Search

Abstract In Southwest China (SWC) soil loss is becoming a serious threat to the sustainable development of resources and the health of the environment. A large number of studies have been conducted to investigate the causes (e.g. environmental factors and human activities) and impacts (e.g. karst rocky desertification) of soil loss. However, the following are some important problems related to soil loss in the karst of SWC: 1) although soil loss is caused by chemical weathering, gravity and water interaction in karst, the mechanisms by which soil loss takes place in the SWC and their interaction are still not clear; 2) subsurface soil loss varies under different geological and hydrological conditions, and there is still no consensus on the common the ratio between surface soil erosion and underground soil loss; 3) traditional methods for soil loss risk assessment are difficult to apply over the whole of SWC; and 4) a large number of studies have been published only in Chinese and remain inaccessible to the international science community. Therefore, a review paper is needed to address this gap and introduce the contributions of Chinese researchers to soil loss in karst. This paper proposes a novel modelling approach for the grouped soil loss styles and presents a comprehensive discussion for the mechanism, causes, and risk assessment of the various soil loss styles in SWC karst. Graphical abstract Unlabelled Image Highlights • Soil is eroded not only in the surface but also in the underground rock layer. • Surface/subsurface soil loss and soil leakage are typical soil loss styles in karst. • Novel modelling is proposed to discretize the grouped soil loss styles in karst. [ABSTRACT FROM AUTHOR]

Paleo-altitude reconstructions using oxygen isotopes provide past surface elevation data that can constrain the topographic evolution of mountain belts. This paper presents oxygen and carbon stable isotopic compositions of gastropod fossils, carbonate cements, and shelly sandstones in the Anjihai and Shawan formations located in the Sikeshu drainage basin, northern Tianshan, western China. Correlations between the stable isotopic compositions and lithofacies suggest that the Sikeshu drainage basin was mainly an open hydrological depositional environment. Based on the oxygen isotopic compositions of the gastropod fossils and petrographic examination, the δ13C and δ18O values have not been reset by late-stage diagenesis. Paleo-altitudes were reconstructed using the corrected paleo-surface water δ18O psw values recovered from gastropod fossils, carbonate cements, and shelly sandstones. An empirical formula was developed to calculate the paleo-latitudes when the Anjihai and Shawan formations were deposited. The paleo-altitude of the Sikeshu drainage basin had attained an elevation exceeding 2000 m by 28 Ma, similar to the present-day elevation in this area. This conclusion contributes to our understanding of the extent of surface uplift in the Cenozoic of Tianshan and the climatic differences on either side of Tianshan during the Cenozoic. Unlabelled Image • This study is the first in Tianshan to provide quantitative evidence for uplift and exhumation. • δ18O date of gastropod fossil analyzed in situ by NanoSIMS rule out diagnetic reset. • Paleoelevation of the north of Tianshan have exceeded 2000 m at least before 28 Ma. [ABSTRACT FROM AUTHOR]

The occurrence of earthquakes has the potential to significantly modify river morphology and cause notable effects on the aquatic ecosystem. Prior to the 2008 Wenchuan earthquake with a magnitude of 8.0, the Shiting River located in the upper basin of the Yangtze River was stable. Rapid bed incision or degradation has continued since the earthquake, and weirs and bank revetments were constructed to mitigate degradation. The impacts of channel morphology adjustment and human activities on biodiversity are not clear, which prohibits efficient river restoration. This paper evaluated the morphological quality of the Shiting River by employing the Morphological Quality Index (MQI), an approach recommended in the European Water Framework Directive. Macroinvertebrate diversity was analyzed in relation to the impacts of morphological adjustment and human activities based on field data collected in 2021. The results showed that the channel degraded and deepened after the earthquake, the planform changed from multi-threaded to single-threaded, and swamps on the floodplains disappeared. The quality of channel morphology in 2021 was evaluated using the MQI and was found to be ranging from poor to moderate. The diversity metrics of macroinvertebrates, specifically the Shannon, Simpson, and Evenness indices, exhibited a positive and significant correlation with four MQI sub-indices, which represented channel geomorphological functionality, artificiality, morphological pattern, and substrate, respectively. However, there was no significant correlation observed between the diversity metrics and the overall MQI score. The results imply that channels with superior morphological functionality, slower channel adjustment, and lower levels of anthropogenic disturbances tend to exhibit greater macroinvertebrate diversity. The study found a significant decrease in macroinvertebrate richness downstream of weirs and bridges. The community structures upstream and downstream of weirs and bridges were also significantly different. This study may help inform the ecological response of rivers to dynamic morphological changes and intense anthropogenic disturbances following disastrous earthquakes. • Post-earthquake channel morphological quality was poor to moderate. • Channels with higher morphological quality had higher macroinvertebrate diversity. • Morphology changes and human activities impacted macroinvertebrates. • Higher macroinvertebrate richness upstream of weirs and bridges than downstream. [ABSTRACT FROM AUTHOR]

