major control on the spatial distribution of erosion risk in the study area. Areas with elevated LS values are expected to experience greater soil loss, especially when combined with sparse vegetation cover or limited support practices.
Cover-management (C): The C-factor map (Figure 6e) presents clear spatial differences related to land cover and management, with values ranging from 0.0623 to 0.2273. Lower C values are associated with areas that have denser vegetation or better ground protection, indicating a lower susceptibility to erosion. By contrast, higher C values occur in more exposed surfaces, including cultivated land or orchard areas with reduced understory cover. This pattern shows that vegetation cover plays an important protective role in the sub-catchment. In practical terms, areas with higher C values are more vulnerable to erosion because the soil surface is less protected against raindrop impact and overland flow.
The spatial distribution of the support-practice factor (P) is shown in Figure 6f. P values vary between 0.6 and 1.0, reflecting the presence or absence of erosion-control practices within the sub-catchment. Areas assigned P = 1.0 represent land with no identified support practice and therefore no reduction in soil loss relative to the standard condition. In contrast, areas with P = 0.6 correspond to terraced fields, where conservation measures contribute to lowering erosion risk. The map indicates that these lower P values are confined to specific parts of the sub-catchment, particularly in the central to northern zones and in smaller patches toward the southeast, whereas most of the area remains characterized by P = 1.0. This confirms that the protective effect of support practices is spatially limited and mainly associated with terraced land.
The final RUSLE soil loss map (Figure 7) shows a clear spatial variation in estimated annual soil erosion across the study sub-catchment. Most of the area falls within the lowest erosion class (0.0–0.6 t ha−1 yr−1), indicating generally low to moderate soil loss under the present environmental and land-use conditions. However, several distinct zones of increased erosion risk are visible, mainly in the central, northwestern, and southeastern parts of the sub-catchment, where soil loss values rise to 1.3–3.2 t ha−1 yr−1, and locally reach the highest class of 3.4–8.5 t ha−1 yr−1. Such areas are likely associated with the combined effect of steeper slopes, greater flow concentration, reduced vegetation cover, and less effective support practices [22]. Overall, the results suggest that although the sub-catchment is relatively small, it contains localized erosion-prone areas that may require targeted soil conservation measures.

3.3. Terrain Elevation Change (DEM of Difference)

To investigate short-term changes in soil surface morphology, a DEM of Difference (DoD) approach was applied using multitemporal LiDAR-derived elevation data. The analysis focused on identifying patterns of surface displacement and soil movement between survey dates rather than directly quantifying soil loss.
A key preprocessing step involved generating a bare-earth DEM through vegetation filtering. The comparison between the raw and filtered DEMs (Figure 8) demonstrates that vegetation removal significantly improves the representation of ground surface morphology, allowing clearer identification of slope features, microtopography, and flow pathways. This step was essential for ensuring that subsequent elevation differences reflected actual terrain changes.
Comparison of the DEMs from 11 December 2024 and 6 May 2025 (Figure 9) indicates that, while the overall topographic structure remained consistent, localized differences in surface elevation occurred across the plot. These variations suggest short-term redistribution of soil material along microtopographic features.
The DoD map (Figure 10) shows both positive and negative elevation changes, reflecting spatially heterogeneous patterns of erosion and deposition within the plot. Changes are mainly associated with small-scale terrain features and flow pathways, indicating localized sediment redistribution over short distances.
In this study, DoD results are interpreted primarily as indicators of soil displacement and surface dynamics rather than as direct estimates of soil loss. Overall, the DoD qualitatively suggests net sediment loss across the plot, although part of this signal may reflect internal sediment redistribution rather than actual export.

