Soil phosphorus (P) is an essential nutrient that controls wetland productivity and ecological functions. However, the effects of soil P forms on P availability during wetland degradation are relatively unknown. Soil samples from differently degraded marshes, including relatively pristine marsh (RPM), lightly degraded marsh (LDM), moderately degraded marsh (MDM), and heavily degraded marsh (HDM), were collected to investigate the changes in soil P forms and its regulation on P availability in the Zoige Plateau, China. We observed that compared with RPM, the main changes in total P concentration were a significant increase of 31.6%–44.2% in the 0–30 cm soil layers of LDM and MDM, and the available P concentration increased in LDM and MDM but decreased in HDM with a lower P activation coefficient. Marsh degradation increased the concentration and proportion of dicalcium phosphates, P occluded in iron hydroxides, and organic P but decreased those of iron oxide surfaces adsorbed P and apatite P. Soil available P was mainly related to organic P and P non-occluded in iron oxide minerals that might also be non-negligible direct source of available P. The transformation from apatite P to organic P was an important regulation mechanism of P availability in soils during marsh degradation. This study revealed the risk of P limitation in heavily degraded marsh soils and established the mechanism by which marsh degradation significantly influences soil P availability. Therefore, some measure of on improving P availability should be implement for the ecological restoration of heavily degraded marsh in the future, such as grazing exclusion and the application of organic fertiliser.

Snow avalanches are a major component of the mountain cryosphere that frequently create a menace for the road network. Deposit characteristics determine the extent of damage and disruption to communication networks, but the factors controlling snow-deposit volumes remain largely unknown. This study investigates the influence of meteorological and snowpack conditions on snow-avalanche deposits and road-network vulnerability based on 1986 deposit volumes from 182 paths located in two regions of the French Alps between 2003 and 2017: Guil and Haute-Maurienne valleys. During the period, 195 avalanches impacted the road network in these areas, leading to major disruptions. In Haute-Maurienne, correlations between deposit volumes and meteorological and snowpack conditions are high in winter. However, the relationships differ with path elevation and orientation. Results do not show any significant relationship between volumes and meteorological or snowpack conditions for the spring season. Focusing on deposits that disturbed the road network, winter and spring reveal a distinct influence of meteorological and snow variables compared to the overall data set, with snowfall intensity as the predominant control variable of deposit volumes leading to road cuts. When the same analysis is conducted by considering Guil valley separately or by aggregating Haute-Maurienne with Guil valley area data, results do not show any significant relationship, highlighting the specific regional nature of relations between deposit volumes and meteorological and snowpack conditions.

One of the core challenges to achieve Land Degradation Neutrality (LDN) is to spatially identify, and strategically prioritise, the areas to implement actions to avoid, reduce and reverse land degradation. To achieve this, a tool for a participatory and data-driven assessment considering both the biophysical, and socio-economic dimensions of land degradation across scales was developed for Ecuador. In this paper, we present the methodology and results obtained, including the spatially explicit interactive tool developed to integrate indicators that support the scaling-up of sustainable land management (SLM). The process involved specialists from various national and international institutions, as well as decision makers from the public sector and other relevant stakeholders. Cloud computing allowed the integration of five main sources of data: (1) the results of a participatory land degradation assessment based on an expert knowledge questionnaire following the Land Degradation Assessment in Drylands (LADA) and World Overview of Conservation Approaches and Technologies (WOCAT) methodology; (2) the Hand-in-Hand Initiative Ecuador typology maps based on poverty maps and estimated agricultural potential and efficiency scores using household surveys on agricultural production; (3) National data sets on land cover and land use, soil properties, and hydro climatic indicators; (4) global satellite-derived LDN indicators, such as Land Productivity Dynamics; and (5) Documented SLM practices from WOCAT Global SLM Database. The tool is based on a Google Earth Engine application and allows decision makers to easily compare results and obtain statistics at different spatial scales and landscapes, including 647 Land Use Systems delimited by experts. It also includes a multi-criteria module to identify areas with specific characteristics to prioritise different types of interventions to achieve the country´s LDN targets. Convergence of local and global evidence allowed the identification of hotspots of degradation as well as areas of false positives/negatives - if only global or remote sensing indicators were considered. The participatory process contributed to strengthening multi-sector cooperation mechanisms and to guaranteeing ownership of the tool and the results. The system will support Ecuador’s efforts to monitor and report progress towards LDN to the United Nations Convention to Combat Desertification. The system´s code is shared as a repository at Earth Engine and can be adapted to and used by other countries and regions.