The coastal mountain system (CMS) of southeastern China has been a natural place to study paleotopographic changes primarily due to its marked topographical contrast since the Late Cretaceous. For a quantitative reconstruction of the paleotopographic evolution of the CMS, this paper provides key information from along the Ningde–Jianyang and Huilai–Longchuan cross-sections using apatite (U-Th)/He (AHe) and zircon (U-Th)/He (ZHe) data combined with thermal history and paleotopographic modeling. For the Ningde–Jianyang section, the AHe dates are in the range of 44.3–70.8 Ma, and the ZHe dates are between 58.5 and 102.9 Ma. For the Huilai–Longchuan section, the resulting AHe dates are in the range of 37.9–73.8 Ma, and ZHe dates are between 65.5 and 119.7 Ma. The modeling results suggest that: 1) the CMS had an asymmetric topography during the Late Cretaceous with the SE segment being significantly higher (peak elevation = 3.8 km) than the NW segment (peak elevation = ~2 km). The marked decreases in elevation and relief of the CMS occurred during 80–40 Ma, followed by relatively minor topographic changes since ~40–30 Ma; 2) The SE segment of the southern CMS experienced significantly faster exhumation (~0.44 km/Myr) than northern CMS (0.16 km/Myr) during 70–60 Ma, which was probably related to thinning of the crust and orogenic collapse linked to the initial-rifting of the South China Sea; 3) The SE segment of southern CMS experienced slower exhumation than northern CMS during 60–40 Ma, likely resulting from the lower elevations in the southern CMS than in the northern CMS during 60–40 Ma. • The coastal mountain system forms a climatic boundary in southeastern China. • Thermochronological data collected along two horizontal transects. • Transect thermochronology results used for topography modeling. • Rapid exhumation from 70-60 Ma along the south 2-D transect linked to rifting. • Rapid exhumation from 60–40 Ma along the north 2-D transect linked to climate change. [ABSTRACT FROM AUTHOR]

Abstract Massive bed degradation (20 m in 7 years) has been observed in the Shi‑ting River, Sichuan Province, China, since the 2008 Wenchuan Ms. 8.0 earthquake. The reason for the massive bed degradation has not been well understood. A hypothesis has been proposed that relates bed degradation to the augmentation of sand supply after the earthquake. The effect of sand on gravel mobility (magic sand effect) has long been observed in laboratory experiments. In this paper, we study whether the augmentation of sand supply and its magic sand effect can lead to the observed massive degradation at decadal scales. A one-dimensional river morphodynamic model is implemented to study the problem in general at field scale and in context of the gravel-bed Shi‑ting River. Sediment transport is calculated with the Wilcock and Crowe (2003) relation, in which the magic sand effect is explicitly embedded in terms of a function relating reference Shields number to the surface sand fraction. We find that the augmentation of gravel supply leads to bed aggradation, whereas the augmentation of sand supply can indeed lead to bed degradation and surface fining. The magnitude and timescale of bed degradation are not sensitive to the sand supply rate but are sensitive to the flood intermittency factor , i.e., the fraction of time the river is in flood. However, an unrealistic flood intermittency factor (≥0.3) would be required in order to match the observed timescale of bed degradation, thus indicating that the magic sand effect might not be the governing reason for massive bed degradation in the Shi‑ting River, and by implication in gravel-bed rivers in general. Our simulation results also indicate that despite the fact that magic sand effects are not explicitly included in most sediment transport relations, they are at least partly built in via the hiding function that is contained in most sediment transport relations for gravel-sand mixtures. While here we use the Shi‑ting River as an example, our results have applicability to gravel-bed rivers subjected to augmentation in sand supply in general. Highlights • Augmentation of gravel supply leads to bed aggradation • Augmentation of sand supply leads to enhanced gravel transport and bed degradation • Magic sand effects alone cannot induce the observed river incision at decadal scale • Magic sand effects can be partly incorporated with a dimensionless hiding function [ABSTRACT FROM AUTHOR]

In this paper we present new and review already existing landslide and earthquake data for a large part of the Tien Shan, Central Asia. For the same area, only partial databases for sub-regions have been presented previously. They were compiled and new data were added to fill the gaps between the databases. Major new inputs are products of the Central Asia Seismic Risk Initiative (CASRI): a tentative digital map of active faults (even with indication of characteristic or possible maximum magnitude) and the earthquake catalogue of Central Asia until 2009 that was now updated with USGS data (to May 2014). The new compiled landslide inventory contains existing records of 1600 previously mapped mass movements and more than 1800 new landslide data. Considering presently available seismo-tectonic and landslide data, a target region of 1200 km (E–W) by 600 km (N–S) was defined for the production of more or less continuous geohazards information. This target region includes the entire Kyrgyz Tien Shan, the South-Western Tien Shan in Tajikistan, the Fergana Basin (Kyrgyzstan, Tajikistan and Uzbekistan) as well as the Western part in Uzbekistan, the North-Easternmost part in Kazakhstan and a small part of the Eastern Chinese Tien Shan (for the zones outside Kyrgyzstan and Tajikistan, only limited information was available and compiled). On the basis of the new landslide inventory and the updated earthquake catalogue, the link between landslide and earthquake activity is analysed. First, size–frequency relationships are studied for both types of geohazards, in terms of Gutenberg–Richter Law for the earthquakes and in terms of probability density function for the landslides. For several regions and major earthquake events, case histories are presented to outline further the close connection between earthquake and landslide hazards in the Tien Shan. From this study, we concluded first that a major hazard component is still now insufficiently known for both types of geohazards – and even totally unknown for most giant landslides: the time information. Thus, dating rockslides and any large mass movement is a crucial factor in order to better constrain related hazards. Second, we highlight the role of coupled hazards in the Tien Shan – inside the mountain range the largest disasters were caused by earthquake-triggered landslides and massive earth flows. Therefore, it is of prime importance for these mountain regions to complement single hazard assessment by coupled hazard scenarios. On the basis of these scenarios, risk evaluations may then be completed considering direct impacts on artificial dams, hydropower schemes, mining and road infrastructure as well as remote effects from possible dam breaches and mobilisation of mining and nuclear waste storages. Basic hazard and risk components are partly analysed in this paper and partly in a companion paper ‘Tien Shan geohazards database: Landslide susceptibility and impacts’. [ABSTRACT FROM AUTHOR]

