A simulation-based two-stage interval-stochastic programming model for water resources management in Kaidu-Konqi watershed,China

This study presented a simulation-based two-stage interval-stochastic programming （STIP） model to support water resources management in the Kaidu-Konqi watershed in Northwest China. The modeling system coupled a distributed hydrological model with an interval two-stage stochastic programing （ITSP）. The distributed hydrological model was used for establishing a rainfall-runoff forecast system, while random parameters were pro- vided by the statistical analysis of simulation outcomes water resources management planning in Kaidu-Konqi The developed STIP model was applied to a real case of watershed, where three scenarios with different water re- sources management policies were analyzed. The results indicated that water shortage mainly occurred in agri- culture, ecology and forestry sectors. In comparison, the water demand from municipality, industry and stock- breeding sectors can be satisfied due to their lower consumptions and higher economic values. Different policies for ecological water allocation can result in varied system benefits, and can help to identify desired water allocation plans with a maximum economic benefit and a minimum risk of system disruption under uncertainty.

Use of deep learning in forensic sex estimation of virtual pelvic models from the Han population

Accurate sex estimation is crucial to determine the identity of human skeletal remains effectively.Here,we developed convolutional neural network(CNN)models for sex estimation on virtual hemi-pelvic regions,including the ventral pubis(VP),dorsal pubis(DP),greater sciatic notch(GSN),pelvic inlet(PI),ischium,and acetabulum from the Han population and compared these models with two experienced forensic anthropologists using morphological methods.A Computed Tomography(CT)dataset of 862 individuals was divided into the subgroups of training,validation,and testing,respectively.The CT-based virtual hemi-pelvises from the training and validation groups were used to calibrate sex estimation models;and then a testing dataset was used to evaluate the performance of the trained models and two human experts on the sex estimation of specific pelvic regions in terms of overall accuracy,sensitivity,specificity,F1 score,and receiver operating characteristic(ROC)curve.Except for the ischium and acetabulum,the CNN models trained with the VP,DP,GSN,and PI images achieved excellent results with all the prediction metrics over 0.9.All accuracies were superior to those of the two forensic anthropologists in the independent testing.Notably,the heatmap results confirmed that the trained CNN models were focused on traditional sexual anatomic traits for sex classification.This study demonstrates the potential of AI techniques based on the radiological dataset in sex estimation of virtual pelvic models.The excellent sex estimation performance obtained by the CNN models indicates that this method is valuable to proceed with in prospective forensic trials.