A New Approach For Smart Soil Erosion Modelling: Integration of Empirical And Machine Learning Models

Despite advances in artificial intelligence modelling, the lack of soil erosion data and other watershed information is still one of the important factors limiting soil-erosion modelling. Additionally, the limited number of parameters and the lack of evaluation criteria are major disadvantages of empirical soil-erosion models. To overcome these limitations, we introduce a new approach that integrates empirical and artificial intelligence models. Erosion-prone locations (erosion ≥16 tons/ha/year) are identified using RUSLE model and a soil-erosion map is prepared using random forest (RF), artificial neural network (ANN), classification tree analysis (CTA), and generalized linear model (GLM). This study uses 13 factors affecting soil erosion in the Talar watershed, Iran, to increase prediction accuracy. The results reveal that the RF model has the highest prediction performance (AUC=0.95, Kappa=0.87, Accuracy=0.93, and Bias=0.88), outperforming the three machine-learning models. The results show that slope angle, land use/land cover, elevation, and rainfall erosivity are the factors that contribute the most to soil erosion propensity in the watershed. Curvature and topography position index (TPI) were removed from the analysis due to multicollinearity with other factors. The results can be used to improve the identification of hot spots of soil erosion, especially in watersheds for which soil-erosion data are limited.

Can the models keep up with the data? Possibilities of soil and soil surface assessment techniques in the context of process based soil erosion models – A Review

Climate change, accompanied by intensified extreme weather events, results in changes in intensity, frequency and magnitude of soil erosion. These unclear future developments make adaption and improvement of soil erosion modelling approaches all the more important. Hypothesizing that models cannot keep up with the data, this review gives an overview of 44 process based soil erosion models, their strengths and weaknesses and discusses their potential for further development with respect to new and improved soil and soil erosion assessment techniques. We found valuable tools in areas, as remote sensing, tracing or machine learning, to gain temporal and spatial distributed high resolution parameterization and process descriptions which could lead to a more holistic modelling approach. Most process based models are so far not capable to implement cross-scale erosional processes or profit from the available resolution on a temporal and spatial scale. We conclude that models need further development regarding their process understanding, adaptability in respect to scale as well as their parameterization and calibration. The challenge is the development of models which are able to simulate soil erosion processes as close to reality as possible, as user-friendly as possible and as complex as it needs to be.

The local implications of a macrostrategy on the villagers and the environment

Nature subcontracted mankind for temporary management stipulating a rigor that does not disregard the smallest details because this is where the difference between perfection and riskiness lies, which inevitably leads to disasters. It establishes binding general rules and local imperatives to be fulfilled otherwise it takes back the reins. For this reason, anthropic macrostrategies must frame their priorities and objectives according to a sustainable social-ecosystem. Here, we evaluate the environmental effect of the triggered landuse change of the metropolis-portuary-industrial park of Tangier (Strait of Gibraltar) on ecosystem services (i.e. vegetation and water) and anticipate their interactions with indigenous villagers. We established a multifactorial analysis including long-term (1985-2021) land-use dynamic assessment, 16-years pixel-based Mann-Kendall phenological trend, EPM soil erosion modelling and assessment of the total volumes of the detached soil, and an NDVI/NDWI monthly drought monitoring. Later, we compared these outcomes with a social survey with 171 households to analyze their living conditions, and their environmental perception and attitude. We have observed that anthropogenic intervention is the precursor of erosion which is likely to worsen natural weaknesses, which are already at the origin of a massive potential loss of soil estimated at 1.2 kg/m²/year. We correlated the negative phenological trend with the lane of the roads and infrastructure and we observed that severe drought episodes are long, frequent and at short intra and interannual intervals. We found a statistically significant association between the low level of education and the rural seclusion with the environmental degradation and the unavailability of water which can evoke serious risks. We anticipate the urgent need for an inclusive reform that implements a behavioral culture, encourages education, and creates social facilitation to build an upward spiral that produces better conditions and more opportunities for the rural society. Broadly, Managers should integrate scientific instructions to master the details and continuously improve macrostrategies to achieve integral and lasting success.

