未来土地利用情景下城市地表径流变化模拟预测研究

【目的】在快速城市化背景下,土地利用格局发生改变,建筑用地等不透水面比例增大,导致地表径流深度增加。因此,理解土地利用变化与地表径流的相互关系在城市防灾减灾工作中具有重要作用。【方法】以杭州市为例,利用CLUMondo模型,得到2030年土地利用情景模拟结果,并在此基础上,利用SCS模型模拟预测了未来地表径流深度信息,探讨了土地利用变化对地表径流的影响。【结果】结果显示:(1)到2030年,耕地面积减少到2070.72 km^(2),共减少283.55 km^(2);建筑用地增加到1227.74 km^(2),共增加了528.19 km^(2),林地和草地均出现了不同程度的减少。(2)在同等降水条件情景下,随着土地利用格局的变化,研究区平均地表径流深度呈逐年增加趋势,但增幅存在差异。在研究区西部的林地地区地表径流增加不明显,在研究区东北部的中心城区,地表径流的范围与深度显著增加。其中杭州主城区径流深度增加最为明显。【结论】2005年到2030年杭州的土地利用结构变化中,耕地与建筑用地变化最为明显,耕地面积不断减少,建筑用地面积不断增加;地表径流深度整体呈增加趋势,但在不同时段和区域都存在一定的差异性,其中杭州主城区径流深度增加最为明显。2005—2030年间杭州降水径流变化的原因与杭州城市化过程中的土地利用转化特别是耕地向建筑用地的转化密切相关。研究成果可为优化城市土地利用政策、缓解洪涝灾害提供理论依据。

An improved area-based approach for estimating plot-level tree DBH from airborne LiDAR data

The diameter at breast height(DBH) of trees and stands is not only a widely used plant functional trait in ecology and biodiversity but also one of the most fundamental measurements in managing forests. However, systematically measuring the DBH of individual trees over large areas using conventional ground-based approaches is labour-intensive and costly. Here, we present an improved area-based approach to estimate plot-level tree DBH from airborne Li DAR data using the relationship between tree height and DBH, which is widely available for most forest types and many individual tree species. We first determined optimal functional forms for modelling heightDBH relationships using field-measured tree height and DBH. Then we estimated plot-level mean DBH by inverting the height-DBH relationships using the tree height predicted by Li DAR. Finally, we compared the predictive performance of our approach with a classical area-based method of DBH. The results showed that our approach significantly improved the prediction accuracy of tree DBH(R^(2)=0.85–0.90, rRMSE=9.57%–11.26%)compared to the classical area-based approach(R^(2)=0.80–0.83, rRMSE=11.98%–14.97%). Our study demonstrates the potential of using height-DBH relationships to improve the estimation of the plot-level DBH from airborne Li DAR data.

Spatio-temporal variations in trends of vegetation and drought changes in relation to climate variability from 1982 to 2019 based on remote sensing data from East Asia

Studying the significant impacts on vegetation of drought due to global warming is crucial in order to understand its dynamics and interrelationships with temperature,rainfall,and normalized difference vegetation index(NDVI).These factors are linked to excesses drought frequency and severity on the regional scale,and their effect on vegetation remains an important topic for climate change study.East Asia is very sensitive and susceptible to climate change.In this study,we examined the effect of drought on the seasonal variations of vegetation in relation to climate variability and determined which growing seasons are most vulnerable to drought risk;and then explored the spatio-temporal evolution of the trend in drought changes in East Asia from 1982 to 2019.The data were studied using a series of several drought indexes,and the data were then classified using a heat map,box and whisker plot analysis,and principal component analysis.The various drought indexes from January to August improved rapidly,except for vegetation health index(VHI)and temperature condition index(TCI).While these indices were constant in September,they increased again in October,but in December,they showed a descending trend.The seasonal and monthly analysis of the drought indexes and the heat map confirmed that the East Asian region suffered from extreme droughts in 1984,1993,2007,and 2012among the study years.The distribution of the trend in drought changes indicated that more severe drought occurred in the northwestern region than in the southeastern area of East Asia.The drought tendency slope was used to describe the changes in drought events during 1982–2019 in the study region.The correlations among monthly precipitation anomaly percentage(NAP),NDVI,TCI,vegetation condition index(VCI),temperature vegetation drought index(TVDI),and VHI indicated considerably positive correlations,while considerably negative correlations were found among the three pairs of NDVI and VHI,TVDI and VHI,and NDVI and TCI.This ecological and climatic mechanism provides a good basis for the assessment of vegetation and drought-change variations within the East Asian region.This study is a step forward in monitoring the seasonal variation of vegetation and variations in drought dynamics within the East Asian region,which will serve and contribute to the better management of vegetation,disaster risk,and drought in the East Asian region.