科学合理布设耕地质量（CLQ）监测点，可以及时准确地提供耕地质量现状和变化信息，对保障国家粮食安全具有重要意义。传统的 CLQ 监测点选择方法是基于土地利用斑块的 CLQ。由于大补丁可能有不同等级，被选为监控点会降低监控CLQ的可靠性。此外，传统的监测点部署方法主要只考虑CLQ，而忽略了道路可达性和地形等因素，导致部分监测点无法进入。因此，为了提高 CLQ 监测的可靠性，本研究提出了一种部署 CLQ 监测点的新方法。第一的，像素尺度的 CLQ 是使用遗传算法 - 反向传播神经网络 (GA-BPNN) 模型基于具有 30 m 空间分辨率的 Landstat8 数据估计的。其次，采用分层抽样模型确定最佳样本点。最后，应用改进的空间模拟退火算法（ISSA），同时考虑坡度和道路可达性，以优化监测点的位置。本研究在中国广东省广州市从化区进行。结果突出表明 (1) 与测量的 CLQ 的准确度相比，准确度 (R 应用改进的空间模拟退火算法（ISSA），同时考虑坡度和道路可达性，以优化监测点的位置。本研究在中国广东省广州市从化区进行。结果突出表明 (1) 与测量的 CLQ 的准确度相比，准确度 (R 应用改进的空间模拟退火算法（ISSA），同时考虑坡度和道路可达性，以优化监测点的位置。本研究在中国广东省广州市从化区进行。结果突出表明 (1) 与测量的 CLQ 的准确度相比，准确度 (R2  =   0.63, RMSE = 79.32, NRMSE = 13.77%) 用遥感技术估计的 CLQ 是可靠的，不同等级的像素级 CLQ 数据比斑块级 CLQ 数据更合理。(2)基于分层抽样模型，最终在研究区确定了132个监测点。（3）与空间模拟退火算法（SSA）和标准网格法相比，本研究提出的方法总分更高（F=94.61）。此外，获得的样本点主要位于道路和平坦地形附近。这样可以有效避开人迹罕至的地方。因此，基于本研究提出的新方法的结果为获得最佳CLQ监测点提供了科学依据和技术支持。

The change in the structure of the polar landscape since the termination of the Little Ice Age (ca. 1900) is expressed by widespread retreat of glaciers, progressive exposure of glacial landforms at ice margins and opening ice marginal zones to increasing paraglacial and periglacial processes operating synchronously in adjacent areas. This study provides insights into the rate of post-LIA deglaciation and associated paraglacial transformation in foreland of Gåsbreen, a glacier situated in north-western Sørkapp Land, region characterized by one of fastest deglaciation rates in the entire Svalbard Archipelago. During the investigated period, 1938-2020, Gåsbreen was in a recession that accelerated after 1990 and as a result the area of its marginal zone almost tripled from 2.2 km² to 5.8 km². This process had a significant impact on the development of the relief in glacier foreland. The dynamics of landscape transformation in the glacier marginal zone manifested in degradation in the surface of ice-cored moraines and the forms that are underlain by dead-ice. Mass movements and debris flow on ice-cored moraines and fluvioglacial processes had a great influence on this transformation. Larger volume of proglacial waters intensified denudation, transport and accumulation of sediments, which resulted, in: an increase in the surface of sandurs and proglacial riverbeds, an increase in the area of lakes, extending and changing of the course of rivers in marginal zone.

