Variation of As concentration between soil types and rice genotypes and the selection of cultivars for reducing As in the diet

Human exposure to toxic heavy metals via the food chain is of increasing concern. In the present study, the effects of soil type and genotype on variation in arsenic (As) concentrations of different organs were investigated by using nine rice cultivars grown in two soils, with two levels of As contamination. There were significant genotypic differences (P < 0.05) in As concentrations of all organs, and As concentrations of polished grain were significantly affected by genotype and soil type. The As concentration in polished grain was higher in red paddy soil under As treatment, with range from 0.24 to 1.03 mg kg⁻¹, and the As concentration of three cultivars exceeded the concentration of Chinese Food Hygiene Standard (0.7 mg kg⁻¹). The As concentrations in stems, leaves and polished grain were all significantly and positively correlated. The As concentrations in polished grain were positively and significantly (P < 0.01) correlated with As root-grain translocation factor. The results indicated that As concentration in grain was partially governed by As uptake and the transfer of As from root to grain. The grain As concentration of the nine cultivars was significantly correlated between the two soil types at different levels of As contamination. Some genotypes, such as japonica rice (e.g. Ning jing 1 and Nan jing 32) had consistently low grain As concentrations. The results suggest the possibility of breeding the As rice cultivars to produce grain for safe consumption from soils with slight and moderate levels of As.     

对一起烧山案法律定性的看法

由一起烧山案引入,详细地分析了失火罪、破坏生产经营罪、重大责任事故罪这三种在林业生产中经常会涉及的犯罪的四个构成要件,并比较了它们的异同,对基层执法人员的实际工作具有一定的借鉴作用。     

不同热力背景对城市降雨（暴雨）的影响（Ⅲ）：—基于人工神经网络的集成预报模型

在对降雨系统时序规律分析的基础上,提出了基于人工神经网络的集成降雨时序分析与预报模型,根据上海市降雨和实际测点布置的特点,提出了两套降雨预报策略,根据热力背景的不同,建立了城市降雨空间差异预报模型,经初步验证,该模型具有较高精度,可为城市规划,排水市政工程建设服务。     

长江经济带经济高质量发展差异性机理及对策分析

基于驱动力—压力—状态—影响—响应(DPSIR)模型,建立经济高质量发展系统作用机理分析框架和概念模型,运用结构方程模型对长江经济带上、中、下游三大区域经济高质量发展路径和关键因素进行探索。结果表明:①长江经济带下游地区作用机理为"驱动力—压力—状态—影响—响应"单向链式作用路径,资源消耗强度、贫困差异和生态环境质量为最关键因素。②长江经济带中游地区为"驱动力—压力—状态"和"影响—响应"跳跃式作用路径,产业结构为最关键因素。③长江经济带上游地区为"状态—影响—响应"的中断式作用路径,生态资源、科技创新和城市化为最关键因素。     

基于MSLSTM-DA模型的水质自动监测异常数据报警

提出一种基于多元堆叠长短时记忆网络-差值分析(MSLSTM-DA)模型对地表水质异常数据进行报警的方法.该方法首先建立MSLSTM模型对水质指标数据进行预测,再基于预测结果的残差分布建立DA模型,并确定各个指标的数据异常阈值,当实测数据与预测数据差值大于阈值时进行数据报警.以长江流域监测断面的水质数据进行了方法有效性验证.结果表明,构建的预测模型对５个指标的MAE、MAPE均值比BP神经网络预测模型降低21.0%,17.8%,比LSTM模型降低16.8%,17.9%.皮尔逊系数均值比BP神经网络、LSTM模型的分别高5.9%,4.4%.5个指标共检出水质异常数据37条,其中34条经人工判断确实存在有异常,报警准确率高达91.9%.