Landslide inventory and susceptibility modelling using geospatial tools,in Hunza-Nagar valley,northern Pakistan

A comprehensive landslide inventory and susceptibility maps are prerequisite for developing and implementing landslide mitigation strategies. Landslide susceptibility maps for the landslides prone regions in northern Pakistan are rarely available. The Hunza-Nagar valley in northern Pakistan is known for its frequent and devastating landslides. In this paper, we have developed a landslide inventory map for Hunza-Nagar valley by using the visual interpretation of the SPOT-5 satellite imagery and mapped a total of 172 landslides. The landslide inventory was subsequently divided into modelling and validation data sets. For the development of landslide susceptibility map seven discrete landslide causative factors were correlated with the landslide inventory map using weight of evidence and frequency ratio statistical models. Four different models of conditional independence were used for the selection of landslide causative factors. The produced landslides susceptibility maps were validated by the success rate and area under curves criteria. The prediction power of the models was also validated with the prediction rate curve. The validation results shows that the success rate curves of the weight of evidence and the frequency models are 82% and 79%, respectively. The prediction accuracy results obtained from this study are 84% for weight of evidence model and 80% for the frequency ratio model. Finally, the landslide susceptibility index maps were classified into five different varying susceptibility zones. The validation and prediction result indicates that the weight of evidence and frequency ratio model are reliable to produce an accurate landslide susceptibility map, which may be helpful for landslides management strategies.

New evidence for the links between the local water cycle and the underground wet sand layer of a mega-dune in the Badain Jaran Desert, China

Scientists and the local government have great concerns about the climate change and water resources in the Badain Jaran Desert of western China. A field study for the local water cycle of a lake-desert system was conducted near the Noertu Lake in the Badain Jaran Desert from 21 June to 26 August 2008. An underground wet sand layer was observed at a depth of 20–50 cm through analysis of datasets collected during the field experiment. Measurements unveiled that the near surface air humidity increased in the nighttime. The sensible and latent heat fluxes were equivalent at a site about 50 m away from the Noertu Lake during the daytime, with mean values of 134.4 and 105.9 W/m2 respectively. The sensible heat flux was dominant at a site about 500 m away from the Noertu Lake, with a mean of 187.7 W/m2, and a mean latent heat flux of only 26.7 W/m2. There were no apparent differences for the land surface energy budget at the two sites during the night time. The latent heat flux was always negative with a mean value of –12.7 W/m2, and the sensible heat flux was either positive or negative with a mean value of 5.10 W/m2. A portion of the local precipitation was evaporated into the air and the top-layer of sand dried quickly after every rainfall event, while another portion seeped deep and was trapped by the underground wet sand layer, and supplied water for surface psammophyte growth. With an increase of air humidity and the occurrence of negative latent heat flux or water vapor condensation around the Noertu Lake during the nighttime, we postulated that the vapor was transported and condensed at the lakeward sand surface, and provided supplemental underground sand pore water. There were links between the local water cycle, underground wet sand layer, psammophyte growth and landscape evolution of the mega-dunes surrounding the lakes in the Badain Jaran Desert of western China.

Estimation of Land Surface Temperature over the Tibetan Plateau Using AVHRR and MODIS Data

Estimation of large-scale land surface temperature from satellite images is of great importance for the study of climate change. This is especially true for the most challenging areas, such as the Tibetan Plateau （TP）. In this paper, two split window algorithms （SWAs）, one for the NOAA’s Advanced Very High Resolu-tion Radiometer （AVHRR）, and the other for the Moderate Resolution Imaging Spectroradiometer （MODIS）, were applied to retrieve land surface temperature （LST） over the TP simultaneously. AVHRR and MODIS data from 17 January, 14 April, 23 July, and 16 October 2003 were selected as the cases for winter, spring, summer, and autumn, respectively. Firstly, two key parameters （emissivity and water vapor content） were calculated at the pixel scale. Then, the derived LST was compared with in situ measurements from the Coordinated Enhanced Observing Period （CEOP） Asia-Australia Monsoon Project （CAMP） on the TP （CAMP/Tibet） area. They were in good accordance with each other, with an average percentage error （PE） of 10.5% for AVHRR data and 8.3% for MODIS data, meaning the adopted SWAs were applicable in the TP area. The derived LST also showed a wide range and a clear seasonal difference. The results from AVHRR were also in agreement with MODIS, with the latter usually displaying a higher level of accuracy.

