Comparative study of different wavelets for developing parsimonious Volterra model for rainfall-runoff simulation

Although the Volterra models are non-parsimonious ones, they are being used because they can mimic dynamics of complex systems. However, applying and identification of the Volterra models using data may result in overfitting problem and uncertainty. In this investigation we evaluate capability of different wavelet forms for decomposing and compressing the Volterra kernels in order to overcome this problem by reducing the number of the model coefficients to be estimated and generating smooth kernels. A simulation study on a rainfall?runoff process over the Cache River watershed showed that the method performance is successful due to multi-resolution capacity of the wavelet analysis and high capability of the Volterra model. The results also revealed that db2 and sym2 wavelets have the same high potential in improving the linear Volterra model performance. However, QS wavelet was more successful in yielding smooth kernels. Moreover, the probability of overfitting while identifying the nonlinear Volterra model may be less than the linear model.     

Automatic mapping of lunar landforms using DEM-derived geomorphometric parameters

Developing approaches to automate the analysis of the massive amounts of data sent back from the Moon will generate significant benefits for the field of lunar geomorphology. In this paper, we outline an automated method for mapping lunar landforms that is based on digital terrain analysis. An iterative self-organizing (ISO) cluster unsupervised classification enables the automatic mapping of landforms via a series of input raster bands that utilize six geomorphometric parameters. These parameters divide landforms into a number of spatially extended, topographically homogeneous segments that exhibit similar terrain attributes and neighborhood properties. To illustrate the applicability of our approach, we apply it to three representative test sites on the Moon, automatically presenting our results as a thematic landform map. We also quantitatively evaluated this approach using a series of confusion matrices, achieving overall accuracies as high as 83.34% and Kappa coefficients (K) as high as 0.77. An immediate version of our algorithm can also be applied for automatically mapping large-scale lunar landforms and for the quantitative comparison of lunar surface morphologies.     

利用大地电磁资料对玄武岩成像的可行性

The magnetotelluric （MT） method has been among the favorite supporting tools for seismic imaging of sub-salt and sub-basalt targets. In this paper we present an example from Kachchh, India （where basaltic rocks overlie Mesozoic sedimentary rocks）, and discuss the feasibility of using MT method as an exploration tool in this geological setting. Our results highlight the difference in magnetotelluric response caused by the thin intrabasalt layering. The key issue addressed in this paper is what MT can and cannot provide in such geological settings. First, we compute apparent resistivity and phase response curves using representative resistivity-depth models and borehole data from the study area. Later, we compare these results to assess the plausibility of using MT to image the sub-volcanic sediments at Kachchh. Finally, we substantiate our discussion through one-dimensional inversion of the field observed MT data from this region that exhibits poor sensitivity of MT for thin basalt layers.     

Identification of buried sinkholes using refraction tomography at Ft. Campbell Army Airfield,Kentucky

Karst aquifers are highly susceptible to contamination, with numerous points of entry for contaminants through recharge features such as sinkholes, swallow holes and solutionally enlarged fractures. These recharge features may be filled or obscured at the surface, requiring the use of geophysical or remote sensing techniques for their identification. This study uses seismic refraction data collected at the Ft. Campbell Army Airfield (CAAF), Kentucky, USA, to test the hypothesis that refraction tomography is a useful tool for imaging bedrock depressions beneath thick overburden (greater than 20 m of unconsolidated sediment). Southeast of the main taxiway of CAAF seismic velocity tomograms imaged a bedrock low, possibly a closed depression, at a depth of 25 m that had been earlier identified through delay-time analysis of the same refraction data. Tomography suggests the bedrock low is about 250-m wide by 10-m deep at its widest point. High rates of contaminant vapor extraction over the western extension of this feature suggest a high concentration of contaminants above, and within, this filled bedrock low, the base of which may contain solutionally enlarged fractures (i.e. karst conduits) that could funnel these contaminants to the upper or lower bedrock aquifers. This study thus demonstrates the viability of seismic refraction tomography as a tool for identification of filled sinkholes and bedrock depressions in karst areas.     