Aeolian desertification is a severe ecological and environmental problem in arid regions. Research on its temporal and spatial distribution, development model, and driving force is necessary to prevent the development of aeolian desertification. In this study, the Moltsog dune field in Mongolia and the Ujimqin dune field in China were selected as the study areas, as both contain dunes with similar natural conditions. Using Landsat data from 1988, 1995, 2002, 2009, 2016, and 2020, the spatial-temporal distribution and degree of development of aeolian desertification in the two dune fields over the past 30 years were compared. Two periods of high-resolution images were then used to compare the surface morphological changes induced by aeolian desertification in the dune fields. Climatic and socio-economic data of the same period were used to compare and analyze the causes of changes in aeolian desertification in these regions. The results show that: (1) Over 30 years, the degree and development rate of aeolian desertification in the Ujimqin dune field were generally higher than those in the Moltsog dune field, and the former had a high degree of fragmented aeolian desertification patches with an expanding range. (2) The main form of aeolian desertification is the reactivation of fixed dunes, which includes the development of blowouts on the flat grassland under the influence of human activities in the Ujimqin dune field. (3) The desertification in Moltsog is mainly affected by climatic factors, while that in Ujimqin is mainly affected by human activities. The latter is specifically affected by the high grazing intensity before 2000 and increased mining activities after 2000. These findings provide a reference for comparing the aeolian desertification process and meaningful information for preventing and managing aeolian desertification and enabling the sustainable development of dune fields in arid regions.

Planted vegetation is considered one of the most important soil-forming factors in the reclamation of degraded post-mining lands for forestry. The objective of this study was to compare the effect of N-fixing tree species and non-N-fixing species on the organic phosphorus (P org) forms in technosols developing from various substrates. Samples were taken from the uppermost layer (depth 0 – 5 cm) of technosols afforested with black locust ( Robinia pseudoaccaccia), black alder ( Alnus glutinosa), silver birch ( Betula pendula) and Scots pine ( Pinus sylvestris). Samples of the tree foliage and the O horizons were taken as well. The studied technosols developed from Quaternary sands (Sands), fly ashes after lignite combustion (Ashes) and Miocene clays (Clays). The soil samples were measured for the contents of labile (P Olab) and moderately labile organic phosphorus (P Omod), phosphorus contained in fulvic and humic acids (P fulv and P hum, respectively) and residual organic phosphorus (P Ores). The foliage and O horizon samples were measured for the concentrations of C, N and P. The N-fixing trees had higher P concentration in the foliage than non-N-fixing species. However, in the O horizon the highest P concentration was determined under birch and not under N-fixers. The effect of tree species on the organic P (P org) concentrations in the mineral soil was limited with significantly lower P org concentrations under pine. Soil under this species contained less P fulv, P hum and P Ores. However, the percentages of P Olab, P Omod, P fulv, P hum and P Ores in soil P org were nearly the same under all tree species. The largest effect on the P org concentration and on the contents of particular P org fractions was from the substrate type. Sands contained much less P org than the Clays and Ashes but their percentage of P Olab in P org was much higher than in two other substrates. We conclude that N-fixing trees do not affect the concentration of labile organic P and the major factor controlling this P org fraction is the soil substrate quality.