Comparison of machine learning models for gully erosion susceptibility mapping

Gully erosion is a disruptive phenomenon which extensively affects the Iranian territory,especially in the Northern provinces.A number of studies have been recently undertaken to study this process and to predict it over space and ultimately,in a broader national effort,to limit its negative effects on local communities.We focused on the Bastam watershed where 9.3%of its surface is currently affected by gullying.Machine learning algorithms are currently under the magnifying glass across the geomorphological community for their high predictive ability.However,unlike the bivariate statistical models,their structure does not provide intuitive and quantifiable measures of environmental preconditioning factors.To cope with such weakness,we interpret preconditioning causes on the basis of a bivariate approach namely,Index of Entropy.And,we performed the susceptibility mapping procedure by testing three extensions of a decision tree model namely,Alternating Decision Tree(ADTree),Naive-Bayes tree(NBTree),and Logistic Model Tree(LMT).We dichotomized the gully information over space into gully presence/absence conditions,which we further explored in their calibration and validation stages.Being the presence/absence information and associated factors identical,the resulting differences are only due to the algorithmic structures of the three models we chose.Such differences are not significant in terms of performances;in fact,the three models produce outstanding predictive AUC measures(ADTree=0.922;NBTree=0.939;LMT=0.944).However,the associated mapping results depict very different patterns where only the LMT is associated with reasonable susceptibility patterns.This is a strong indication of what model combines best performance and mapping for any natural hazard-oriented application.

Characteristics of land-atmosphere energy and turbulentfluxes over the plateau steppe in central Tibetan Plateau

The land-atmosphere energy and turbulence exchange is key to understanding land surface processes on the Tibetan Plateau(TP). Using observed data for Aug. 4 to Dec. 3, 2012 from the Bujiao observation point(BJ) of the Nagqu Plateau Climate and Environment Station(NPCE-BJ), different characteristics of the energy flux during the Asian summer monsoon(ASM) season and post-monsoon period were analyzed. This study outlines the impact of the ASM on energy fluxes in the central TP. It also demonstrates that the surface energy closure rate during the ASM season is higher than that of the post-monsoon period. Footprint modeling shows the distribution of data quality assessments(QA) and quality controls(QC) surrounding the observation point. The measured turbulent flux data at the NPCE-BJ site were highly representative of the target land-use type. The target surface contributed more to the fluxes under unstable conditions than under stable conditions. The main wind directions(180° and 210°) with the highest data density showed flux contributions reaching 100%, even under stable conditions. The lowest flux contributions were found in sectors with low data density, e.g., 90.4% in the 360° sector under stable conditions during the ASM season. Lastly, a surface energy water balance(SEWAB) model was used to gap-fill any absent or corrected turbulence data. The potential simulation error was also explored in this study. The Nash-Sutcliffe model efficiency coefficients(NSEs) of the observed fluxes with the SEWAB model runs were 0.78 for sensible heat flux and 0.63 for latent heat flux during the ASM season, but unrealistic values of-0.9 for latent heat flux during the post-monsoon period.