Numerical modeling of landform evolution is the key to understanding the formation of land surfaces. However, landform evolution occurs under certain initial topographic conditions, while existing landform evolution models often use a plane or a randomly produced surface to establish the initial topographic conditions. This misunderstanding of the initial topographic conditions causes the landform evolution results to deviate greatly from the real terrain. Therefore, the initial topographic conditions must be reconstructed and integrated into the current numerical landform evolution model. In this paper, the Loess Plateau of China was selected as a typical sample area. The paleotopography was reconstructed and integrated into the numerical landform evolution model. Based on the evolutionary mechanism of loess landforms, modules of surface erosion, matter diffusion and precipitation were selected to construct a numerical model of loess landform evolution. Moreover, randomly produced terrain was used as the initial conditions during the landform evolution process and compared with the paleotopography-constrained evolution results. In addition, the results of the landform evolution process were evaluated with different terrain factors, and the uncertainty of the model was assessed by comparing different model conditions. The results show that, due to the control of paleotopography, the evolution results can maintain the main morphological structure of valleys in the loess landscape. Without considering the paleotopography, a large difference exists between the evolution results and the true ground terrain. These results show that paleotopography is an important feature of loess landform evolution modeling and is indispensable for reasonably expressing the process of landform evolution. Finally, the numerical modeling of loess landform evolution with the consideration of paleotopography can not only help to achieve a better understanding of the loess landform evolution process but also provide a reference for feature dependent landform evolution modeling in different landform applications. • Paleotopography affects the evolution of loess landform in the Loess Plateau. • We integrate the paleotopography into the numerical loess landform evolution model. • Our results can maintain the main morphological structure of loess landform. • Our research provides a reference for feature-concerned landform evolution modeling. [ABSTRACT FROM AUTHOR]

Ground fissures are a geological hazard with complex formation mechanisms. Increasing amounts of human activity have created more ground fissures, which can destroy buildings and threaten human security. Some ground fissures indicate potentially devastating earthquakes, so we must pay attention to these hazards. This paper documents recently discovered ground fissures in the Hetao basin. These ground fissures are located along the frontal margins of the terraces of the Sertengshan piedmont fault. These fissures are 600–1600 m long, 5–50 cm wide, and at most 1 m deep. These ground fissures emerged after 2010 and ruptured newly constructed roads and field ridges. The deep geodynamic mechanisms within this extensional environment, which is dominated by NE-SW principal compressive shear, involve N-S tensile stress, which has produced continuous subsidence in the Hetao basin and continuous activity along the Sertengshan piedmont fault since the late Quaternary. Trenches across the ground fissures reveal that the fissures are the latest manifestation of the activity of preexisting faults and are the result of creep-slip movement along the faults. The groundwater level in the Hetao basin has been dropping since the 1960s because of overexploitation, resulting in subsidence. When the tensile stress exceeds the ultimate tensile strength of the strata, the strata rupture along preexisting faults, producing ground fissures. Thus, the Sertengshan piedmont fault planes are the structural foundation of the ground fissures, and groundwater extraction induces the development of ground fissures. [ABSTRACT FROM AUTHOR]

Cirques are typical erosional landforms of glaciers and have been used as bases for paleoclimate and paleoenvironment reconstruction and for understanding the interactions between glacial erosion, climate, and topography. The availability of high-resolution digital elevation models (DEMs) provides the opportunity to map large populations of cirques for regional and global scale analysis. However, cirque outlines are still mainly determined based on manual digitization, which is time consuming and labor intensive. This paper introduces an automated method to recognize and delineate cirques using DEMs based on a series of hydrological and morphological analyses, including delineating stream network, filtering streams, determining potential cirque threshold points, and delineating cirque outlines. A semi-automated tool is also developed based on user-specified threshold points or cross sections. The related tools are coded in python and imported into ArcGIS as a toolbox, AutoCirque, with user friendly interfaces. Comparison in a test area of the eastern Tian Shan, China, indicated that the population statistics are relatively consistent between manually digitized and auto-delineated cirques. Detailed comparisons for 11 selected cirques indicated that the AutoCirque-delineated and manually digitized cirque outlines are similar in shape with an average boundary offset of approximately one DEM cell size (30 m) and a 70–90% overlap-fit percentage. The derived cirque metrics are also similar, especially for elevation, slope, and aspect related metrics. This toolbox can significantly speed up the analytical processes, remove the subjectivity in delineating cirque outlines, and allow for the comparison of cirque morphology and metrics at regional and global scales. • This paper presents an automated method to delineate cirque outlines from DEMs. • The developed ArcGIS toolbox incudes both fully automated and semi-automated tools. • Delineated cirques have consistent population statistics with manual-digitized ones. • Manual- and auto-derived cirques are similar in shape with high overlap percentage. • This toolbox allows for large population analyses of cirques at different scales. [ABSTRACT FROM AUTHOR]