： The problem of grassland degradation resulting from livestock production at the expense of local grasses has become a serious environmental problem worldwide. To maintain livestock production and sustainable grassland development, a better grasp of the extent and intensity of grassland utilization needs to be realized. We developed a model to monitor the intensity of grazing in a typical grazing area (Selinco basin) of the Qinghai-Tibet Plateau. Based on combining the number of livestock at the township scale with their matching livestock habitat area and location, the biomass consumed by livestock was then assigned to the productivity supply map (NPP) by a mathematical iterative algorithm from the perspective of different foraging habits of different livestock. The objective is to more accurately measure the overall grazing pressure on the pasture as well as to estimate livestock pasture utilization. The model has confirmed that there are obvious spatial differences in the intensity of grassland utilization in the Selinco basin, which is significantly lower except in the southwestern region. Overall grazing area derived from the statistics on the pixel is 150, thousands km 2. Among them, the area of moderate grazing area is 130, thousands km 2, accounting for 87% of the total grazing area, and the area of overgrazing area is 20, thousands km 2, accounting for 13% of the total grazing area. Our model can quantify human activities spatially and provide a reliable and accurate scientific basis for livestock production development and ecological environment management.

The effects of different straw returning and nitrogen addition levels on soil quality are important for proper coastal saline soil remediation. Two maize/wheat straw returning levels (1.0 × 10 4 kg ha -1 (2S) and 5.0 × 10 3 kg ha -1 (S)) and three inorganic nitrogen addition levels (300 kg ha -1 (N2), 150 kg ha -1 (N) and 75 kg ha -1 (N1/2))—were studied, with 150 kg ha -1 inorganic nitrogen and without straw addition treatment as the control (CK), to elucidate the response of soil physical and chemical properties to the two factors. Dry-sieving technique was applied to fractionate the soils into silt-plus-clay particles (< 0.053 mm, CS), microaggregates (0.053–0.25 mm, MI), small macroaggregates (0.25–2.0 mm, SM), and large macroaggregates (> 2 mm, LM). After four consecutive wheat-maize cycles, different straw and N fertilizer treatments obviously decreased the salinity contents, increased the total nutrient contents, and optimized the soil structure of the saline soil. The saline soil reclamation effects showed significant distinctions among the different straw and N fertilizer treatments. The 2SN2 treatment displayed the greatest effects in regard to decreasing salinity, increasing the total soil nutrient contents and optimizing the soil structure, which resulted in the best remediation effect. Straw returning play a major role in decreasing soil salinity and enhancing saline soil aggregate formation. N fertilizer addition supplies rich nutrients for straw decomposition, and promotes soil microbial growth and reproduction, which brought about C sequestration in coastal saline soil. During the coastal saline soil remediation process in the Yellow River Delta, it is suggested to prioritize straw returning and moderate N fertilizer addition, and live together with moderate P fertilizer application.

Agricultural Policy has been a major driver of Land Use Changes in Portugal, especially the south-eastern Alentejo: the commons division in the early 20 th century, autarkic policies from the fascist regime, and subsidies for forestry, cereal, and cattle production since joining the common market. This study focuses on the Serra de Mértola, an area with poor soils historically subject to land degradation. Data from the Vale Formoso Erosion Centre (1961-present) was used to compute erosion rates under different land uses representative of the study area. Higher values were found in vertical Fallow (964kg/ha/yr), Wheat (90kg/ha/yr) and horizontal Fallow (66kg/ha/yr), with lower values associated with Spontaneous Vegetation (3kg/ha/yr), Quercus (4kg/ha/yr), and Pines (7kg/ha/yr). Analysis of trends in Landsat NDVI values allowed to discriminate between land cover change and natural oscillations in weather and vegetation growth: in dry years NDVI values decrease for winter and spring, maintaining lower values throughout the year, while natural vegetation recovery shows a slow but steady increase until it stabilizes or breaks due to human action. Breaks in NDVI values were correlated with occasional ploughing in pastureland and spontaneous vegetation as well as the different moments in the agricultural year for cereal production. Three tendencies were observed: in Cultivated areas, ploughing provides higher sediment availability in valley floors, resulting in aggradation; Pastureland shows rock fragments on slopes and lower sediment availability, leading to incision on valley floors; Spontaneous Vegetation/Abandoned Land leads to a sharp decrease of sedimentary yield, with incision until its natural growth stabilizes hillslope processes.