Soil and water conservation in terraced and non-terraced cultivations - a massive comparison of 50 vineyards

Understanding the soil and water conservation (SWC) impact of steep-slope agricultural practices (e.g. terraces) has arguably never been more relevant than today, in the face of widespread intensifying rainfall conditions. In northern Italy, a diverse mosaic of terraced and non-terraced cultivation systems have historically developed from local traditions and more recently from the introduction of machinery. Previous studies suggested that each vineyard configuration is characterised by a specific set of soil degradation patterns. However, an extensive analysis of SWC impacts by different vineyard configurations is missing, while this is crucial for providing robust guidelines for future-proof viticulture. Here, we provide a unique extensive comparison of SWC in 50 vineyards, consisting of 10 sites of 5 distinct practices: slope-wise cultivation (SC), contour cultivation (CC), contour terracing (CT), broad-base terracing (BT) and diagonal terracing (DT). A big-data analysis of physical erosion modelling based on high-resolution LiDAR data is performed, while four predefined SWC indicators are systematically analysed and statistically quantified. Regular contour terracing (CT) ranked best across all indicators, reflecting a good combination of flow interception and homogeneous distribution of runoff and sediment under intense rainfall conditions. The least SWC-effective practices (SC, CC, and DT) were related to vineyards optimised for trafficability by access roads or uninterrupted inter-row paths, which create high upstream-downstream connectivity and are thus prone to flow accumulation. The novel large-scale approach of this study offers a robust comparison of SWC impacts under intense rainstorms, which is becoming increasingly relevant for sustainable future management of such landscapes.

Maximum Entropy Distribution of Rainfall Intensity and Duration – MEDRID: a method for precipitation temporal downscaling for sediment delivery assessment

Scarcity of precipitation data is yet a problem in erosion modelling, especially when working in remote and data scarce areas. While much effort was made to use remote sensing and reanalysis data, they are still considered to be not completely reliable, notably for sub-daily measures such as duration and intensity. A way forward are statistical analyses, which can help modellers to obtain sub-daily precipitation characteristics by using daily totals. In this paper, we propose a novel method (Maximum Entropy Distribution of Rainfall Intensity and Duration – MEDRID) to assess the duration and intensity of sub-daily rainfalls relevant for modelling of sediment delivery ratios. We use the generated data to improve the sediment yield assessment in seven catchments with area varying from 10−3 to 10+2 km2 and broad timespan of monitoring (1 to 81 years). The best probability density function derived from MEDRID to reproduce sub-daily duration is the generalised gamma distribution (NSE = 0.98), whereas for the rain intensity it is the uniform (NSE = 0.87). The MEDRID method coupled with the SYPoME model (Sediment Yield using the Principle of Maximum Entropy) represents a significant improvement over empirically-based SDR models, given its average absolute error of 21 % and a Nash-Sutcliffe Efficiency of 0.96 (rather than 105 % and −4.49, respectively

Machine Learning Aided Prediction of Rain Erosion Damage on Wind Turbine Blade Sections

Rain erosion of wind turbine blades represents an interesting topic of study due to its non-negligible impact on annual energy production of the wind farms installed in rainy sites. A considerable amount of recent research works has been oriented to this subject, proposing rain erosion modelling, performance losses prediction, structural issues studies, etc. This work aims to present a new method to predict the damage on a wind turbine blade. The method is applied here to study the effect of different rain conditions and blade coating materials, on the damage produced by the rain over a representative section of a reference 5MW turbine blade operating in normal turbulence wind conditions.

Risk and Resources: An Evaluation of the Ability of National Soil Datasets to Predict Post-Depositional Processes in Archaeological Sites and Heritage at Risk

Previous studies have demonstrated the vast range of physical, chemical and biological processes that influence the preservation of archaeological sites, yet characterisation at the site-level remains largely unexplored. National datasets on soil type, land use and erosion modelling have the potential to predict localised impacts but remain an untapped resource in the evaluation of heritage at risk. Using early medieval Scotland as a case study, this paper explores in detail some of the primary factors which have impacted the archaeological record and the degree to which site-based evidence contained in excavation reports compares with national datasets (Land Cover Map 2015, Soil Information for Scottish Soils and Soils of Scotland Topsoil pH) and coastal erosion models (Dynamic Coast National Coastal Change Assessment and Coastal Erosion Susceptibility Model). This provides valuable information on the preservation of Scotland’s early medieval settlement, as well as a methodology for using national datasets in the remote assessment of post-depositional factors across the broader archaeological landscape. Results indicate that agriculture, bioturbation and aggressive soil conditions are among the most significant factors impacting Scotland’s archaeological remains. While the national datasets examined have the potential to inform heritage management strategies on these processes, their use is limited by a number of theoretical and methodological issues. Moving forward, site-specific studies that characterise the preservation environment will be crucial in developing baseline assessments that will advance both local and global understandings of destructive factors and soil-mediated decay.