Landslide characteristics and its impact on tourism for two roadside towns along the Kathmandu Kyirong Highway

Frequent landslide events affect the Kathmandu Kyirong Highway(KKH),one of the most strategic Sino-Nepal highways,with multiple social effects.Amongst them,the impacts on local tourism,although being substantial,have not been studied so far.The aim of this research is to analyze the characteristics of such landslides and their influence on road damages and/or blockages as well as on local tourism industry.We analyzed the co-seismic landslides triggered by the Gorkha Earthquake,2015(7.8 Mw),the post-seismic landslides that occurred during the monsoons following the earthquake,as well as landslides which occurred or reactivated in 2018,with relation to the damage that they caused to the highway.High resolution satellite images from 2015 to 2018,and field data were used for the analysis.The Langtang avalanche that locates off the highway was also mapped due to its high impacts on tourism.Between 2015 and 2018,the number of road damaging landslides in the Betrawati-Rasuwagadhi section of KKH(where Dhunche and Syafrubesi towns are located)was 101 in the main track(MT)and 103 in the new track(NT),with respective average density of 1.46/km and 3.63/km.The dominant observed landslide types were debris slides and rock falls.Landslides were mostly concentrated in the locations with the following characteristics:1)having higher elevated area,2)being located with the‘main central thrust’and other lineaments’belts,3)belonging to the Proterozoic lesser Himalayan rocks,4)having a slope gradient of 25°-45°,5)having northern,western and southern slope aspect,6)being subjected to average annual rainfall of higher than 1,000 mm,and 7)having less than 4 km distance from the past earthquake epicenters.The results further indicated that 7 rain-induced and 4 co-and post-seismic landslides have great impact on tourist flows.An impact analysis was also assessed through a door to door questionnaire survey with local hotel operators from Dhunche and Syafrubesi towns(n=29+31).The results reveal that out of six rigorously affected sectors by landslides leading to road blockage,tourism business is the most impacted livelihood sector in these towns.The reduction of visitors in different hotels ranged from 50%-100%in Dhunche and 70%-100%in Syafrubesi for the first year aftermath of the tremor.This is higher than the respective 5%-50%tourist reduction due to raininduced landslides.Using as a reference the base year 2014,the income loss of hotels in both towns was found to be 50%-100%in 2015,20%-100%in 2016,5%-75%in 2017,and similar to 35%in 2018.These results provide insights on the synergic effect of contributing factors for cut slope as well as down slope instability along mountainous motorways and their impact on income sources for local communities.

Landslide susceptibility assessment at Kathmandu Kyirong Highway Corridor in pre-quake, co-seismic and post-quake situations

Kathmandu Kyirong Highway(KKH)is one of the most strategic Sino-Nepal highways.Lowcost mitigation measures are common in Nepalese highways,however,they are not even applied sufficiently to control slope instability since the major part of this highway falls still under the category of feeder road,and thus less resources are made available for its maintenance.It is subjected to frequent landslide events in an annual basis,especially during monsoon season.The Gorkha earthquake,2015 further mobilized substantial hillslope materials and damaged the road in several locations.The aim of this research is to access the dynamic landslide susceptibility considering pre,co and post seismic mass failures.We mapped 5,349 multi-temporal landslides of 15 years(2004-2018),using high resolution satellite images and field data,and grouped them in aforementioned three time periods.Landslide susceptibility was assessed with the application of’certainty factor’(CF).Seventy percent landslides were used for susceptibility modelling and 30%for validation.The obtained results were evaluated by plotting’receiver operative characteristic’(ROC)curves.The CF performed well with the’area under curve’(AUC)0.820,0.875 and 0.817 for the success rates,and 0.809,0.890 and 0.760 for the prediction rates for respective pre,co and post seismic landslide susceptibility.The accuracy for seismic landslide susceptibility was better than pre and post-quake ones.It might be because of the differences on completeness of the landslide inventory,which might have been possibly done better for the single event based co-seismic landslide mapping in comparison with multitemporal inventories in pre and post-quake situations.The results obtained in this study provide insights on dynamic spatial probability of landslide occurrences in the changing condition of triggering agents.This work can be a good contribution to the methodologies for the evaluation of the dynamic landslide hazard and risk,which will further help to design the efficient mitigation measures along the mountain highways.

Fully Convolutional Networks for Street Furniture Identification in Panorama Images

Panoramic images are widely used in many scenes,especially in virtual reality and street view capture.However,they are new for street furniture identification which is usually based on mobile laser scanning point cloud data or conventional 2D images.This study proposes to perform semantic segmentation on panoramic images and transformed images to separate light poles and traffic signs from background implemented by pre-trained Fully Convolutional Networks(FCN).FCN is the most important model for deep learning applied on semantic segmentation for its end to end training process and pixel-wise prediction.In this study,we use FCN-8s model that pre-trained on cityscape dataset and finetune it by our own data.Then replace cross entropy loss function with focal loss function in the FCN model and train it again to produce the predictions.The results show that in all results from pre-trained model,fine-tuning,and FCN model with focal loss,the light poles and traffic signs are detected well and the transformed images have better performance than panoramic images in the prediction according to the Recall and IoU evaluation.