The sulfur cycle is one of the geochemical element cycles in which microorganisms play a key driving role. The microbial function of soil S cycling in response to desert degradation, however, remains largely unknown. We used metagenomics to analyze the characteristics of microbial communities and their functional genes involved in the S cycles under natural water gradients with three typical halophytes shrubs in the Ebinur Lake Basin Desert, China. Our results showed that the rhizosphere effect, water gradient, and halophyte type played a major role in shaping the S cycle. On the whole, in the rhizosphere type and low water environment, the functional genes involved in the S cycle had high abundance, and the SOX system in Alhagi sparsifolia had a high expression level. In the S cycle network structure, as the soil water content decreased, the complexity in S gene networks increased, showing the characteristics of clustering and high connectivity. Indicates the strengthening mode in microbial interactions with the water content. Interestingly, the negative correlation of the network changed with the water content, and there was more competition among communities under the low water gradient and more cooperation under the high water gradient. Through the correlation between environmental factors and the network, nitrate (NO 3 −) and soil available S (AS) constrained most S gene ecology networks. The key species involved in the S cycle were halophilic microorganisms. These results can enhance the understanding of soil S biogeochemical processes and contribute to the mitigation of desertification by improving soil conservation.

Vegetation restoration in ecologically fragile areas has a significant carbon sequestration effect, and the driving factors affecting it are complex. In particular, it is usually difficult to achieve quantitative assessment at the regional scale for this part of human activity intervention in the ecological environment. The Chinese government’s ex-situ poverty alleviation and relocation project has relocated approximately 10 million people from areas with a fragile ecological environment to urban centralized resettlement, which is a typical case of weakened environmental intervention by human activities. Guizhou Province, an ecologically fragile region in southwest China, was selected as the study area, with a relocated population of 1.92 million, accounting for approximately 20% of the total relocated population in China. The carbon sequestration (CS) model of vegetation photosynthesis and spatial analysis of geographic information were used to quantify the contribution of human activities to the natural restoration of vegetation carbon sequestration at the regional scale caused by the weakening of environmental interventions, based on the data of net primary productivity (NPP) of vegetation from 2000 to 2020. The results show that the implementation of the ex-situ poverty alleviation and relocation project acts as an external force to drive vegetation restoration and carbon sequestration, which increases the slope of the carbon density change trend (from k=30.9 to k=57.41), resulting in an overall carbon density increase of 26.51 tCkm -2. The results of the regional spatial analysis showed that the correlation coefficients between carbon density and relocation intensity in the 5-year and 10-year change intervals before and after relocation were r=0.976 (p<0.01) and r=0.949 (p<0.05), respectively, indicating a significant positive correlation between vegetation carbon sequestration and relocation intensity. Based on this, the carbon sequestration contribution of vegetation in 84 counties (districts) in Guizhou Province that implemented ex-situ poverty alleviation and relocation projects was calculated, showing that 79 counties (districts) contributed positively, accounting for 94%. The average contribution of carbon sequestration by vegetation restoration in each county (district) was 0.0556 Tg, and the CO 2 emissions during the emission reduction period were 0.2059 Tg. The other five counties (districts) with a negative contribution to carbon sequestration were distributed in regions with relatively stable ecosystems and mature forests. This shows that human intervention in the environment changes significantly in fragile ecological areas.

This paper present results on soil contamination with petroleum hydrocarbons (PHs) on soil sorptivity and hydrophobicity under different soil moistures. The micro-infiltrometer method was used in laboratory experiment to determine the soil water repellency index (R) and the water drop penetration time (WDPT) test. The increase in PHs contamination contributed to soil repellency and caused a decrease in water sorptivity. The negative effect of contamination with PHs on soil sorptivity depended on soil moisture and was marked especially clearly after exceeding the critical moisture threshold. However, contamination by PHs did not reveal significant changes when ethanol was used instead of water. The R index and the WDPT test revealed a similar trend, inversely related to the level of soil contamination with PHs. The total amount of water available to plants in non-contaminated soil was 19.04%, while contamination equal to 100 g kg -1 caused a decrease to 6.36%. Hydrophobization of water-conducting pore surfaces by petroleum hydrocarbons severely reduced infiltration and destroyed the existing hydrological system of naturally hydrophilic soil. The almost three-fold decrease in total amount of water has a fundamental influence on increasing the risk of soil drought. The soil water repellency causes a decrease in resistance to droughts and slows the alimentation process of soil water retention. The results obtained indicated that the interrelations presented between the level of PHs contamination, soil sorptivity, water repellency, and soil moisture are key to predicting the environmental effects of contamination and effective soil remediation.