Reconstructing three-dimensional palaeoglacier surface, thickness, and volume provides useful information for quantifying palaeo-climate conditions during glacial stages and assessing the impact of glacier change on water resources. A glacial flowline model has been implemented in Excel and ArcGIS to reconstruct palaeoglaciers based on digital elevation models and geomorphic constraints, such as moraines and trimlines that represent the margins and local heights of palaeoglaciers, respectively. However, significant efforts are still needed in using these tools for palaeoglacier reconstruction, such as digitizing flowlines, adjusting shear stress, and deriving shape factors. This paper presents an automated method, PalaeoIce, to reconstruct palaeoglaciers using geomorphic constraints. Coded in python, PalaeoIce is an ArcGIS toolbox with a set of tools to generate glacial flowlines, optimize shear stress, derive shape factors, calculate ice thickness values along flowlines, and interpolate palaeo ice surfaces based on geomorphic constraints. The use of PalaeoIce is demonstrated using extant glaciers with measured thickness values and moraines/trimlines of the Little Ice Age (LIA) and Marine Isotope Stage (MIS) 2 in the eastern Tian Shan, China. The reconstructed ice thickness values of Glacier No. 1 in this area show a reasonable agreement with field measurements with 1.4 % and 5.4 % differences between measured and reconstructed mean ice thickness values using the two methods included in PalaeoIce: a revised ice thickness model for extant glaciers and a palaeoglacier reconstruction method based on glacier outlines, respectively. Two approaches are then illustrated for palaeoglacier reconstructions using two sets of geomorphic constraints: one with the fully reconstructed glacier outlines during the LIA, and the other with just the terminal moraine and limited trimlines during MIS 2. PalaeoIce can significantly improve the efficiency of palaeoglacier reconstructions and the understanding of the impact of glacier change on water resources. • An automated method, PalaeoIce, is presented for palaeoglacier reconstruction. • Tools are developed to create flowlines, modify DEM, and reconstruct palaeoglaciers. • Reconstructed ice thickness data are validated by field measurements in Tian Shan. • Target geomorphic features are of importance for palaeoglacier reconstruction. [ABSTRACT FROM AUTHOR]

This paper is the second part of a new geohazards analysis applied to a large part of the Tien Shan, Central Asia, focused on landslide susceptibility computations that are based on recently compiled geographic, geological and geomorphological data. The core data are a digital elevation model, an updated earthquake catalogue, an active fault map as well as a new landslide inventory. The most recently added digital data are a new simplified geological map, an annual precipitation map, as well as river and road network maps that were produced for the Kyrgyz and Tajik parts of the Tien Shan. On the basis of these records we determine landslide densities with respect to morphological (M), geological (G), river distance (R), precipitation (P), earthquake (E) and fault (F) distance factors. Correlations were also established between scarp locations and the slope angle, distance to rivers, curvature. These correlations show that scarps tend to be located on steeper slopes, farther from rivers and on more convex terrain than the entire landslides. On the basis of the landslide density values computed for each class of the aforementioned factors, two landslide susceptibility maps are created according to the Landslide Factor analysis: the first one considers correlations between the landslide occurrences and the first four factors (MGRP); the second one is based on the first map (MGRP) combined with the seismo-tectonic influence (+ E + F) on landslide distributions. From the comparison of these two maps with actual landslide distributions we infer that the distances to rivers as well as to faults and past earthquakes most strongly constrain the susceptibility of slopes to landslides. We highlight several zones where the landslide susceptibilities computed for the MGRP + E + F factors fit better the observed concentration of landslides than those computed for the MGRP factors alone. For a few zones, both maps produce high landslide susceptibilities that do not well reflect the observed low sub-regional landslide activity; for some cases, we consider that some influencing factors must not have been well taken into consideration, for others we show that we simply had missed landslide detections. At the scale of the mountain range, the computed landslide susceptibility maps fit the observed landslide distributions relatively well, but these maps only represent the spatial component of landslide hazards. Temporal aspects are not considered by this analysis. [ABSTRACT FROM AUTHOR]

Although temperature is considered a key factor influencing rock decay, little is known about the quantitative relationship between a rock's decay rate and temperature, which makes it difficult to quantitatively predict the rate of rock decay under varied temperature during wetting–drying cycle conditions. Thus, the aim of the present paper was to observe the decay rate of various purple mudstones and to develop a model with which to calculate their decay rates under varied temperature conditions. Three types of purple mudstone were sampled from the Tuodian group (J 3 t), Matoushan group (K 2 m) and Lufeng group (J 1 l), located in the Chuxiong district of Yunnan province, south-western China. All samples were manually cut into cubes of 50 × 50 × 50 mm, before being subjected to 39 wetting–drying cycles in a 2 mm sieve under artificial constant temperature, natural daily temperature variation and artificial temperature alternation (minimum and maximum) treatments in the laboratory. Decay rates were calculated by weighing the mass remaining in the sieve after each treatment cycle. The results showed that the decay rates of the tested rocks rose with an increase in temperature and temperature difference (maximum–minimum), with the rank order of decay rate being J 3 t > K 2 m > J 1 l. Quantitative analysis revealed rock decay rate to be significantly related to temperature by a power function, and exponentially related to temperature difference. Our results suggested that temperature difference has a stronger effect on purple mudstone decay than that of temperature magnitude during the wetting–drying processes. [ABSTRACT FROM AUTHOR]

River confluences are key nodes in fluvial systems. Past studies on large river confluences were focused on the Paranà, Amazon, and Yangtze rivers, while the Yarlung Zangbo River (YZR), the longest plateau river in China, has not been investigated yet. This paper investigates a large multi-channel confluence between the YZR and one of its major tributaries, the anabranching Niyang River (NR). A remote-sensing study of this area demonstrated that the confluence planform is very stable compared with its upstream and downstream reaches that suffered from sandbar migration, meandering, and the development of a head-cut. A field survey using Acoustic Current Doppler Profiling (ADCP) was conducted at the confluence in June 2019, when the YZR and NR had discharges of about 1393 m3/s and 1299 m3/s in total, respectively. Although the discharge ratio of this confluence attained a value of 0.93, the existence of an estuary delta with a large area of floodplain largely weakened the impact of the NR on the dynamics of the YZR. The inflows of the multiple channels of the NR separated by the delta created an evident and long mixing interface and produced the left-side secondary cells in the YZR. In contrast, the curved riverbank played a more important role in the formation of deep scour holes, generating heterogeneity of the velocity distribution and the fast-mixing of flows. This study leads to an advancement in our current knowledge about multi-channel confluences. • Remote sensing images show that the plateau river confluence planform is very stable. • A field survey on flow and morphological features of the confluence was conducted. • The estuary delta with large floodplain largely weakened the confluence dynamics. • The curved river bank played an important role in river bed scours and flow mixing. [ABSTRACT FROM AUTHOR]