Exploration and Application of Ubiquitous Mapping in Engineering Practice Courses in the Intelligence Age

In the current era of digital surveying and mapping to intelligent surveying and mapping,ubiquitous surveying and mapping has brought many opportunities and challenges to college engineering course teaching.With the development of ubiquitous surveying and mapping,college engineering practice courses urgently need to respond to ubiquitous surveying and mapping.The research aims to integrate the development of ubiquitous surveying and mapping into the teaching of engineering practice courses in colleges,including promoting Android,Brower/Server(B/S),and Client/Server(C/S)to build a platform for practice courses.This also incorporates real development cases in measurement data processing such as gravity field refinement.In this way,the teaching level of engineering practice courses in colleges can be improved,and new ideas can be put forward for cultivating surveying and mapping talents in the new era in colleges.Finally,it can also provide new ideas for the organization of surveying and mapping practice courses under the background of the pandemic.

Expansion of traditional land-use and deforestation:a case study of an adat forest in the Kandilo Subwatershed,East Kalimantan,Indonesia

Deforestation issues are more problematic when indigenous(adat) communities,living within a forest,have lived there for many generations.These adat communities,who employ traditional land-use,are frequently accused of encroaching on the forest.To understand existing and future trends in the spatial patterns of the expansion of traditional land-use and deforestation,we conducted a case study in the Kandilo Subwatershed using mixed methods with image interpretation,spatial modelling and sociocultural surveys to examine the interrelationships between physical conditions,community characteristics and traditional land-use expansion.We investigated community characteristics through household interviews,communication with key informants,and discussions with focusgroups.By using an area production model,we were able to analyze the effect of improved farming systems,policy intervention and law enforcement on traditional land-use expansion and deforestation.Based on our examination of a 20-year period of traditional land-use activities in adat forests,the evidence indicated that the steeper the slope of the land and the farther the distance from the village,the lower the rate of deforestation.Our study found that customary law,regulating traditional land-use,played an important role in controlling deforestation and land degradation.We conclude that the integration of land allocation,improved farming practices and enforcement of customary law are effective measures to improve traditional land productivity while avoiding deforestation and land degradation.

Habitat requirements of Boreal Owl(Aegolius funereus)and Pygmy Owl(Glaucidium passerinum)in rear edge montane populations on the Balkan Peninsula

Rear-edge populations of montane species are known to be vulnerable to environmental change,which could affect them by habitat reduction and isolation.Habitat requirements of two cold-adapted boreo-alpine owl species—Boreal Owl(Aegolius funereus)and Pygmy Owl(Glaucidium passerinum)—have been studied in refugial montane populations in the western Rhodopes,South Bulgaria.Data on owl presence and forest stand attributes recorded in situ have been used to identify significant predictors for owl occurrence.The results revealed Boreal Owl’s preference for comparatively dense forests(high canopy closure values),big trees(diameter at breast height≥50 cm)and large amount of fallen dead wood in penultimate stage of decay.For Pygmy Owl the only significant explanatory variable was the total amount of fallen dead wood.Results suggest preference of both owl species for forests with structural elements typical of old-growth forests(i.e.,veteran trees,deadwood),the Pygmy Owl being less prone to inhabit managed forests.Being at the rear edge of their Palearctic breeding range in Europe both Boreal and Pygmy Owls are of high conservation value on the Balkan Peninsula.Hence,additional efforts are needed for their conservation in the light of climate change and resulting alteration of forest structural parameters.Current findings can be used for adjusting forest management practices in order to ensure both,sustainable profit from timber and continuous species survival.

Precipitation Regime Shift Associated with the Pacific Decadal Oscillation in the Maritime Continent

Recent changes in precipitation regime in South-East Asia are a subject of ongoing discussion. In this article, for the first time, evidence of a precipitation regime shift during the mid-1970s in the Northern Hemispheric part of South-East Asia is demonstrated. The detection of regime shifts is made possible by using a new comprehensive dataset of daily precipitation records (South-East Asian Climate Assessment and Dataset) and applying a novel Bayesian approach for regime shift detection. After the detected regime shift event in the mid-1970s, significant changes in precipitation distribution occurred in the Northern Hemispheric regions—Indochina Peninsula and the Philippines. More specifically, dry days became up to 10% more frequent in some regions. However, no precipitation regime shift is detected in Southern Hemisphere regions—Java and Northern Australia, were the number of observed dry days increased gradually.