4. Discussion

The main strength of this study is that soil erosion was examined using a multi-method approach, combining direct field measurements, RUSLE modelling, and LiDAR/DoD-based surface analysis. This is important because soil erosion is a complex process [49] influenced by multiple factors [50] acting across different spatial and temporal scales; therefore, a single method is rarely sufficient to capture its full behavior. Recent studies from Mediterranean environments show that runoff plots provide direct field evidence of sediment loss, while GIS- and remote sensing-based models represent the spatial distribution of erosion risk; together, these approaches offer a more complete understanding of erosion processes [51,52]. Runoff plots are especially valuable because they provide direct event-scale measurements of gross soil loss [7,53], while RUSLE offers spatially distributed predictions of erosion risk [19,54,55], and DoD methods [56] help visualize how the soil surface changes through time.
Among the methods tested, the runoff-plot experiment provided the most direct measured evidence of soil loss from the monitored subplot [52,57]. Runoff plots are widely used to obtain direct measurements of runoff and soil loss from bounded areas, especially at the event scale [58,59]. However, their results are sensitive to plot design, operation, and local spatial variability, and should therefore be interpreted with due caution [58,60]. In the present study, the runoff-plot data served as the main measured reference for event-driven soil loss. Technical problems in the collection system caused some missing observations during the first monitoring year. Since this period did not coincide with intense rainfall events, major soil-loss events were unlikely to have been missed.
The RUSLE results showed good agreement with the field measurements once the comparison was made at the appropriate spatial scale. This is one of the most important outcomes of this study. The measured annual soil loss values of 0.497 t ha−1 for 2023–2024 and 0.768 t ha−1 for 2024–2025 fall within, or very close to, the 0.6–1.3 t ha−1 yr−1 range predicted by RUSLE for the pixels corresponding to the experimental plot. This does not mean that the two approaches are identical, but it does indicate that they are in the same order of magnitude and that the model reproduced the local erosion signal reasonably well. This agreement is particularly meaningful because RUSLE is one of the most widely used empirical models for estimating soil erosion [22], yet its performance depends strongly on the quality of the input layers used to derive the R, K, LS, C, and P factors [33]. Reliable RUSLE predictions require well-prepared input data and, whenever possible, independent validation against field observation [24,35]. In the present study, the quality of the local input data was high, and this likely explains why the model outputs were consistent with the measured plot-scale values.
This also helps justify the decision to apply RUSLE to a small, well-defined sub-catchment rather than to the entire Xerias watershed. RUSLE was originally developed for estimating long-term average annual soil loss [33] and is generally more defensible when applied at plot to small-catchment scales [61], especially when local information is available for parameterization and some form of field validation is possible. When the model is transferred to larger areas without validation, uncertainty increases because the input factors become more generalized and because deposition processes are not explicitly represented [19]. In that respect, our study contributes by showing that a sub-catchment-scale application linked to direct field measurements can produce realistic erosion estimates under Mediterranean olive-growing conditions. The contribution is therefore not only the map itself, but also the demonstration that local validation improves confidence in the mapped results.
The LiDAR and DoD analysis added a different but complementary perspective. Unlike runoff plots, which measure exported sediment, the DoD approach mainly reveals surface elevation change [44], allowing the identification of zones where the soil surface was lowered or raised between surveys. For this reason, in the present work the DoD results are interpreted primarily in terms of soil displacement and local surface redistribution, rather than as a direct estimate of soil loss. This interpretation is consistent with other DoD-based studies, which show that differenced DEMs are particularly effective for mapping where topographic change occurs and for identifying short-distance redistribution patterns across small plots and hillslopes [62]. In our plot, the DoD pattern supports the general expectation that material tends to move from relatively higher to relatively lower positions, indicating downslope redistribution of soil within the monitored surface.
At the same time, the DoD part of the study should be interpreted as an exploratory but promising component of the overall methodology. The monitored area was relatively small (100 m2), and the comparison covered only a five-month period, which limits the degree to which broader or longer-term erosion patterns can be inferred. In addition, the usefulness of DoD products depends strongly on the quality of the DEMs and on their spatial resolution, because high-resolution surfaces are needed to preserve microtopographic features that control flow concentration and local erosion. Previous UAV- and LiDAR-based studies have shown that DEM resolution has a strong influence on slope representation, LS estimation, and the visibility of microtopographic change [63,64]. For that reason, the DoD results in this study are best viewed as a detailed indicator of surface behaviour at plot scale, rather than as a stand-alone measure of annual soil loss. Even so, they are valuable because they reveal the internal spatial organization of terrain change in a way that neither runoff plots nor RUSLE can show on their own.
Taken together, the three approaches show that soil erosion assessment benefits from a holistic framework. The field experiment supplies direct measurements, RUSLE provides spatial extrapolation and identifies erosion-prone areas, and LiDAR/DoD reveals fine-scale redistribution of the soil surface. This combination responds to a common limitation in erosion research, namely that many studies rely on only one method and therefore capture only one dimension of the process. Recent Mediterranean reviews explicitly call for this kind of multi-method, multi-scale strategy, because each method provides only a partial but useful view of erosion dynamics [5,65]. Our work contributes to this discussion by showing that, even in a relatively small Mediterranean sub-catchment, an integrated design can link direct observations with spatial modelling and high-resolution terrain analysis. Future work can build on this framework by extending monitoring duration, enlarging the number of plots, and further refining terrain-based analysis, with the aim of reducing remaining limitations and strengthening comparisons among methods.