The Yellow River (YR) flows from Northwest China to North China, crossing three topographic steps: the Tibetan Plateau, the Chinese Loess Plateau (CLP), and the North China Plain (NCP). Determining the river's geomorphological evolution is significant for understanding the formation of macro-geomorphic patterns in China. However, the formation age of YR channelization into the Bohai Sea is highly controversial and uncertain, with estimates ranging from the Late Miocene to the Late Pleistocene. This study conducted grain size analysis, whole-rock geochemistry, and detrital zircon U-Pb dating on ~5.2 Ma sediments from borehole G4 drilled in the Bohai Bay. The geochemical results from borehole G4 indicate that chemical weathering was weak during 5.2–1.6 Ma, but showed diverse characteristics since ~1.6 Ma. Moreover, concordant detrital zircons from borehole G4 (~1.6 Ma) yield a major age population of 350–550 Ma, which, however, is scarcely produced by detrital zircons from the nearby Yanshan and Taihang Mountains. Geochemical signals and detrital zircon U-Pb geochronology suggest that the YR reached the NCP after cutting through the Sanmen Gorge and entered the Bohai Bay since ~1.6 Ma. The massive presence of sediments with an age population of 650–1350 Ma in the top of borehole G4 demonstrates that the present topography of the middle and upper reaches of the YR initially formed at ~0.6 Ma. Subsequently, changes in provenance since ~1.6 Ma influenced the records of paleoclimatic signals indicated by elemental geochemistry in the Bohai Bay. Because of detrital inputs from the YR, the (La/Sm) N and (La/Yb) N ratios of borehole G4 abruptly increased after ~1.6 Ma. Other geochemical indicators exhibited no noticeable variation trend, which differs from the decreasing trend of deep-sea δ18O records, the Chemical Index of Alteration (CIA), and the 87Sr/86Sr ratio of the loess from the Chinese Loess Plateau after the YR entered the NCP at ~1.6 Ma. This paper proposes that the YR gradually triggered fluvial erosion up to its upper reaches after entering the NCP, which is primarily in response to climate change. • Detrital zircon U-Pb geochronology and elemental geochemistry of borehole sediments in Bohai Bay indicate the geomorphological evolution of Yellow River • Yellow River cut through the Sanmen Gorge and reached the North China Plain at ~1.6 Ma • The current geomorphic of the middle and upper reaches of Yellow River was initially formed ~0.6 Ma • Changes in provenance since ~1.6 Ma have influenced the records of paleoclimatic signals in Bohai Bay [ABSTRACT FROM AUTHOR]

Debris flow is a major geohazard in mountainous regions and poses a significant threat to life and property. The damage caused by debris flows have increased with the expansion of human settlements and activity into the mountainous regions of China. In regards to risks from debris flows, previously unrecognized low-frequency debris flow catchments constitute an especially significant threat. According to our investigation, only about 500 catchments have debris flow records in >2000 catchments of Bailong River basin. The main purpose of this paper is to introduce a new methodology using Artificial Intelligence (AI) that can simultaneously input parameters related to geomorphological conditions and material conditions to better distinguish low-frequency debris flow catchments (LFD s) from medium-high frequency debris flow catchments (MHFD s). A total of 449 prototypical debris flow catchments, 15 parameters, and 9 commonly used learning machines were used to build identification models. Debris flow catchments are divided into 4 cases (LO1-LO4) based on different sample ratios of LFD s and MHFD s, which are input into each classifier one by one. Based on model evaluation, the CHAID model in the case LO2 performs best, which only uses five parameters (formation lithology index, land use index, vegetation coverage index, drainage density and landslide density index) to predict LFD s. The results indicate that LFD s are mainly distributed in areas with less landslide distribution and better vegetation coverage compared with MHFD s. However, the distribution of LFD s is concentrated on FLI (formation lithology index) =4, which is the weak lithology area. The tree classifier seems to be better at classifying fluvial processes. The model developed in this paper can help us quickly find LFD s in similar areas, and help to assess the risk of debris flows. • A simple decision tree model was built to identify low-frequency debris flows. • Geomorphological and material parameters were used simultaneously for modeling. • Materials seem to be the main influencing factor of low-frequency debris flows. • The tree classifier seems to be better at classifying fluvial processes. [ABSTRACT FROM AUTHOR]