两倍的青藏高原蒸散发增长速率

Estimation of evapotranspiration(ET_(a))change on the Tibetan Plateau(TP)is essential to address the water requirement of billions of people surrounding the TP.Existing studies have shown that ET_(a)estimations on the TP have a very large uncertainty.In this article,we discuss how to more accurately quantify ET_(a)amount and explain its change on the TP.ET_(a)change on the TP can be quantified and explained based on an ensemble mean product from climate model simulations,reanalysis,as well as ground-based and satellite observations.ET_(a)on the TP experienced a significant increasing trend of around 8.4±2.2 mm(10 a)^(-1)(mean±one standard deviation)during 1982–2018,approximately twice the rate of the global land ET_(a)(4.3±2.1 mm(10 a)^(-1)).Numerical attribution analysis revealed that a 53.8%TP area with the increased ET_(a)was caused by increased temperature and 23.1%part was due to soil moisture rising,because of the warming,melting cryosphere,and increased precipitation.The projected future increase in ET_(a)is expected to cause a continued acceleration of the water cycle until 2100.

Assessment of SoilGrids data for soil erosion estimation at watershed scale: A case study in northern Thailand

Soil erosion has been identified as one of the most destructive forms of land degradation,posing a threat to the sustainability of global economic,social and environmental systems.This underscores the need for sustainable land management that takes erosion control and prevention into consideration.This requires the use of state-of-the-art erosion prediction models.The models often require extensive input of detailed spatial and temporal data,some of which are not readily available in many developing countries,particularly detailed soil data.The soil dataset Global Gridded Soil Information(SoilGrids)could potentially fill the data gap.Nevertheless,its value and accuracy for soil erosion modelling in the humid tropics is still unknown,necessitating the need to assess its value vis-à-vis field-based data.The major objective of this study was to conduct a comparative assessment of the value of SoilGrids and field-based soil data for estimating soil loss.Soil samples were collected from five physiographic positions(summit,shoulder,back slope,foot slope,and toe slope)using the soil catena approach.Samples were collected using a 5-cm steel sample ring(undisturbed)and a spade(disturbed).Data of the landform,predominant vegetation types,canopy cover,average plant height,land use,soil depth,shear strength,and soil color were recorded for each site.The soil samples were subjected to laboratory analysis for saturated hydraulic conductivity,bulk density,particle size distribution,and organic matter content.Pedotransfer functions were applied on the SoilGrids and field-based data to generate soil hydrological properties.The resultant field-based data were compared with the SoilGrids data for corresponding points/areas to determine the potential similarities of the two datasets.Both datasets were then used as inputs for soil erosion assessment using the revised Morgan-Morgan-Finney model.The results from both datasets were again compared to determine the degree of similarity.The results showed that with respect to point-based comparison,both datasets were significantly different.At the hillslope delineation level,the field-based data still consistently had a greater degree of variability,but the hillslope averages were not significantly different for both datasets.Similar results were recorded with the soil loss parameters generated from both datasets;point-based comparison showed that both datasets were significantly different,whereas the reverse was true for parcel/area-based comparison.SoilGrids data are certainly useful,especially where soil data are lacking;the utility of this dataset is,however,dependent on the scale of operation or the extent of detail required.When detailed,site-specific data are required,SoilGrids may not be a good alternative to soil survey data in the humid tropics.On the other hand,if the average soil properties of a region,area,or land parcel are required for the implementation of a particular project,plan,or program,SoilGrids data can be a very valuable alternative to soil survey data.

