Modeling typhoon- and earthquake-induced landslides in a mountainous watershed using logistic regression

Landslides can be caused by storms and earthquakes. Most logistic regression models proposed in recent years have been targeted at rainfall-induced landslides. In areas such as Taiwan, where landslides can be triggered by typhoons (tropical cyclones) and earthquakes, a rainfall-induced model is insufficient because it provides only a partial explanation of landslide occurrence and overlooks the potential effect of earthquakes on typhoon-triggered landslides. This study used landslides triggered by a major earthquake and a typhoon prior to the earthquake to develop an earthquake-induced model and a typhoon-induced model. The models were then validated by using landslides triggered by three typhoons after the earthquake. According to the results, typhoon-triggered landslides tended to be near stream channels and earthquake-triggered landslides were more likely to be near ridge lines. Moreover, a major earthquake could still affect the locations of typhoon-triggered landslides 6 years after the earthquake. This study therefore demonstrates that an earthquake-induced model both sheds light on the environmental factors for triggering landslides, and augments a rainfall-induced model in its predictive capability in areas such as Taiwan.     

An advanced analytical model for high damping rubber bearings

Base‐isolation technologies have been developed over the years in attempts to mitigate the effects of earthquakes on structures and potentially vulnerable contents in earthquake prone areas of the world. The high damping rubber bearing (HDRB) is a relatively recent and evolving technology of this kind. The isolator shifts the fundamental period of the base‐isolated structure to a value beyond the range of the plentiful energy‐containing periods of earthquake motions and supplies significant damping to dissipate energy caused by motions. Nevertheless, the highly non‐linear mechanical behaviour of the HDRB is so complex, especially at large strains, that it is difficult to model it analytically. In this paper, an extensive study of experimental tests for identifying the mechanical characteristics of the HDRB is presented. By modifying the Wen's model to include the rate‐dependent effects, an advanced analytical model in an incremental form for the HDRB is also proposed. A very good agreement between the analytical and experimental results has been obtained. It is illustrated that the proposed mathematical model can predict well the mechanical behaviour of HDRB bearings, even at large shear strain. Copyright © 2003 John Wiley & Sons, Ltd.     

Impairment of sea urchin sperm quality by UV-B radiation: predicting fertilization success from sperm motility

Sperm quality of the sea urchin, Anthocidaris crassispina, after exposure to environmentally realistic UV-B irradiances, was assessed by changes in sperm motility (measured by the computer-assisted sperm analysis (CASA) system), and related to subsequent fertilization success. Percentage motile sperm of A. crassispina declined significantly after exposure to a UV-B dose of 16.2 kJ m(-2), while sperm motion velocity as measured by curvilinear velocity (VCL), straight line velocity (VSL), and average path velocity (VAP) showed significant reduction after exposure to a UV-B dose of 5.4 kJ m(-2). A parallel study showed that fertilization success was significantly reduced after sperm were exposed to UV-B doses > or = 5.4 kJ m(-2). Notably, the four sperm motility parameters were strongly correlated with fertilization success (P < 0.001), followed the increasing order: VSL (r = 0.8) < % motile sperm (r = 0.804) < VCL (r = 0.912) < VAP (r = 0.928). Fertilization success is best predicted by VAP using the exponential model: y = 8.678 + 90.202/[1 + exp(82.83 - x)/10.27)] (r(2) = 0.95). Thus, impairment of sperm motility of sea urchin, as measured by the CASA method, can be used to predict reproductive success and ecological effects.     

An intelligent and autonomous MEMS IMU/GPS integration scheme for low cost land navigation applications

An intelligent scheme to integrate inertial navigation system/global positioning system (GPS) is proposed using a constructive neural network (CNN) to overcome the limitations of current schemes, namely Kalman filtering (KF). The proposed CNN technique does not require prior knowledge or empirical trials to implement the proposed architecture since it is able to construct its architecture “on the fly,” based on the complexity of the vehicle dynamic variations. The proposed scheme is implemented and tested using Micro-electro-mechanical systems inertial measurement unit data collected in a land-vehicle environment. The performance of the proposed scheme is then compared with the multi-layer feed-forward neural networks (MFNN) and KF- based schemes in terms of positioning accuracy during GPS signal outages. The results are then analyzed and discussed in terms of positioning accuracy and learning time. The preliminary results presented in this article indicate that the positioning accuracy were improved by more than 55% when the MFNN and CNN-based schemes were implemented. In addition, the proposed CNN was able to construct the topology by itself autonomously on the fly and achieve similar prediction performance with less hidden neurons compared to MFNN-based schemes.     

Variation of initial Th/Th and limits of high precision U-Th dating of shallow-water corals