Mons Rümker is a preferred candidate landing region for China's Chang'e-5 (CE-5) mission, from where it is of great significance to select safe landing areas. Lunar terrain factors from Digital Elevation Model (DEM) data, limited by their low resolutions (~10 m/pixel), are inapplicable to evaluating the lunar landing area safety, in spite of the fact that lunar remote sensing imagery has higher resolution (~50 cm/pixel). In this paper, we extracted terrain factors in divided square girds by the aid of the high-resolution Lunar Reconnaissance Orbiter (LRO) Narrow-angle Camera (NAC) images, namely, flat area percentage (Fap), distribution pattern of uneven objects (NNI) based on the double-threshold Otsu method, and roughness based on gray level histogram analysis. Mons Rümker can be divided into four geological units, named as LD, B1, B2, and B3, respectively. Unit B1 has a higher roughness and a lower Fap. Unit B2 and B3 are characterized with the highest Fap and the lowest roughness. NNIs of Unit B1, B2, and B3 are >1 while LD's is <1. Thus, the distribution patterns of uneven objects in Unit LD are clustered but dispersed in all Unit B1, B2, and B3. This paper tends to take Fap and roughness as the main terrain factors to evaluate the safety for CE-5 landing in Mons Rümker with NNI being a supplement to Fap. According to Standard 1–3 mentioned in this paper, we would classify the divided square grids of the Mons Rümker region as the safe or unsafe areas, and then discriminate five potential landing areas for CE-5 probe safe landing. Five potential landing areas (red polygons marked with number 1–5). The blue, purple and red colors are the safe grids classified in line with Standard 1–3, separately. The white solid line indicates the boundary of Mons Rümker. The yellow stars show two candidate landing sites for CE-5 mission mentioned by Zhao et al. (2017). Unlabelled Image • Three terrain factors (Fap, NNI and roughness) were derived from the NAC mosaic image of Mons Rümker; • Mons Rümker can be divided into four geological units, which are named as LD, B1, B2, and B3; • Five potential landing areas for CE-5 probe safe landing were recognized based on Standard 1-3. [ABSTRACT FROM AUTHOR]

Coastal morphodynamics and environmental changes have been obviously altered by global warming, sea-level fluctuations, and human activities. Palynological analysis is one of the most important tools for providing information on vegetation, climate, and environmental change, which can be relatively sensitive indicators of human activity during the Holocene. However, Holocene coastal morphodynamics and palynological evidence from the Pearl River Delta and coastal areas has not been sufficiently studied. In this paper, we carry out analyses using detailed grain-size and palynological records of pollen, spores, freshwater algae, and marine dinoflagellate cysts from Borehole QZK13 in the Pearl River Estuary (PRE), southern China. The results show information relating coastal morphodynamics, vegetation evolution, climate changes, and anthropogenic activities during the Holocene based on radiocarbon chronology. In general, the PRE experienced marine transgressive processes; sea-level and the coastline tended to migrate landward between 10,700 and 4250 cal yr BP. Meanwhile, abundant arboreal pollen (mainly Castanopsis , Quercus -evergreen, Oleaceae, and Castanea) and small amounts of non-arboreal pollen (mainly Poaceae, Cyperaceae, and Artemisia) were observed in Borehole QZK13, which suggests that subtropical evergreen broadleaved forests were distributed throughout the PRE and adjacent areas, and relatively warm and moist climatic conditions were present during the early-middle Holocene. After that, there were relatively cool and wet climatic conditions, and marine regressive processes occurred in the PRE during the Late Holocene (ca. 4250–960 cal yr BP) indicated by an abrupt increase in aquatic herbs and Dicranopteris. Moreover, we hypothesize that the Neolithic "cultural fault" in the PRE is likely related to frequent storm surges induced by extreme typhoon events that occurred ca. 3000–2700 yr BP. • Grain size, pollen, spores, freshwater algae and dinoflagellate cysts from the Pearl River Estuary were investigated. • Three major phases shown by palynological results were corresponding to the changes in vegetation and climate. • We hypothesized that the Neolithic "cultural fault" is probably related to the storm surges induced by extreme typhoon events. • Natural vegetation cover has suffered large-scale destruction by ancient humans during the Late Holocene. [ABSTRACT FROM AUTHOR]

Seasonal or interannual precipitation differences lead to changes in the water level of a tufa lake, while the underwater topography affects its local depth. Therefore, topo-bathymetry is important to study and protect the aquatic environment of tufa lakes. However, traditional field-based topo-bathymetric surveying methods (e.g., sounding rod or sonar) inevitably disturb the fragile lake ecosystem. In recent years, the emerging remote sensing technology of unmanned aerial vehicles (UAV) has provided a cost-effective solution for measuring topo-bathymetry without disturbance. In this paper, taking Spark Lake in Jiuzhaigou, China, as an example, we captured red-green-blue (RGB) images using a fixed-wing UAV and produced a digital elevation model (DEM) prior to the Jiuzhaigou Earthquake using Structure-from-Motion (SfM) photogrammetry. The underwater topography of Spark Lake was obtained by refraction correction and water color inversion based on the DEM and orthophoto, respectively. For refraction correction, a water depth correction model based on Snell's Law was used. For water color inversion, general band ratio models were replaced by a band difference model (blue band - green band). The qualities of the resulting DEMs produced by the two methods were evaluated against the topography of the drained Spark Lake after the earthquake, and the corresponding DEMs of difference (DoD) were also analyzed. The coefficient of determination (R2) and root mean square error (RMSE) are 0.88 and 1.32 m for refraction correction, and 0.86 and 1.37 m for water color inversion, respectively. The results demonstrated the feasibility and effectiveness of applying UAV-acquired RGB imagery and the two optical remote sensing methods to topo-bathymetric mapping of transparent tufa lakes. • UAV-acquired RGB images were applicable for testifying topo-bathymetry of tufa lake. • Refraction correction demonstrated a good performance for water depths up to 12 m. • The difference between blue and green bands was the optimal spectral combination. • Water color inversion exhibited a fine performance across the entire lake depth range. [ABSTRACT FROM AUTHOR]