5. Conclusions

This study showed that soil erosion assessment in Mediterranean olive-growing areas can be improved through the combined use of direct field measurements, RUSLE modelling, and LiDAR-based terrain analysis. The field experiment provided direct evidence of actual soil loss at the subplot scale, while the RUSLE model offered a spatially distributed estimate of erosion risk across the study sub-catchment. The agreement between the measured annual soil loss values and the RUSLE-predicted range at the experimental plot suggests that, when supported by high-quality local input data, RUSLE can provide realistic estimates of soil erosion under local conditions.
The LiDAR–DoD approach contributed an additional and complementary perspective by identifying short-term changes in surface elevation and patterns of local soil redistribution within the experimental plot. Although this method was not used here as a direct measure of soil loss, it proved useful for describing soil displacement and surface behavior over time. Taken together, the three approaches showed that soil erosion is a complex process that cannot be fully captured by a single method and that a multi-method framework can support a more complete understanding of erosion dynamics.
Overall, the findings highlight the importance of integrated erosion assessment not only for estimating soil loss, but also for improving our understanding of land degradation processes in vulnerable Mediterranean agroecosystems. Since soil erosion is one of the main drivers of land degradation, especially in sloping cultivated areas, the combined methodology applied in this study may support more targeted soil conservation and sustainable land management strategies. Future research should further refine these approaches, reduce remaining limitations, and expand monitoring in order to strengthen erosion assessment at both plot and sub-catchment scales.

Author Contributions

Conceptualization, C.P. and P.T.N.; methodology, C.P. and P.T.N.; software, C.P. and P.T.N.; validation, C.P. and P.T.N.; formal analysis, C.P. and P.T.N.; investigation, C.P. and P.T.N.; resources, C.P. and P.T.N.; data curation, C.P. and P.T.N.; writing—original draft preparation, C.P.; writing—review and editing, P.T.N.; visualization, C.P. and P.T.N.; supervision, P.T.N.; project administration, C.P. and P.T.N.; funding acquisition, P.T.N. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The original contributions presented in this study are included in the article. Further enquiries can be directed to the corresponding authors.

Acknowledgments

The authors would like to acknowledge the Navarino Environmental Observatory (NEO) for supporting this work through the provision of facilities and research infrastructure. We also thank TEMES for providing meteorological data from its monitoring network, which contributed to the implementation of this study.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
RUSLERevised Universal Soil Loss Equation
DEMDigital Elevation Model
DoDDEM of Difference
LiDARLight Detection and Ranging
GISGeographic Information System(s)
OMOrganic Matter
RTKReal-time kinematic positioning