Precise assessment of soil organic carbon (SOC) storage requires understanding how vegetation and soil physicochemical properties differ in SOC fractions. Therefore, we aimed to analyze the dynamics of aggregate-associated, liable organic carbon (LOC) fractions corresponding to depth to clarify the effect of vegetation and soil properties on water stable aggregate (WSA) mineral adsorption in subtropical, red soil with five vegetation restoration regimes. The results showed that the large macro-aggregate fraction dominated the degraded red soil, which had the highest content of dissolved organic carbon (DOC). WSA-associated, easily oxidized organic carbon (EOC) varied from 6.26 to 20.02 g/kg and was not affected by vegetation types. Schima superba pure forest (SP) significantly increased DOC (0.38 g/kg on average) and particulate organic carbon (POC, 7.92g/kg on average), which had the highest biomass. Along with soil depth, WSA-associated POC declined, while exhibiting a growth trend with decreasing particle size, e.g., the highest POC was found in silt + clay fraction. The RDA ordination indicated that soil porosity and TN were the main soil parameters that explained the most variance. Meanwhile, the vegetation biomass, except for litter, were all significantly positively correlated with silt + clay fractions. Leaf biomass played the most important role on DOC in macro-aggregate with a 53.42% contribution. For aggregate-related POC, the largest contribution was from the interactions between branch biomass and pH (47.78%) followed by TN (35.1%) of micro-aggregate-related POC. Leaf biomass, silt + clay fractions, and TN can be used as indicators to evaluate the impact of vegetation restoration on WSA-associated SOC fractions. Broad leaved forest or combined with indigenous coniferous species was a better choice for SOC sequestration improvement in the study area.

Organic substitution management (OSM) is critical for soil conservation and nitrogen efficiency. However, assessment of soil N pool and system NUE under paddy-upland rotation systems after OSM introduction is still lacking, especially given the system’s alternating wetting-drying soil environment and double cropping. Here, the response of soil nitrogen and crop nitrogen use efficiency to OSM was evaluated in 2018, 2019 and 2020, basing on a long-term rice-fava bean rotation experiment started in 2015. Soil nitrogen, soil enzymes and microbial functions were also used to identify the key factors influencing nitrogen use in response to OSM. The results showed that in comparison to chemical fertilization, OSM increased the soil total nitrogen in both cropping seasons, with increases of 13.1% for the rice season and 16.3% for the fava bean season, respectively. However, nitrogen use efficiency (NUE) showed season-specific responses to OSM. OSM decreased NUE by 18.2% in the rice season but decreased by 8.0% in the fava bean season. Based on the SEM analysis results, we suggest that the opposite response of NRE and NUE to OSM across the studied seasons was mainly attributed to the difference in crop nitrogen requirements and soil hydrological conditions. In conclusion, in paddy-upland rotation areas where have increasing demand for food and chemical fertilizers are currently abused, the use of OSM has potential to curb regional soil degradation due to its high capacity of improving soil quality while ensuring no yield reduction.