This paper presents an approach to improve the previously proposed 2-D topography modeling (Ding et al., 2019) that is primarily based on inversion of low-temperature thermochronological data. The new approach requires thermochronological data from both horizontal transect and vertical profiles to yield a more comprehensive modeling of topographic evolution. This modeling allows for (1) reconstruction of paleotopography evolution and (2) calculation of spatial distribution of paleo-geothermal gradients over time; therefore, it is possible to deduce more reliable exhumation histories even for non-steady state topographic evolution. The deduced exhumation histories are used to re-evaluate the topography modeling results. This approach when applied to the Dabie orogen in Eastern China suggests an uneven topographic evolution along the Xishui-Lu'an section and comparable exhumation histories for the Meichuan, the Tianzhushan, the Mingtangshan and the Tiantangzhai vertical profiles since the Late Cretaceous. The modeling results suggest that the location of the highest elevation in the NE-SW transect has been almost identical for the last 80 Ma although the elevation of the peak has reduced by ~3.3 km over the same period. The elevation reduction in the NE segment during 70–40 Ma is caused by extensional activity of the Xiaotian-Mozitan fault. The elevation reduction in the SW segment during 50–30 Ma is probably related to climate change, where the Paleogene topography change in the Dabie orogen played an important role introducing humid and warm air from the Pacific Ocean. • The Dabie orogen forms a tectonic and climatic boundary in Eastern China. • Thermochronological data collected along horizontal transects and vertical profiles • Transect/profile thermochronology results used for topography modeling • Rapid exhumation from 70 to 40 Ma along NE 2-D transect segment linked to faulting • Rapid exhumation from 50 to 30 Ma along SE 2-D transect segment linked to climate change [ABSTRACT FROM AUTHOR]

Revised wind erosion equation (RWEQ) is widely applied to determining wind erosion and achieves reliable results. However, the spatial variability of wind erosivity and wind erosion has not been sufficiently investigated in previous studies. In this paper, RWEQ was used to determine wind erosion and its spatial variation at the landscape scale in the Xilingele grassland, which is a typical temperate grassland in North China suffering from increasing wind erosion problems. Based on field investigation and laboratory measurement, land-surface parameter data were obtained. Computational fluid dynamics (CFD) wind modelling was integrated to acquire the spatial variation of wind data, which considers the influence of topography and land use. These data were used to determine the five factors of RWEQ and wind erosion modulus. Results showed that the mean wind erosion modulus was 5.33 t ha−1 yr−1, and 97.16% of the area had light or no wind erosion, which was mostly distributed in the low plains and the leeward areas. However, there were hotspots of wind erosion which were mainly distributed in the windward slopes and the summit areas. There were no significant differences among the wind erosion moduli of different land uses, but land use affected the influence of topography on wind erosion. The maximum modulus varied greatly from 0 to 259.41 t ha−1 yr−1 for different land uses, and the most severe wind erosion occurred in moderately grazed grassland (MG), followed by heavily grazed grassland (HG), arable land (AL), lightly grazed grassland (LG) and ungrazed grassland (UG). Statistical tests indicated that topography had a significant influence on wind erosion, which was not detected in previous studies due to the lack of high-resolution topography and wind data. The combination of RWEQ and CFD wind modelling suggests that future wind erosion studies should further consider the important influence of topography. To achieve sustainable development in temperate grassland, rational grazing is critical, and special attention should be paid to windward slopes and high elevation areas. • RWEQ was used to determine wind erosion and its spatial variation at landscape scale. • CFD wind modelling was employed to acquire spatial variation of the wind erosivity. • Wind erosion in the temperate grassland is generally weak but with local hotspots. • Topography and land use are the major factors affecting landscape scale wind erosion. • CFD is a proper solution to model the influence of topography which was overlooked. [ABSTRACT FROM AUTHOR]

In ancient China, the course of rivers changed frequently under the influence of the natural environment. The study of changes in paleochannels provides the basis for river regulation and water conservancy, and also supplies an important scientific foundation for the evolution of the geographical environment. However, it is extremely difficult to determine the location of paleochannels due to the limitations of traditional research methods and modern remote sensing interpretation. This paper introduces a new method, Detrended Digital Elevation Model Interpretation, for supporting interpretation of ancient rivers at large scales. A detrended DEM was obtained by removing the overall trend from the original DEM by subtracting a "detrending surface" derived from the original image using a guiding filter. The paleochannels of the Yellow River in the North China Plain were interpreted by GIS analysis with the help of historical materials. The detrended DEM was found to be less affected by the surface environment and had higher identification accuracy compared with an optical image and the colored DEM. Moreover, the detrended DEM could be used to interpret ancient rivers at large scales. Finally, the flow path of ancient rivers in the Eastern Han Dynasty (or near the Jindi River) was hypothesized using the interpretation results. The resulting paleochannel data can be used to revise historical river maps and provide reference for researching and managing the Yellow River. • Detrended Digital Elevation Model Interpretation is a new method for supporting interpretation of paleochannels in a large scale. • Detrending solves the ambiguity of feature details at low altitude of the original DEM by removing the down-valley slope of DEM. • The detrended DEM retains the topographic structure and shows the details of objects without being affected by the surface spectrum and occlusion. • The paleochannel of Yellow River in the Eastern Han Dynasty in Yanggu area is supposed to be located in the north side of Jindi river. [ABSTRACT FROM AUTHOR]