References

  1. Li, L.; Du, S.; Wu, L.; Liu, G. An Overview of Soil Loss Tolerance. CATENA 2009, 78, 93–99. [Google Scholar] [CrossRef]
  2. Carr, T.W.; Balkovič, J.; Dodds, P.E.; Folberth, C.; Skalský, R. The Impact of Water Erosion on Global Maize and Wheat Productivity. Agric. Ecosyst. Environ. 2021, 322, 107655. [Google Scholar] [CrossRef]
  3. Andualem, T.G.; Hewa, G.A.; Myers, B.R.; Peters, S.; Boland, J. Erosion and Sediment Transport Modeling: A Systematic Review. Land 2023, 12, 1396. [Google Scholar] [CrossRef]
  4. Guerra, C.A.; Rosa, I.M.D.; Valentini, E.; Wolf, F.; Rosa, I.M.D.; Filipponi, F.; Karger, D.N.; Xuan, A.N.; Mathieu, J.; Lavelle, P.; et al. Global Vulnerability of Soil Ecosystems to Erosion. Landsc. Ecol. 2020, 35, 823–842. [Google Scholar] [CrossRef] [PubMed]
  5. García-Ruiz, J.M.; Nadal-Romero, E.; Lana-Renault, N.; Beguería, S. Erosion in Mediterranean Landscapes: Changes and Future Challenges. Geomorphology 2013, 198, 20–36. [Google Scholar] [CrossRef]
  6. Samela, C.; Imbrenda, V.; Coluzzi, R.; Pace, L.; Simoniello, T.; Lanfredi, M. Multi-Decadal Assessment of Soil Loss in a Mediterranean Region Characterized by Contrasting Local Climates. Land 2022, 11, 1010. [Google Scholar] [CrossRef]
  7. Peñuela, A.; González-Sánchez, E.J.; Milazzo, F. Soil Erosion in Mediterranean Olive Groves: A Review. Soil 2026, 12, 93–111. [Google Scholar] [CrossRef]
  8. Scholten, T.; Seitz, S. Soil Erosion and Land Degradation. Soil Syst. 2019, 3, 68. [Google Scholar] [CrossRef]
  9. Ferreira, C.S.S.; Seifollahi-Aghmiuni, S.; Destouni, G.; Ghajarnia, N.; Kalantari, Z. Soil Degradation in the European Mediterranean Region: Processes, Status and Consequences. Sci. Total Environ. 2022, 805, 150106. [Google Scholar] [CrossRef]
  10. Gomez, J.A.; Amato, M.; Celano, G.; Koubouris, G.C. Organic Olive Orchards on Sloping Land: More than a Specialty Niche Production System? J. Environ. Manag. 2008, 89, 99–109. [Google Scholar] [CrossRef]
  11. Deitch, M.J.; Sapundjieff, M.J.; Feirer, S.T. Characterizing Precipitation Variability and Trends in the World’s Mediterranean-Climate Areas. Water 2017, 9, 259. [Google Scholar] [CrossRef]
  12. Vaezi, A.R.; Ahmadi, M.; Cerdà, A. Contribution of Raindrop Impact to the Change of Soil Physical Properties and Water Erosion under Semi-Arid Rainfalls. Sci. Total Environ. 2017, 583, 382–392. [Google Scholar] [CrossRef]
  13. Márquez-García, F.; Hayas, A.; Peña, A.; Ordóñez-Fernández, R.; González-Sánchez, E.J. Influence of Cover Crops and Tillage on Organic Carbon Loss in Mediterranean Olive Orchards. Soil Tillage Res. 2024, 235, 105905. [Google Scholar] [CrossRef]
  14. Chontos, K.; Pantazis, C.; Berg, H. Impact of Soil Management Practices on Olive Orchard Soil Health and Arthropod Diversity in Messenia, Greece. Agronomy 2026, 16, 404. [Google Scholar] [CrossRef]
  15. González-Rosado, M.; Parras-Alcántara, L.; Aguilera-Huertas, J.; Lozano-García, B. Soil Productivity Degradation in a Long-Term Eroded Olive Orchard under Semiarid Mediterranean Conditions. Agronomy 2021, 11, 812. [Google Scholar] [CrossRef]
  16. Ahmad, N.S.B.N.; Mustafa, F.B.; Yusoff, S.Y.M.; Didams, G. A Systematic Review of Soil Erosion Control Practices on the Agricultural Land in Asia. Int. Soil Water Conserv. Res. 2020, 8, 103–115. [Google Scholar] [CrossRef]
  17. Parsons, A.J. How Reliable Are Our Methods for Estimating Soil Erosion by Water? Sci. Total Environ. 2019, 676, 215–221. [Google Scholar] [CrossRef]