From spatio-temporal landslide susceptibility to landslide risk forecast

The literature on landslide susceptibility is rich with examples that span a wide range of topics.However,the component that pertains to the extension of the susceptibility framework toward space–time modeling is largely unexplored.This statement holds true,particularly in the context of landslide risk,where few scientific contributions investigate risk dynamics in space and time.This manuscript proposes a modeling protocol where a dynamic landslide susceptibility is obtained via a binomial Generalized Additive Model whose inventories span nine years(from 2013 to 2021).For the analyses,the data cube is organized with a mapping unit consisting of 26,333 slope units repeated over an annual temporal unit,resulting in a total of 236,997 units.This phase already includes several interesting modeling experiments that have rarely appeared in the landslide literature(e.g.,variable interaction plots).However,the main innovative effort is in the subsequent phase of the protocol we propose,as we used climate projections of the main trigger(rainfall)to obtain future estimates of yearly susceptibility patterns.These estimates are then combined with projections of urban settlements and associated populations to create a dynamic risk model,assuming vulnerability=1.Overall,this manuscript presents a unique example of such a modeling routine and offers a potential standard for administrations to make informed decisions regarding future urban development.

Adopting the margin of stability for space–time landslide prediction–A data-driven approach for generating spatial dynamic thresholds

Shallow landslide initiation typically results from an interplay of dynamic triggering and preparatory conditions along with static predisposition factors.While data-driven methods for assessing landslide susceptibility or for establishing rainfall-triggering thresholds are prevalent,integrating spatiotemporal information for dynamic large-area landslide prediction remains a challenge.The main aim of this research is to generate a dynamic spatial landslide initiation model that operates at a daily scale and explicitly counteracts potential errors in the available landslide data.Unlike previous studies focusing on space–time landslide modelling,it places a strong emphasis on reducing the propagation of landslide data errors into the modelling results,while ensuring interpretable outcomes.It introduces also other noteworthy innovations,such as visualizing the final predictions as dynamic spatial thresholds linked to true positive rates and false alarm rates and by using animations for highlighting its application potential for hindcasting and scenario-building.The initial step involves the creation of a spatio-temporally representative sample of landslide presence and absence observations for the study area of South Tyrol,Italy(7400 km2)within well-investigated terrain.Model setup entails integrating landslide controls that operate on various temporal scales through a binomial Generalized Additive Mixed Model.Model relationships are then interpreted based on variable importance and partial effect plots,while predictive performance is evaluated through various crossvalidation techniques.Optimal and user-defined probability cutpoints are used to establish quantitative thresholds that reflect both,the true positive rate(correctly predicted landslides)and the false positive rate(precipitation periods misclassified as landslide-inducing conditions).The resulting dynamic maps directly visualize landslide threshold exceedance.The model demonstrates high predictive performance while revealing geomorphologically plausible prediction patterns largely consistent with current process knowledge.Notably,the model also shows that generally drier hillslopes exhibit a greater sensitivity to certain precipitation events than regions adapted to wetter conditions.The practical applicability of the approach is demonstrated in a hindcasting and scenario-building context.In the currently evolving field of space–time landslide modelling,we recommend focusing on data error handling,model interpretability,and geomorphic plausibility,rather than allocating excessive resources to algorithm and case study comparisons.

On the use of explainable AI for susceptibility modeling:Examining the spatial pattern of SHAP values

Hydro-morphological processes(HMP,any natural phenomenon contained within the spectrum defined between debris flows and flash floods)are globally occurring natural hazards which pose great threats to our society,leading to fatalities and economical losses.For this reason,understanding the dynamics behind HMPs is needed to aid in hazard and risk assessment.In this work,we take advantage of an explainable deep learning model to extract global and local interpretations of the HMP occurrences across the whole Chinese territory.We use a deep neural network architecture and interpret the model results through the spatial pattern of SHAP values.In doing so,we can understand the model prediction on a hierarchical basis,looking at how the predictor set controls the overall susceptibility as well as doing the same at the level of the single mapping unit.Our model accurately predicts HMP occurrences with AUC values measured in a ten-fold cross-validation ranging between 0.83 and 0.86.This level of predictive performance attests for an excellent prediction skill.The main difference with respect to traditional statistical tools is that the latter usually lead to a clear interpretation at the expense of high performance,which is otherwise reached via machine/deep learning solutions,though at the expense of interpretation.The recent development of explainable Al is the key to combine both strengths.In this work,we explore this combination in the context of HMP susceptibility modeling.Specifically,we demonstrate the extent to which one can enter a new level of data-driven interpretation,supporting the decision-making process behind disaster risk mitigation and prevention actions.