Loess caves, landslides, collapses, bank erosion and soil erosion, which are independently and interactively shaping the modern loess landscape, are widely distributed across the Chinese Loess Plateau. The study area is located in a small watershed in Huining County, Gansu Province, which is one of the areas with the strongest development of loess caves across the Loess Plateau. This area is an ideal site for studying geological hazards and geomorphic evolution on the Loess Plateau. Seventeen landslides and 176 loess caves with a density of 887 units/km2 were investigated with a UAV (unmanned aerial vehicle) and mapped by GIS (geographic information system). Using high-resolution UAV images and topographic data, we carried out an interpretation of loess caves, bank erosion and landslides; studied the spatial distribution and developmental patterns of loess caves, landform morphology and the relationship between loess caves and landslides based on GIS spatial analysis, mathematical statistics and field surveys; and analyzed the influence of loess microtopography on slope landforms. Finally, six typical evolutionary models of soil erosion-loess cave development-landslide occurrence-barrier dam formation are preliminarily proposed: the primary slope stage, early cave stage, accelerated cave stage, cave connecting stage, landslide creeping stage, and landslide-dam forming stage. This study is a useful exploration and an attempt to reveal the influence of loess cave development on slope stability and disaster chain effects. • This paper completed the high-resolution interpretation of loess geological hazards based on UAV. • Spatial distribution, developmental regularity and influences of loess caves were revealed. • Typical evolution model of the loess disaster chain was preliminarily proposed. [ABSTRACT FROM AUTHOR]

Weathering is one of the main preparatory causal factors for rock-avalanche events and weathering assessment of involved materials can refine understanding of geomorphic processes of this class. The weathering degree of the deposit of the Touzhai basalt rock avalanche in China, is quantitatively evaluated herein through an integrated geochemical analysis, coupled with meso- and micro-scopic examination of fragments, which can help to understand the inhomogeneity of weathering in rock slopes. The clasts of 12 size classes of the debris making up the deposit, ranging from meter-sized boulder to fines of <25 μm, are assessed with 16 geochemical parameters and eight weathering indices. The results show that the debris can be subdivided into two distinct groups: fresh (or fresh-slightly weathered) and weathered (moderately-highly weathered). Group 1 includes five classes of several meters, 15-10 cm, 5-4 cm, 2-1 cm and 1-0.5 cm, and Group 2 covers a particle size range of <0.5 cm to <25 μm. Of special interest is that fresh debris accounts for at least ~80 wt.% of the rock-avalanche deposit. The unexpectedly significant proportion of fresh debris appears to be a necessary prerequisite for the rock-avalanche occurrence and can be attributed to the aphanitic texture and low permeability of the basalt. On the basis of the low bulk density and strength of saprolite (completely weathered rocks), the slopes containing ~ 80-90% of unweathered materials may fail as rock avalanche-debris flows. They may collapse as rockslide-debris flows if saprolite crusts continue to thicken. In addition, the slopes made entirely, of saprolite, would not host catastrophic failures. • The weathering of a rock-avalanche deposit in China is quantitatively assessed. • The inhomogeneity of weathering of rock masses is taken into account in this paper. • Surprisingly high proportion of fresh debris is found in the loose earthy deposit. • The weathered states of rock masses can control the style of slope failures. [ABSTRACT FROM AUTHOR]

Landslides play an important role in landscape evolution of the Loess Plateau, northern China, but they have not been systematically analysed across the region. The contribution to regional rates of erosion and sediment budgets is poorly known as a consequence. Co-seismic landslides are a significant natural hazard, and exacerbate the threat posed by active faults in the region, but this aspect of the landslides is also poorly quantified. This paper presents inventories for ~80,000 landslides in the Loess Plateau, and analyses the distributions and extents. Four study areas focus on the epicentres of major (M > 7) historical earthquakes, from 1303 CE (Hongdong), 1556 (Huaxian), 1718 (Tongwei) and 1920 (Haiyuan), which are likely to have been major triggers for landslides in these regions. A fifth study area focuses on landslides in the aftermath of localised storms in 2010 in the Tianshui region, to allow comparison of landslides clearly triggered by rainfall with landslides in the regional inventories. Landslides predominantly resulted from failure of the regional loess blanket that gives the region its name, or the soil profile developed above it. Rainfall-generated landslides are dominantly surficial soil landslides. Landslides preserved in the four epicentral areas are larger and deeper, and resemble global bedrock landslides in the dimensions. Total volume of landslide in each of the four epicentral areas are on the order of 3–7 km3; these volumes are likely to be the cumulative totals of repeated earthquakes in the same regions, with each event reactivating landslide scarps on hillsides. Comparison of landslide distributions and modern settlement patterns suggests that future M > 7 earthquakes striking the Loess Plateau may trigger landslides that could directly hit 30–50% of the communities. • Analysis of 80,000 landslides in the Loess Plateau. • Storms generate surficial soil landslides, easy to recover; historical earthquakes trigger both soil landslides and larger, deeper ones, only the latter can be interpreted. • Future M > 7 earthquakes may trigger landslides that could directly hit 30–50% of the communities in the epicentral area. [ABSTRACT FROM AUTHOR]

This paper examines the relationship between the extent of permafrost during the Last Permafrost Maximum (LPM, largely between 22 and 17 ka BP) and the altitudinal and latitudinal temperature gradients that controlled its extent. The region chosen, namely, the Qinghai-Tibet Plateau and the transition to the northern adjacent plains in the North China Plain, is of relatively limited extent in order to avoid regional effects. This area extends over 15 degrees of latitude while altitudes vary from 1000 to 1500 m above sealevel (asl) in the northern plains to more than 4000 m asl in the central-to-south part of the Qinghai-Tibet Plateau. The respective lowest southernmost permafrost occurrences at a northern and a southern position are used for reconstruction of the latitudinal and altitudinal gradients. As a first approximation, we conclude that the elevation compensation during the LPM for a 1° latitude change was between 309 m and 385 m. This is considerably higher than the value calculated for existing permafrost in the region that uses thermal data at a depth of zero amplitude change (139–198 m altitudinal compensation for 1° latitude). [ABSTRACT FROM AUTHOR]