Ecosystem carbon sequestration services (CSSs) are the most important ecosystem services (ESs) to mitigate global warming. Multiple soil and water conservation projects (SWCPs) have been implemented to restore disturbed ecosystems on the Loess Plateau, China. However, responses of CSSs to SWCPs are unclear due to trade-offs between CSSs and other ESs. Here, we quantified key ESs (i.e. carbon sequestration, water yield, soil conservation and crop production) and the spatio-temporal trade-off relationships by using RS/GIS techniques and ecosystem modeling in the Yanhe Basin, Loess Plateau, during 1990-2020. Additionally, the structural equation model (SEM) was used to estimate the direct and indirect inflences of multiple SWCPs including check dams, terraces and Grain for Green (GFG) on CSSs. Results show that CSSs has improved to 457 t/ha in 2020, which was twice compared to 1990. Here in, 57% of CSSs changes were explained by ESs and SWCPs. That is, water yield (-77%), soil conservation (76%), crop production (22%), GFG (80%), check dams (16%), and terraces (-72%), respectively. In order to balance trade-offs among ESs, GFG project with a focus on vegetation protection need to be prioritised, followed by check dams, and non-agricultural terraces such as [reverse-slope level terrace](http://search.ebscohost.com/login.aspx?direct=true&profile=ehost&scope=site&authtype=crawler&jrnl=10019332&asa=Y&AN=63484555&h=G6mLycMgNlOVPgRoCYkOUk10bJVLBylqio15fIxfweqCrJzP5PEtwMoYDxnyvIpHMRMXk%2BO8V2wSOuIGuR9DpA%3D%3D&crl=c)s and fish-scale pits. Our results provide a mechanistic understanding of how interacting processes of human activities at small catchments scales to influence carbon sequestration, and promote sustainable utlization of ESs in hill and gully regions of the Loess Plateau.

In the last few decades, Silchar city and its surrounding areas have witnessed massive population growth due to local and migrated peoples, which led to the rapid transformation of its land use/land cover (LU/LC) pattern. For detailed micro-level LU/LC analysis, very high-resolution satellite data (VHRS) and on-screen visual interpretation techniques were used. The LU/LC classification of the city has been prepared for 2005, 2011 and 2018 with five major and further seventeen sub-classes. The wards of the city were classified into five zones and the Shannon entropy model applied to analyze the degree and trends of urban built-up development. The obtained results indicate that the Silchar city is compact near the center which follows the infill built-up growth trend. However, as the distance has increased from its center the scattering with infill growth pattern was observed. The accuracy assessment techniques are used to validate the results of the classification. The results of this study can provide detailed information for land-use planners, researchers, policy-decision makers, and municipal authorities.

Soil cracks significantly affect preferential flow; however, there are some uncertainties associated with the effect of soil cracks on preferential flow in karst areas in southwest China. In this study, ground-penetrating radar (GPR) was applied to pedons to investigate the soil crack properties (inclusion, width, and configuration). Blue dye tracing experiments were designed, based on geophysical detection results, to assess the influences of inclusions (sand grains and rock fragments), crack width (1, 1.5, and 2 cm), and configuration (I-shape, V-shape, and Λ-shape) on the preferential flow. Our results indicated that (1) GPR envelope can describe the configuration of isolated soil cracks; (2) the Brilliant Blue FCF (C.I. Food Blue 2) infiltration rate and depth were over 1.5 times slower and 1.2–3.8 times lower those of water, respectively, during infiltration; (3) soil cracks can accelerate infiltration and increase the maximum dye-penetration depth, cumulative infiltration, and wetting front depth by at least an average of 5.2% and 63.2%, respectively; and (4) the I- and Λ-shaped soil crack configurations contributed to preferential flow, while the flow was not observed along the V-shaped configuration crack pore paths. The I-shaped configurations, with a width of 1.5 cm, were filled with rock fragments and had higher preferential flow ratios (18.2%–52.3%) and length indexes (4.0%–33.8%) than those of other configurations. Inclusions, crack widths, and configurations had significant influences on preferential flow ( p < 0.05). The influence of soil crack properties on preferential flow cannot be neglected during vegetation restoration and groundwater security processes in karst areas.