Application of stacking hybrid machine learning algorithms in delineating multi-type flooding in Bangladesh.

Floods are among the most devastating natural hazards in Bangladesh. The country experiences multi-type floods (i.e., fluvial, flash, pluvial, and surge floods) every year. However, areas prone to multi-type floods have not yet been assessed on a national scale. Here, we used locally weighted linear regression (LWLR), random subspace (RSS), reduced error pruning tree (REPTree), random forest (RF), and M5P model tree algorithms in a hybrid ensemble to assess multi-type flood probabilities at a national scale in Bangladesh. We used historical flood data (1988-2020), remote sensing images (e.g., MODIS, Landsat 5-8, and Sentinel-1), and topography, hydrogeology, and environmental datasets to train and validate the proposed algorithms. According to the results, the stacking ensemble machine learning LWLR-RF algorithm performed better than the other algorithms in predicting flood probabilities, with R2 = 0.967-0.999, MAE = 0.022-0.117, RMSE = 0.029-0.148, RAE = 4.48-23.38%, and RRSE = 5.8829.69% for the training and testing datasets. Furthermore, true skill statistics (TSS: 0.929-0.967), corrected classified instances (CCI: 96.45-98.35), area under the curve (AUC: 0.983-0.997), and Gini coefficients (0.966-0.994) were computed to validate the constructed (LWLR-RF) multi-type flood probability maps. The maps constructed via the LWLR-RF algorithm revealed that the proportions of different categories of flooding areas in Bangladesh are fluvial flooding 1.50%, 5.71%, 12.66%, and 13.77% of the total land area; flash floods of 4.16%, 8.90%, 11.11%, and 5.07%; pluvial flooding: 5.72%, 3.25%, 5.07%, and 0.90%; and surge flooding, 1.69%, 1.04%, 0.52%, and 8.64% of the total land area, respectively. These percentages represent low, medium, high, and very high probabilities of flooding. The findings can guide future flood risk management and sustainable land-use planning in the study area.

Detection of lead in soil implying sample heating and laser-induced breakdown spectroscopy.

The emission line intensities enhancement and sensitivity of laser-induced breakdown spectroscopy (LIBS) has been a subject of great interest for the last several years to improve the detection of the trace elements in soil and other environmental samples. Among several other methodologies, LIBS of the heated targets is emerging as one of the effective techniques to achieve the objective. We have investigated the effect of target heating (room temperature, 100°C, and 200°C) on the emission enhancement and plasma parameters of the laser-produced plasma on the soil sample containing 80 ppm lead. In addition, the limit of detection (LOD) of lead in soil has been determined at a fixed target temperature (200°C) and with varying lead concentration (20 ppm, 80 ppm, and 100 ppm) in the soil samples. With increasing the target temperature, not only do the emission line intensities, the excitation temperature, and electron number density increase, but also the spectral lines of Pb emerge, which were absent in the soil spectra recorded at room temperature. The limit of detection of lead in LIBS of the heated soil target has been determined as 3.8 ppm. This study reveals the potential application of the LIBS of a pre-heated target for the detection of lead with an improved LOD in the environmental sample.

Elemental composition analysis of granite rocks using LIBS and LA-TOF-MS.

Qualitative and quantitative analysis of the raw granite rocks acquired from deposits in the Hunza district, Gilgit area of Pakistan were studied using laser-induced breakdown spectroscopy (LIBS) and laser ablation time-of-flight mass spectrometry (LA-TOF-MS). The optical emission spectra of the granite rocks used in artificial jewelry and for flooring tiles show the emission lines of (Si, Ca, K, Fe, Mg, Al, Na, and Li) and (Si, Ca, K, Fe, Mg, Al, Ti, Na, Ba, and Li), respectively. The mass spectra of these granite rocks were also studied with the LA-TOF-MS, revealing analogous elemental compositions. The results obtained using LIBS and LA-TOF-MS for the raw granite samples display their ability as powerful and complementary tools for the compositional analysis of the geological samples.

Geometrical structure and interface dependence of bias stress induced threshold voltage shift in C60-based OFETs.

The influence of the nature of interface between organic semiconductor and gate dielectric on bias stress electrical stability of n-type C60-based organic field effect transistors (OFETs) was studied. The bias stress induced threshold voltage (Vth) shift was found to depend critically on the OFET device structure: the direction of V(th) shift in top-gate OFETs was opposite to that in bottom-gate OFETs, while the use of the dual-gate OFET structure resulted in just very small variations in V(th). The opposite direction of Vth shift is attributed to the different nature of interfaces between C60 semiconductor and Parylene dielectric in these devices. The V(th) shift to more positive voltages upon bias stress in bottom-gate C60-OFET was similar to that observed for other n-type semiconductors and rationalized by electron trapping in the dielectric or at the gate dielectric/C60 interface. The opposite direction of Vth shift in top-gate C60-OFETs is attributed to free radical species created in the course of Parylene deposition on the surface of C60 during device fabrication, which produce plenty of hole traps. It was also realized that the dual-gate OFETs gives stable characteristics, which are immune to bias stress effects.