  18. Boix-Fayos, C.; Martínez-Mena, M.; Arnau, E.; Calvo-Cases, A.; Castillo, V.; Albaladejo, J. Measuring Soil Erosion by Field Plots: Understanding the Sources of Variation. Earth-Sci. Rev. 2006, 78, 267–285. [Google Scholar] [CrossRef]
  19. Räsänen, T.A.; Tähtikarhu, M.; Uusi-Kämppä, J.; Piirainen, S.; Turtola, E. Evaluation of RUSLE and Spatial Assessment of Agricultural Soil Erosion in Finland. Geoderma Reg. 2023, 32, e00610. [Google Scholar] [CrossRef]
  20. Alexiou, S.; Papanikolaou, I.; Schneiderwind, S.; Kehrle, V.; Reicherter, K. Monitoring and Quantifying Soil Erosion and Sedimentation Rates in Centimeter Accuracy Using UAV-Photogrammetry, GNSS, and t-LiDAR in a Post-Fire Setting. Remote Sens. 2024, 16, 802. [Google Scholar] [CrossRef]
  21. Renard, K.G.; Foster, G.R.; Weesies, G.A.; Porter, J.P. RUSLE: Revised Universal Soil Loss Equation. J. Soil Water Conserv. 1991, 46, 30–33. [Google Scholar] [CrossRef]
  22. Benavidez, R.; Jackson, B.; Maxwell, D.; Norton, K. A Review of the (Revised) Universal Soil Loss Equation ((R)USLE): With a View to Increasing Its Global Applicability and Improving Soil Loss Estimates. Hydrol. Earth Syst. Sci. 2018, 22, 6059–6086. [Google Scholar] [CrossRef]
  23. Abdelsamie, E.A.; Abdellatif, M.A.; Hassan, F.O.; El Baroudy, A.A.; Mohamed, E.S.; Kucher, D.E.; Shokr, M.S. Integration of RUSLE Model, Remote Sensing and GIS Techniques for Assessing Soil Erosion Hazards in Arid Zones. Agriculture 2023, 13, 35. [Google Scholar] [CrossRef]
  24. Ganasri, B.P.; Ramesh, H. Assessment of Soil Erosion by RUSLE Model Using Remote Sensing and GIS—A Case Study of Nethravathi Basin. Geosci. Front. 2016, 7, 953–961. [Google Scholar] [CrossRef]
  25. Batista, P.V.G.; Davies, J.; Silva, M.L.N.; Quinton, J.N. On the Evaluation of Soil Erosion Models: Are We Doing Enough? Earth-Sci. Rev. 2019, 197, 102898. [Google Scholar] [CrossRef]
  26. Ta, N.; Wang, C.; Zhao, S.; Zhang, Q. Quantifying Soil Erosion Processes Based on Micro-ΔDEM. Water 2025, 17, 2557. [Google Scholar] [CrossRef]
  27. Ministry of Agriculture Greek Payment Authority of Common Agricultural Policy Aid Schemes. Available online: https://www.opekepe.gr/opekepe-organisation-gr/opekepe-e-services-gr/pliroforiaka-systimata/gewpliroforiako-systima-edafologikwn-dedomenwn (accessed on 24 March 2026).
  28. Walkley, A.; Black, I.A. An Examination of the Degtjareff Method for Determining Soil Organic Matter and a Proposed Modification of the Chromic Acid Titration Method. Soil Sci. 1934, 37, 29–38. [Google Scholar] [CrossRef]
  29. Bouyoucos, G.J. Hydrometer Method Improved for Making Particle Size Analyses of Soils. Agron. J. 1962, 54, 464–465. [Google Scholar] [CrossRef]
  30. USDA. Soil Survey Manual; United States Department of Agriculture: Washington, DC, USA, 1951.
  31. Kakkavou, K.; Gemtou, M.; Fountas, S. Drivers and Barriers to the Adoption of Precision Irrigation Technologies in Olive and Cotton Farming—Lessons from Messenia and Thessaly Regions in Greece. Smart Agric. Technol. 2024, 7, 100401. [Google Scholar] [CrossRef]
  32. Thanopoulos, R.; Drossinou, I.; Koutroumpelas, I.; Chatzigeorgiou, T.; Stavrakaki, M.; Bebeli, P.J. Hilly, Semi-Mountainous and Mountainous Areas Harbor Landraces Diversity: The Case of Messinia (Peloponnese-Greece). Diversity 2024, 16, 151. [Google Scholar] [CrossRef]
  33. Wischmeier, W.H.; Smith, D.D. Predicting Rainfall Erosion Losses, a Guide to Conservation Planning; U.S. Department of Agriculture: Washington, DC, USA, 1978; 62p.
  34. Nicosia, A.; Carollo, F.G.; Di Stefano, C.; Palmeri, V.; Pampalone, V.; Serio, M.A.; Bagarello, V.; Ferro, V. The Importance of Measuring Soil Erosion by Water at the Field Scale: A Review. Water 2024, 16, 3427. [Google Scholar] [CrossRef]