A Machine Learning-based Cloud Detection Algorithm for the Himawari-8 Spectral Image

Cloud Masking is one of the most essential products for satellite remote sensing and downstream applications. This study develops machine learning-based (ML-based) cloud detection algorithms using spectral observations for the Advanced Himawari Imager (AHI) onboard the Himawari-8 geostationary satellite. Collocated active observations from Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) are used to provide reference labels for model development and validation. We introduce both daytime and nighttime algorithms that differ according to whether solar band observations are included, and the artificial neural network (ANN) and random forest (RF) techniques are adopted for comparison. To eliminate the influences of surface conditions on cloud detection, we introduce three models with different treatments of the surface. Instead of developing independent ML-based algorithms, we add surface variables in a binary way that enhances the ML-based algorithm accuracy by ~5%. Validated against CALIOP observations, we find that our daytime RF-based algorithm outperforms the AHI operational algorithm by improving the accuracy of cloudy pixel detection by ~5%, while at the same time, reducing misjudgment by ~3%. The nighttime model with only infrared observations is also slightly better than the AHI operational product but may tend to overestimate cloudy pixels. Overall, our ML-based algorithms can serve as a reliable method to provide cloud mask results for both daytime and nighttime AHI observations. We furthermore suggest treating the surface with a set of independent variables for future ML-based algorithm development.     

Comparison of spectral and spatial methods for a Moho recovery from gravity and vertical gravity-gradient data

In global studies investigating the Earth’s lithospheric structure, the spectral expressions for the gravimetric forward and inverse modeling of the global gravitational and crustal structure models are preferably used, because of their numerical efficiency. In regional studies, the applied numerical schemes typically utilize the expressions in spatial form. Since the gravity-gradient observations have a more localized support than the gravity measurements, the gravity-gradient data (such as products from the Gravity field and steady-state Ocean Circulation Explorer - GOCE - gravity-gradiometry satellite mission) could preferably be used in regional studies, because of reducing significantly the spatial data-coverage required for a regional inversion or interpretation. In this study, we investigate this aspect in context of a regional Moho recovery. In particular, we compare the numerical performance of solving the Vening Meinesz-Moritz’s (VMM) inverse problem of isostasy in spectral and spatial domains from the gravity and (vertical) gravity-gradient data. We demonstrate that the VMM spectral solutions from the gravity and gravity-gradient data are (almost) the same, while the VMM spatial solutions differ from the corresponding spectral solutions, especially when using the gravity-gradient data. The validation of the VMM solutions, however, reveals that the VMM spatial solution from the gravity-gradient data has a slightly better agreement with seismic models. A more detailed numerical analysis shows that the VMM spatial solution formulated for the gravity gradient is very sensitive to horizontal spatial variations of the vertical gravity gradient, especially in vicinity of the computation point. Consequently, this solution provides better results in regions with a relatively well-known crustal structure, while suppressing errors caused by crustal model uncertainties from distant zones. Based on these findings we argue that the gravity-gradient data are more suitable than the gravity data for a regional Moho recovery.     

Reconstruction of paleocoastlines for the northwestern South China Sea since the Last Glacial Maximum