One hundred eighty U-Th data, including 23 isochrons on 24 pristine modern and Holocene corals and 33 seawater samples, were analyzed using sector-field mass spectrometry to understand the variability of initial ²³⁰Th/²³²Th (²³⁰Th/²³²Th₀). This dataset allows us to further assess the accuracy and precision of coral ²³⁰Th dating method. By applying quality control, including careful sampling and subsampling protocols and the use of contamination-free storage and workbench spaces, the resulting low procedural blanks give an equivalent uncertainty in age of only ±0.2-0.3 yr for 1-2 g of coral sample. Using site-specific ²³⁰Th/²³²Th₀ values or isochron techniques, our study demonstrates that corals with an age less than 100 yrs can be ²³⁰Th-dated with precisions of ±1 yr. Six living subtidal coral samples were collected from two continental shelf sites, Nanwan off southern Taiwan in the western Pacific and Son Tra off central Vietnam in the South China Sea; one coral core was drilled from an open-ocean site, Santo Island, Vanuatu, in the western tropical Pacific; and modern and fossil intertidal coral slabs, 17 in total, were cut from six sites around the islands of Simeulue, Lago, North Pagai and South Pagai of Sumatra in the eastern Indian Ocean. The results indicate that the main source of thorium is the dissolved phase of seawater, with variation of ²³⁰Th/²³²Th₀ depending on local hydrology. With intense input of terrestrial material, low ²³⁰Th/²³²Th₀ atomic ratios of 4.9 × 10⁻⁶ and 3.2 × 10⁻⁶ with a 10% variation are observed in Nanwan and Son Tra, respectively. At the Santo site, we find a value of 5.6 × 10⁻⁶ at 4 horizons and one high value of 24 × 10⁻⁶ in a sample from AD 1974.6 ± 0.5, likely due to the upwelling of cold water during a La Niña event between AD 1973 and 1976. The natural dynamics of ²³⁰Th/²³²Th₀ recorded in the intertidal corals at sites in the Sumatran islands are complicated so that this value varies significantly from 3.0 to 9.4 × 10⁻⁶. Three of the 141 modern coral ²³⁰Th ages differ from their true ages by −23 to +4, indicating the presence of detrital material with anomalous ²³⁰Th/²³²Th values. Duplicate measurement of coeval subsamples is therefore recommended to verify the age accuracy. This improved high precision coral ²³⁰Th dating method raises the prospects of refining the age models for band-counted and tracer-tuned chronologies and of advancing coral paleoclimate research.     

Airborne dust distribution in longwall faces

Summary The generation and transportation of airborne dust is governed mainly by the velocity and the movement pattern of the ventilation air in longwall faces. Generally speaking, the higher the air velocity, the more airborne dust will be generated and the further dust of a given size will be carried away from the source. In order to effectively prevent or reduce the production and spread of the airborne dust, the airborne dust concentration distribution at a longwall face must be known in advance. This paper presents the measurement methodology, data analyses and dust distributions in five longwall faces.     

Comparison of static-feedforward and dynamic-feedback neural networks for rainfall–runoff modeling

A systematic comparison of two basic types of neural network, static and dynamic, is presented in this study. Two back-propagation (BP) learning optimization algorithms, the standard BP and conjugate gradient (CG) method, are used for the static network, and the real-time recurrent learning (RTRL) algorithm is used for the dynamic-feedback network. Twenty-three storm-events, about 1632 rainfall and runoff data sets, of the Lan-Yang River in Taiwan are used to demonstrate the efficiency and practicability of the neural networks for one hour ahead streamflow forecasting. In a comparison of searching algorithms for a static network, the results show that the CG method is superior to the standard BP method in terms of the efficiency and effectiveness of the constructed network's performance. For a comparison of the static neural network using the CG algorithm with the dynamic neural network using RTRL, the results show that (1) the static-feedforward neural network could produce satisfactory results only when there is a sufficient and adequate training data set, (2) the dynamic neural network generally could produce better and more stable flow forecasting than the static network, and (3) the RTRL algorithm helps to continually update the dynamic network for learning—this feature is especially important for the extraordinary time-varying characteristics of rainfall–runoff processes.     

Characteristics of the wedge-top depozone of the southern Taiwan foreland basin system

The wedge‐top depozone in the southern Taiwan foreland basin system is confined by the topographic front of the Chaochou Fault to the east and by a submarine deformation front to the west. The Pingtung Plain, Kaoping Shelf and Kaoping Slope constitute the main body of the wedge‐top depozone. In a subaerial setting, the alluvial and fluvial sediments accumulate on top of the frontal parts of the Taiwan orogenic wedge to form the Pingtung Plain proximal to high topographic relief. In a submarine setting, fine‐grained sediments accumulate on the Kaoping Shelf and dominant mass‐wasting sediment forms the Kaoping Slope. Wedge‐top sediments are deformed into a series of west‐vergent imbricated thrusts and folds and associated piggyback basins. A major piggyback basin occurs in the Pingtung Plain. Four smaller piggyback basins appear in the shelf–slope region. Many small‐sized piggyback basins developed over ramp folds in the lower slope region. Pliocene–Quaternary deep marine to fluvial sediments about 5000 m thick have been deposited on top of the frontal orogenic wedge in southern Taiwan. Sedimentary facies shows lateral variations from extremely coarse fluvial conglomerates proximal to the topographic front (Chaochou Fault) to fine‐grained deep marine mud close to the deformation front near the base of the slope. The stratigraphic column indicates that offshore deep‐water mud is gradationally overlain by shallow marine sands and then fluvial deposits. The transverse cross‐section of the wedge‐top depozone in the southern Taiwan is a doubly tapered prism. The northern boundary of the wedge‐top depozone in southern Taiwan is placed along the southern limit of the Western Foothills where the frontal orogenic wedge progressively changes southward to a wedge‐top depozone (Pingtung Plain), reflecting ongoing southward oblique collision between the Luzon Arc and the Chinese margin. The wedge‐top depozone is bounded to the south by the continent–ocean crust boundary. The deep slope west of the Hengchun Ridge can be viewed as an infant wedge‐top depozone, showing initial mountain building and the beginning of wedge‐top depozone.