  35. Renard, K.G.; Foster, G.R.; Weesies, G.A.; McCool, D.K.; Yoder, D.C. Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation; USDA Agriculture: Washington, DC, USA, 1997.
  36. Renard, K.G.; Freimund, J.R. Using Monthly Precipitation Data to Estimate the R-Factor in the Revised USLE. J. Hydrol. 1994, 157, 287–306. [Google Scholar] [CrossRef]
  37. Arnoldus, H.M.J. An Approximation of the Rainfall Factor in the Universal Soil Loss Equation; John Wiley and Sons: Chichester, UK, 1980. [Google Scholar]
  38. Fournier, F. Climat et Érosion; Presses Universitaires: Paris, France, 1960. [Google Scholar]
  39. Arnoldus, J.M.J. Methodology Used to Determine the Maximum Potential Average Annual Soil Loss Due to Sheet and Rill Erosion in Morocco. Food Agric. Org. Soils Bull. 1977, 34, 39–51. [Google Scholar]
  40. USDA. Soil Conservation Service; United States Department of Agriculture: Washington, DC, USA, 1983; p. 650.
  41. Moore, I.D.; Burch, G.J. Physical Basis of the Length-Slope Factor in the Universal Soil Loss Equation. Soil Sci. Soc. Am. J. 1986, 50, 1294–1298. [Google Scholar] [CrossRef]
  42. Moore, I.D.; Wilson, J.P. Length-Slope Factors for the Revised Universal Soil Loss Equation: Simplified Method of Estimation. J. Soils Water Conserv. 1992, 47, 423–428. [Google Scholar] [CrossRef]
  43. Panagos, P.; Borrelli, P.; Meusburger, K.; Alewell, C.; Lugato, E.; Montanarella, L. Estimating the Soil Erosion Cover-Management Factor at the European Scale. Land Use Policy 2015, 48, 38–50. [Google Scholar] [CrossRef]
  44. Anders, N.; Valente, J.; Masselink, R.; Keesstra, S. Comparing Filtering Techniques for Removing Vegetation from UAV-Based Photogrammetric Point Clouds. Drones 2019, 3, 61. [Google Scholar] [CrossRef]
  45. Storch, M.; Kisliuk, B.; Jarmer, T.; Waske, B.; de Lange, N. Comparative Analysis of UAV-Based LiDAR and Photogrammetric Systems for the Detection of Terrain Anomalies in a Historical Conflict Landscape. Sci. Remote Sens. 2025, 11, 100191. [Google Scholar] [CrossRef]
  46. Wheaton, J.M.; Brasington, J.; Darby, S.E.; Sear, D.A. Accounting for Uncertainty in DEMs from Repeat Topographic Surveys: Improved Sediment Budgets. Earth Surf. Process. Landf. 2009, 35, 136–156. [Google Scholar] [CrossRef]
  47. Michail, I.; Pantazis, C.; Solomos, S.; Michailidis, M.; Molassiotis, A.; Gkisakis, V. Cover Crops for Carbon Mitigation and Biodiversity Enhancement: A Case Study of an Olive Grove in Messinia, Greece. Agriculture 2025, 15, 898. [Google Scholar] [CrossRef]
  48. Blake, G.R.; Hartge, K.H. Bulk Density. Methods of Soil Analysis, 2nd ed.; Soil Science Society of America (SSSA): Madison, WI, USA, 1986. [Google Scholar]
  49. Wen, X.; Deng, X. Current Soil Erosion Assessment in the Loess Plateau of China: A Mini-Review. J. Clean. Prod. 2020, 276, 123091. [Google Scholar] [CrossRef]
  50. Epple, L.; Kaiser, A.; Schindewolf, M.; Bienert, A.; Lenz, J.; Eltner, A. A Review on the Possibilities and Challenges of Today’s Soil and Soil Surface Assessment Techniques in the Context of Process-Based Soil Erosion Models. Remote Sens. 2022, 14, 2468. [Google Scholar] [CrossRef]
  51. Alexakis, D.D.; Hadjimitsis, D.G.; Agapiou, A. Integrated Use of Remote Sensing, GIS and Precipitation Data for the Assessment of Soil Erosion Rate in the Catchment Area of “Yialias” in Cyprus. Atmos. Res. 2013, 131, 108–124. [Google Scholar] [CrossRef]
  52. Pantazis, C.; Nastos, P. Comparing Direct Field Measurements of Soil Erosion with RUSLE Model Estimates in Mediterranean Olive Orchards. Environ. Earth Sci. Proc. 2025, 35, 75. [Google Scholar] [CrossRef]