The range of relative sea level rise in the northwestern South China Sea since the Last Glacial Maximum was over 100 m. As a result, lowland regions including the Northeast Vietnam coast, Beibu Gulf, and South China coast experienced an evolution from land to sea. Based on the principle of reconstructing paleogeography and using recent digital elevation model, relative sea level curves, and sediment accumulation data, this paper presents a series of paleogeographic scenarios back to 20 cal. ka BP for the northwestern South China Sea. The scenarios demonstrate the entire process of coastline changes for the area of interest. During the late glacial period from 20 to 15 cal. ka BP, coastline slowly retreated, causing a land loss of only 1×10⁴ km², and thus the land-sea distribution remained nearly unchanged. Later in 15–10 cal. ka BP coastline rapidly retreated and area of land loss was up to 24×10⁴km², causing lowlands around Northeast Vietnam and South China soon to be underwater. Coastline retreat continued quite rapidly during the early Holocene. From 10 to 6 cal. ka BP land area had decreased by 9×10⁴km², and during that process the Qiongzhou Strait completely opened up. Since the mid Holocene, main controls on coastline change are from vertical crustal movements and sedimentation. Transgression was surpassed by regression, resulting in a land accretion of about 10×10⁴km². Supported by Key Laboratory of Marginal Sea Geology, Chinese Academy of Sciences (Grant No. MSGL0711), the Guangdong Natural Science Foundation (Grant No. 04001309) and Open Fund of the Key Laboratory of Marine Geology and Environment, Chinese Academy of Sciences (Grant No. MGE2007KG04)     

Assessment of shallow landslides from Hurricane Mitch in central America using a physically based model

Shallow landslides induced by heavy rainfall events represent one of the most disastrous hazards in mountainous regions because of their high frequency and rapid mobility. Recent advancements in the availability and accessibility of remote sensing data, including topography, land cover and precipitation products, allow landslide hazard assessment to be considered at larger spatial scales. A theoretical framework for a landslide forecasting system was prototyped in this study using several remotely sensed and surface parameters. The applied physical model SLope-Infiltration-Distributed Equilibrium (SLIDE) takes into account some simplified hypotheses on water infiltration and defines a direct relation between factor of safety and the rainfall depth on an infinite slope. This prototype model is applied to a case study in Honduras during Hurricane Mitch in 1998. Two study areas were selected where a high density of shallow landslides occurred, covering approximately 1,200 km². The results were quantitatively evaluated using landslide inventory data compiled by the United States Geological Survey (USGS) following Hurricane Mitch’s landfall. The agreement between the SLIDE modeling results and landslide observations demonstrates good predictive skill and suggests that this framework could serve as a potential tool for the future early landslide warning systems. Results show that within the two study areas, the values of rates of successful estimation of slope failure locations reached as high as 78 and 75%, while the error indices were 35 and 49%. Despite positive model performance, the SLIDE model is limited by several assumptions including using general parameter calibration rather than in situ tests and neglecting geologic information. Advantages and limitations of this physically based model are discussed with respect to future applications of landslide assessment and prediction over large scales.     

A coupling of mixed and continuous Galerkin finite element methods for poroelasticity I: the continuous in time case

In this paper, we formulate a finite element procedure for approximating the coupled fluid and mechanics in Biot’s consolidation model of poroelasticity. Here, we approximate the pressure by a mixed finite element method and the displacements by a Galerkin method. Theoretical convergence error estimates are derived in a continuous in-time setting for a strictly positive constrained specific storage coefficient. Of particular interest is the case when the lowest-order Raviart–Thomas approximating space or cell-centered finite differences are used in the mixed formulation, and continuous piecewise linear approximations are used for displacements. This approach appears to be the one most frequently applied to existing reservoir engineering simulators.     

Stabilization of Expansive Clays Using Granulated Blast Furnace Slag (GBFS) and GBFS-Cement

Expansive clays undergo swelling when subjected to water. This can cause damage, especially to light weight structures, water conveyance canals, lined reservoirs, highways, and airport runways unless appropriate measures are taken. In this study, granulated blast furnace slag (GBFS) and GBFS-cement (GBFSC) were utilized to overcome or to limit the expansion of an artificially prepared expansive soil sample (sample A). GBFS and GBFSC were added to sample A in proportions of 5–25% by weight. The effects of these stabilizers on grain size distribution, Atterberg limits, swelling percentage and rate of swell of soil samples were determined. GBFS and GBFSC were shown to successfully decreasing the total amount of swell while increasing the rate of swell.