  53. Maetens, W.; Poesen, J.; Vanmaercke, M. How Effective Are Soil Conservation Techniques in Reducing Plot Runoff and Soil Loss in Europe and the Mediterranean? Earth-Sci. Rev. 2012, 115, 21–36. [Google Scholar] [CrossRef]
  54. Thapa, P. Spatial Estimation of Soil Erosion Using RUSLE Modeling: A Case Study of Dolakha District, Nepal. Environ. Syst. Res. 2020, 9, 15. [Google Scholar] [CrossRef]
  55. Zaied, M.B.; Jomaa, S.; Ouessar, M. Soil Erosion Estimates in Arid Region: A Case Study of the Koutine Catchment, Southeastern Tunisia. Appl. Sci. 2021, 11, 6763. [Google Scholar] [CrossRef]
  56. Nastos, P.; Pantazis, C.; Solomos, S.; Fountoulakis, I.; Zerefos, C.S. Evaluating Soil Erosion Risk in Olive Orchards through Drone-Based Monitoring and Field Experiments. In Proceedings of the Eleventh International Conference on Remote Sensing and Geoinformation of the Environment (RSCy2025), Paphos, Cyprus, 17–19 March 2025; SPIE: Bellingham, WA, USA, 2025. [Google Scholar] [CrossRef]
  57. Critchley, W.; Siegert, K. A Manual for the Design and Construction of Water Harvesting Schemes for Plant Production; Food and Agriculture Organization of the United Nations: Rome, Italy, 1991. [Google Scholar]
  58. Kinnell, P.I.A. A Review of the Design and Operation of Runoff and Soil Loss Plots. CATENA 2016, 145, 257–265. [Google Scholar] [CrossRef]
  59. Anache, J.A.A.; Wendland, E.C.; Oliveira, P.T.S.; Flanagan, D.C.; Nearing, M.A. Runoff and Soil Erosion Plot-Scale Studies under Natural Rainfall: A Meta-Analysis of the Brazilian Experience. CATENA 2017, 152, 29–39. [Google Scholar] [CrossRef]
  60. Rüttimann, M.; Schaub, D.; Prasuhn, V.; Rüegg, W. Measurement of Runoff and Soil Erosion on Regularly Cultivated Fields in Switzerland—Some Critical Considerations. CATENA 1995, 25, 127–139. [Google Scholar] [CrossRef]
  61. Terefe, B.; Melese, T.; Tsegaye, A.; Afework, A.; Yibeltal, T.; Anagaw, A.; Temesgen, F.; Belay, T.; Assefa, G.; Mencho, B.B. Soil Loss Estimation in Ethiopia: A Comprehensive Review of the RUSLE Model Integrated with Geospatial Technologies. Discov. Sustain. 2025, 6, 334. [Google Scholar] [CrossRef]
  62. Williams, R.D. DEMs of Difference. In Geomorphological Techniques; British Society for Geomorphology: London, UK, 2012. [Google Scholar]
  63. Chidi, C.L.; Zhao, W.; Chaudhary, S.; Xiong, D.; Wu, Y. Sensitivity Assessment of Spatial Resolution Difference in DEM for Soil Erosion Estimation Based on UAV Observations: An Experiment on Agriculture Terraces in the Middle Hill of Nepal. ISPRS Int. J. Geo-Inf. 2021, 10, 28. [Google Scholar] [CrossRef]
  64. Dai, W.; Hu, J.; Wang, B.; Fan, M.; Zhou, Y.; Zhang, M. Effects of Digital Elevation Model Resolution on Unmanned Aerial Vehicle-Based Topographic Change Detection in Human-Altered Landscapes. Drones 2024, 8, 610. [Google Scholar] [CrossRef]
  65. Wang, J.; Yang, J.; Li, Z.; Ke, L.; Li, Q.; Fan, J.; Wang, X. Research on Soil Erosion Based on Remote Sensing Technology: A Review. Agriculture 2025, 15, 18. [Google Scholar] [CrossRef]
Land 15 00794 g001
Figure 1. Location of the study area in a sub-catchment of the Xerias River watershed, Pylos–Nestor, Messenia, Greece.
Figure 1. Location of the study area in a sub-catchment of the Xerias River watershed, Pylos–Nestor, Messenia, Greece.
Land 15 00794 g001
Land 15 00794 g002
Figure 2. Experimental set up of the soil erosion monitoring network.
Figure 2. Experimental set up of the soil erosion monitoring network.
Land 15 00794 g002
Land 15 00794 g003
Figure 3. Workflow of RUSLE model.
Figure 3. Workflow of RUSLE model.
Land 15 00794 g003
Land 15 00794 g004
Figure 4. TEMES weather station network.
Figure 4. TEMES weather station network.
Land 15 00794 g004
Land 15 00794 g005
Figure 5. Temporal variation in event rainfall characteristics (a) and sediment yields in the 100 m2 experimental subplot (b) during the monitoring period (2023–2025).
Figure 5. Temporal variation in event rainfall characteristics (a) and sediment yields in the 100 m2 experimental subplot (b) during the monitoring period (2023–2025).
Land 15 00794 g005
Land 15 00794 g006aLand 15 00794 g006b
Figure 6. Spatial variability of the RUSLE input factors in the study sub-catchment, including rainfall erosivity (a), soil erodibility (b), elevation (c), topographic factor (d), land cover factor (e), and support practice factor (f).
Figure 6. Spatial variability of the RUSLE input factors in the study sub-catchment, including rainfall erosivity (a), soil erodibility (b), elevation (c), topographic factor (d), land cover factor (e), and support practice factor (f).
Land 15 00794 g006aLand 15 00794 g006b
Land 15 00794 g007
Figure 7. Spatial distribution of predicted annual soil loss in the study sub-catchment based on the RUSLE model.
Figure 7. Spatial distribution of predicted annual soil loss in the study sub-catchment based on the RUSLE model.
Land 15 00794 g007
Land 15 00794 g008
Figure 8. Lidar-derived surface elevation before and after vegetation filtering. (a) Raw point-cloud DEM (tree tops and ground vegetation still present). (b) Filtered ground DEM (vegetation removed).
Figure 8. Lidar-derived surface elevation before and after vegetation filtering. (a) Raw point-cloud DEM (tree tops and ground vegetation still present). (b) Filtered ground DEM (vegetation removed).
Land 15 00794 g008
Land 15 00794 g009
Figure 9. Bare-earth digital elevation models (DEMs) of the experimental plot for 11 December 2024 (a) and 6 May 2025 (b).
Figure 9. Bare-earth digital elevation models (DEMs) of the experimental plot for 11 December 2024 (a) and 6 May 2025 (b).
Land 15 00794 g009
Land 15 00794 g010
Figure 10. DEM of Difference (DoD) of the experimental plot showing surface elevation changes between 11 December 2024 and 6 May 2025.
Figure 10. DEM of Difference (DoD) of the experimental plot showing surface elevation changes between 11 December 2024 and 6 May 2025.
Land 15 00794 g010
Table 1. C-factor per cover type.
Table 1. C-factor per cover type.
Land CoverC-Factor
Olive groves0.2273
Sclerophyllous vegetation0.0623
Land principally occupied by agriculture with significant areas of natural vegetation0.1232
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Pantazis, C.; Nastos, P.T. Land Degradation Assessment in an Olive Orchard Using Different Soil Erosion Estimation Methods. Land 2026, 15, 794. https://doi.org/10.3390/land15050794

AMA Style

Pantazis C, Nastos PT. Land Degradation Assessment in an Olive Orchard Using Different Soil Erosion Estimation Methods. Land. 2026; 15(5):794. https://doi.org/10.3390/land15050794

Chicago/Turabian Style

Pantazis, Christos, and Panagiotis T. Nastos. 2026. "Land Degradation Assessment in an Olive Orchard Using Different Soil Erosion Estimation Methods" Land 15, no. 5: 794. https://doi.org/10.3390/land15050794

APA Style

Pantazis, C., & Nastos, P. T. (2026). Land Degradation Assessment in an Olive Orchard Using Different Soil Erosion Estimation Methods. Land, 15(5), 794. https://doi.org/10.3390/land15050794

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop