Testing Backpropagation Neural Network Approach in Interpolating Missing Daily Precipitation

This study proposes a method, backpropagation (BP) neural network, for interpolating missing values in daily precipitation time series. Firstly, the BP neural network is adopted to interpolate missing daily rainfall data at three selected stations in Yantai, Shandong, China. Then, the temporal and spatial variations in precipitation extremes across Shandong are analyzed by utilizing the complete daily rainfall dataset derived from accurate propagation at 24 meteorological stations. The results show that the long-term trends in five selected extreme precipitation indices calculated from interpolated daily rainfall data are generally consistent with those from original nonmissing values. And the spatial patterns of trends in precipitation extremes also show better performance for BP neural network approach in interpolating missing daily rainfall gaps. Those suggest that this BP neural network algorithm can obtain a good fit in terms of space-time variability of regional precipitation extremes, in case that the correlation coefficients between the target stations with missing values and reference stations with complete daily rainfall dataset are relatively large. These findings could be crucial for investigating regional frequency of heavy rainfall and water resource management.     

ASSESSING LAND DEGRADATION RISK THROUGH THE LONG‐TERM ANALYSIS OF EROSIVITY: A CASE STUDY IN SOUTHERN SPAIN

Despite the high variability of the precipitation regime characterizing the Mediterranean area, the records of rainfall depth are usually not appropriate for long‐term calculations of erosivity and soil losses, because they do not reveal details of short lengths or long durations (daily, monthly). In this work, we present a simple approach to calculate annual erosivity through monthly precipitation records. The study area (olive groves on steep slopes) has a high erosion risk associated to the main soil land use, combined with an irregular and erosive rainfall regime. The relationships between rainfall data at intervals of 10 min for a period of 3 years, daily rainfall records over 10 years and a long‐term monthly dataset of 60 years were checked to calculate the annual erosivity values through daily data, Fourier's index and modified Fourier's index values. A good, adjusted linear relationship between modified Fourier's index and the erosivity was found, which allowed us to optimize the use of the 60‐year monthly data series and to carry out a long‐term analysis of the erosivity quantiles in the study area. The estimated mean erosivity showed a return period of between 2 and 5 years and a variation coefficient of over 50 per cent, which illustrate its high variability and frequency. This approach to calculate erosivity and the use of quantiles could be applied in other areas with month‐long data series in order to study and model the erosion risk using suitable temporal periods. Copyright © 2011 John Wiley & Sons, Ltd.     

Temporal variability of rainfall in a semiarid environment in Brazil and its effect on sediment transport processes

Purpose

The temporal variabilities of both soil erosion by water and sediment redistribution in watersheds are directly related to rainfall characteristics. The purpose of this work was to assess the temporal pattern of rainfall in a semiarid watershed in Brazil and explain how this feature controls soil erosion and sediment yield.

Materials and methods

Daily and 5-min rainfall records were used to assess the temporal pattern down to the sub-hourly scale. To study the effect of the rainfall on sediment processes, erosivity and sediment yield at the Aiuaba (12 km²) and Benguê (933 km²) watersheds, Brazil were determined. Erosivity was calculated based on the rainfall kinetic energy method, while sediment yield was estimated from sediment rating curves and daily water discharge measurements.

Results and discussion

A large portion of annual rainfall is restricted to a few rain events and strong concentration in the sub-daily scale occurs, producing high erosivity. The temporal concentration of erosivity is greater than that of rainfall; the 10th percentile of the highest magnitude events encompasses 51% of the precipitation, but 80% of the erosivity. The temporal concentration of sediment yield is more pronounced; 88 and 98% of the sediment yield for the Aiuaba and Benguê watersheds, respectively, are within the 10th percentile of events.

Conclusions

The strong temporal concentration of precipitation causes events with high intensity and erosivity, thus allowing for soil detachment. Nonetheless, the low runoff rates limit downstream sediment transport. Such behavior produces a much higher temporal concentration of sediment yield, which reaches its maximal after a sequence of rainy days, when hydrological connectivity is enhanced and the sediments are propagated throughout the entire transport-limited system.     

基于IMERG反演降水数据估算广东省降雨侵蚀力

近年来遥感反演降水产品的时空分辨率不断提高，为估算区域尺度上具有空间连续性的降雨侵蚀力提供了新的可能。但以往研究在应用遥感降水产品估算降雨侵蚀力时多忽略了其与站点观测数据间的差异和对其纠偏的可能性。该研究以广东省86个气象站2001—2020年的逐时降水资料估算的降雨侵蚀力为观测值，评估两套IMERG（integrated multi-satellite retrievals for GPM）遥感降水产品-GPM_3IMERGHH（0.1°，逐30-min）和GPM_3IMERGDF（0.1°，逐日）对广东省降雨侵蚀力的估算精度并量化偏差，再结合拟合纠偏确定基于遥感反演降水数据估算广东省降雨侵蚀力的最优方法。结果表明：这两套产品均不适宜直接估算降雨侵蚀力指标，不同时间尺度、不同方法直接应用时精度均较低，克林-古普塔效率系数（Kling-Gupta efficiency, KGE）小于等于0.51。但多年平均和极端次事件降雨侵蚀力与对应观测值间具有强相关性（皮尔逊相关系数大于等于0.78），具备纠偏的潜力。因此，本研究发展线性模型对IMERG估算结果进行纠偏，交叉验证结果表明纠偏后GPM_3IMERGHH估算多年平均降雨侵蚀力（R因子）的KGE可达0.79，10年一遇EI₃₀的KGE可达0.64，优于采用站点日降水估算降雨侵蚀力并插值的精度（KGE分别为0.60和0.59），与采用站点小时降水估算降雨侵蚀力并插值的精度相近（KGE分别为0.77和0.66）。当前研究结果充分展示了遥感反演降水在土壤水蚀领域的应用潜力和前景。     

辽河流域降雨侵蚀力的时空变化分析

降雨侵蚀力是反映流域降雨侵蚀能力的综合指标之一。根据辽河流域10个气象站的日降雨量资料,利用日降雨侵蚀力模型估算辽河流域的降雨侵蚀力。结果表明：辽河流域降雨侵蚀力的空间变异与降雨量的空间分布趋势基本一致,由东南向西北递减,变化于1000—3800MJ·mm／（hm^2·h·a）之间;降雨侵蚀力年内集中度高,6—8月3个月约占全年的80％;降雨侵蚀力年际变化大,年际变率Cv在0．367—0．649之间,采用时序系列的Mann—Kendall检验表明,降雨侵蚀力并无显著变化趋势;特别是在流域水土流失严重的西辽河地区,年降雨侵蚀力较小,但年内集中程度大,年际变化更突出。     

浙江省降雨侵蚀力变化特征分析

根据浙江省83个气象站1980-2009年逐日雨量资料,采用日降雨侵蚀力模型,运用Mann-Kendall非参数检验和径向基函数插值等方法,研究了该省降雨侵蚀力变化的时空分布特征。结果表明,近30a来浙江省多年降雨侵蚀力长期变化趋势不显著,但冬季呈显著上升趋势,1和12月为上升趋势的主要贡献月份;浙江省多年平均降雨侵蚀力由西北向东南递增,月、季降雨侵蚀力变化趋势显著的集中区主要位于浙中浙北地区,但夏季降雨侵蚀力较大,上升趋势显著的集中区位于该省东南部,该地区降雨潜在侵蚀动能较大,是水土保持的重点区域。     

陕北黄土高原月降雨理论分布的初步研究

利用陕北两个典型气象站点多年降雨强度数据,使用数学方法描述月雨强分布的形状,从而满足水文模型、土壤侵蚀模型等对降雨强度数据的需求。通过探讨小时雨强在月时间尺度上应符合的理论分布函数,并采用统计学参数估计方法进行参数拟合,来选择合适的数学方法对月雨强分布进行描述。使用最大似然法对常见的7种理论分布函数进行拟合的结果表明,LOGN、GAMA、WBL等理论分布函数都可以较好地再现月降雨强度分布,90%以上的月份COE值大于0.7,一半以上的月份COE值大于0.9。另外,WBL分布对降雨强度峰值亦有良好的表现能力,因此推荐在陕北黄土高原降雨分布的模拟过程中使用。     

A simple method for generating daily rainfall data

In recent years, agricultural scientists have developed considerable interest in modeling and simulation of rainfall as new ways of analyzing rainfall data and assessing its impact on agriculture. Among the proposed methods, a combination of Markov chain and gamma distribution function is recognized as a simple approach and is demonstrated to be effective in generating daily rainfall data for many environments. Unfortunately, this method requires that many years of daily weather records be available for estimating the model parameters. Thus the availability of the weather data limits the applicability of the simulation method. When these model parameters are evaluated over time and at different places, however, certain general characteristics are revealed. First, the transitional probability of a wet day followed by a wet day tends to be greater but parallel to the transitional probability of a dry day followed by a wet day. This phenomenon leads to a linear relationship of the transitional probabilities to the fraction of wet days per month. Second, the beta parameter in a gamma distribution function, which is used to describe the amount of rainfall, is closely related to the amount of rain per wet day owing to the positive skewness of the rainfall distribution. Based on these relationships, a simple method is introduced, by which model parameters can be estimated from monthly summaries instead of from daily values. The suggested method, therefore, provides a convenient vehicle for applying weather simulation models to areas in which its use had been impossible because of the unavailability of long series of daily weather data.     

多源强降雨灾情可信度智能判别方法

真实的灾情信息是有效防范和减轻强降雨灾害损失的重要参考。本研究以过程降雨强度（R）为指标,构建1984-2020年河北省县级多源气象灾情与致灾过程相匹配的强降雨灾害事件库。经过人工质控,获得真实灾情信息2305组,伪灾情信息263组。采用相关分析法确定与灾害发生程度（灾度）显著相关的降雨关键特征因子,基于单类支持向量机和十折交叉检验法,随机抽取10次样本,建立强降雨灾情气象因子致灾判别模型,并进行检验优化,以探索智能化、易用性的多源强降雨灾情可信度智能判别方法。结果表明：（1）与灾度显著相关的降雨关键特征因子共计11个,分别为最大降雨量、最小降雨量、过程平均降雨量、日均降雨量、平均小时雨强、1h最大雨量、3h最大雨量、6h最大雨量、12h最大雨量、24h最大雨量及前10日降雨总量,均通过了0.01水平显著性检验。（2）采用11个因子建立10个致灾判别模型（M1-M10）,依据真实灾情判别准确率确定最优模型为M9,其证真率为96.4%,证伪率为67.6%,表明该模型对灾情真伪判定较为片面,应进一步优化。（3）通过自相关检验,以最大降雨量、平均小时雨强、1h最大雨量及前10日降雨总量4个因...     

基于Hurst指数与相关系数的降雨侵蚀力变异识别与分级方法

变化环境下区域降雨侵蚀力的时空变化问题对区域水土流失防治工作提出了新的挑战。降雨侵蚀力序列不再是纯随机序列,往往存在趋势、跳跃或者周期的变化,在对降雨侵蚀力序列分析与计算时,现有研究往往采用单一的检验方法,缺乏对降雨侵蚀力序列各类成分的综合比较,所得到的结果可信度及其程度如何无法判断。该研究提出了基于Hurst系数和相关系数的降雨侵蚀力序列联合分析方法。该方法首先计算降雨侵蚀力序列的Hurst系数,引用水文序列变异的概念,从统计学角度将降雨侵蚀力序列确定性成分分为三级(无变异、弱变异和强变异)。然后通过多种检验方法综合检验,将得到的结果与原序列进行相关性分析提取相关系数最大的确定性成分(趋势、跳跃和周期),对其进行剔除,重复上述步骤,将降雨侵蚀力序列中的确定性成分进行一一分解,最终得出的降雨侵蚀力序列将是一个随机序列与确定性序列的组合。实际应用中,根据长江流域174个气象站点1961—2014年逐日降雨资料,对流域内各气象站点年降雨侵蚀力序列进行确定性成分分析与分级结果表明:长江流域174个气象站点中有130个站点降雨侵蚀力序列无明显变异,有31个站点降雨侵蚀力序列出现弱变异,有13个站点降雨侵蚀力序列出现强变异。以重庆奉节站为例进行综合检验,分析结果为整体强变异,该站年降雨侵蚀力序列存在复合周期和跳跃成分,其中复合周期为5 a和16 a,向下的跳跃点为2011年。该研究为变化环境下区域降雨侵蚀力预测提供理论依据。     

黑龙江省降雨侵蚀力空间分布规律

利用黑龙江省16个国家级气象站,1960-2000年日降雨量资料,分析黑龙江省侵蚀性降雨和降雨侵蚀力的空间分布规律。在16个气象站中,日降雨量达到侵蚀性标准（≥12mm／d）的降雨时间为9～15d／a,最大值同最小值之间相差近0．7倍;日降雨量达到侵蚀性标准的年降雨量为192～387mm,最大值同最小值之间相差l倍。16个气象站年降雨侵蚀力多年平均值为794～2144MJ·mm／（hm^2·h·a）,最大值同最小值之间相差近2倍。降雨侵蚀力空间分布从西北到中南部逐渐升高,东部低于中部,年降雨侵蚀力空间分布基本与年降雨量空间分布相似。年内降雨侵蚀力分布主要集中在6—9月,7月份下半月或8月份上半月达到最高值,6—9月降雨侵蚀力占全年比率为88％～95％,其中西部比东部略高。     

两种Penman-Monteith公式计算草坪草参考腾发量的适用性

为了揭示ASCE和FAO56两种Penman-Monteith公式在计算小时参考作物腾发量（ET0）时的差异,开展了充分供水草坪草腾发量观测试验。基于自动气象站的小时气象数据和蒸渗仪试验结果,在对比两公式计算结果差异基础上,以实测的日草坪腾发量为标准评价了2种计算公式小时ET0的日累积结果及以日的计算结果。结果表明：2种Penman-Monteith公式计算的小时ET0结果存在一定差异,ET0较高的时段差异也比较大。白天FAO56 Penman-Monteith公式的计算结果低于ASCE Penman-Monteith公式的计算结果,夜晚则正好相反,原因在于Cd取值的差异。与实测日ET0结果相比2种公式小时时段的ET0结果的累积值误差均比较大,ASCE的改进并没有使Penman-Monteith在计算结果上取得实质性的改进,相比之下以日为时段的Penman- Monteith公式（ASCE同FAO56）取得了与实测结果最为一致的效果。进一步根据实测的小时ET0数据以及更长序列的日ET0实测结果,评价FAO56 Penman-Monteith和ASCE Penman-Monteith结果的地区适用性将是今后研究内容之一。     

日降雨对降雨侵蚀力年雨量简易算法的影响分析

使用重庆市沙坪坝气象站1951—2010年日降雨数据构建降雨侵蚀力年雨量简易算法,分析年雨量简易算法预测精度,确定了年雨量算法预测精度最高时对应的日雨量,并进一步探讨了年降雨侵蚀力和降雨参数的时间变化特征。结果表明:(1)年雨量与年降雨侵蚀力呈指数关系。日雨量≥25mm的年雨量与年降雨侵蚀力的关系最为密切;采用日雨量≥25mm的年雨量算法预测年降雨侵蚀力的精度均优于其他日雨量对应算法。(2)1951—2010年年降雨侵蚀力与年降雨参数随时间变化趋势不显著。对于某一变化趋势时段,日雨量≥25mm的年雨量与时间的相关系数和年降雨侵蚀力与时间的相关系数数值相近。可使用日雨量≥25mm的年雨量作为替代指标分析水土流失对气候变化的响应。该文研究结果可为提高降雨侵蚀力简易算法预测精度,深入理解降雨侵蚀力对气候变化的响应以及区域水土流失防治提供参考。     

European emission data with high temporal and spatial resolution

The estimation of acidic deposition strongly depends on the availability of accurate emission data. The atmospheric models, that calculate concentrations and depositions of pollutants, need data in a high temporal resolution (e.g. daily, 6-hourly or even hourly data). However, although some progress has been achieved concerning annual emission data for Europe (e.g. CORINAIR 90), only very little information is available about the temporal variation of these emissions during a year. Therefore, within the EUROTRAC-GENEMIS project special emphasis was laid on the development of methods to generate emissions with a high temporal and spatial resolution. As results the temporal and spatial distribution of SO₂- and NO_X-emissions are shown. The results indicate, that the emissions vary considerably over time and that the use of simple patterns for the temporal disaggregation is not sufficient for modelling and assessment of acidic deposition.     

渭河流域降雨侵蚀力时空分布特征

[目的]揭示渭河流域降雨侵蚀力的时空变化特征,为区域水土保持规划提供依据。[方法]根据渭河流域及其周边范围30个气象站点1957—2014年逐日降雨资料,采用章文波日降雨量侵蚀模型计算各站点的降雨侵蚀力,分析其空间分布规律和年内分布特征。[结果]渭河流域多年平均降雨侵蚀力值分布范围为806.25~3 510.81 MJ·mm/(hm2·h),平均值1 798.97 MJ·mm/(hm2·h),与多年平均侵蚀性降雨的空间分布基本一致,总体呈现西北低东南高的趋势。渭河流域降雨侵蚀力年内变化呈单峰型,主要集中在7—9月,占全年降雨侵蚀力的63.91%。北部黄土高原地区和关中平原发生水土流失的时期集中在7—9月,而秦岭北麓地区5—10月均有可能发生较大的水土流域,侵蚀风险由西北向东南递增。流域降雨侵蚀力年际波动较大,年际变率Cv值在34%~56%之间,整体而言,流域西北部地区的降雨侵蚀力年际变化幅度大于东南部地区。除洛川、长武、环县、平凉4个站点降雨侵蚀力在研究时段内有所增大外,其余地区降雨侵蚀侵蚀力呈不同速率的减小趋势。[结论]渭河流域降雨侵蚀力时空分布差异显著,尽管流域降雨侵蚀力呈减弱趋势,由于流域地处黄土高原,水土保持与水源涵养工作仍需高度重视。     

金沙江流域降雨侵蚀力时空分布特征

[目的]分析金沙江流域降雨侵蚀力的时空分布的变化特征,为优化流域土壤流失定量评估及水土保持规划编制工作提供支持。[方法]利用气象台站降水资料验证了TRMM降水数据估算降雨侵蚀力在金沙江流域内的适用性,并结合气象站、TRMM和DEM数据,在Arc/Info软件提供的地图代数运算功能支持下,利用日雨量模型估算降雨侵蚀力开展分析和研究。[结果]1998—2015年TRMM 3B42降水数据和气象站降水数据估算金沙江流域多年平均的总体精度达到了82%,说明二者估算降雨侵蚀力的结果在合理误差范围内,金沙江流域降雨侵蚀力大体呈由东南向西北递减的趋势,地区差异大。总体上,高程越小的地区,降雨侵蚀力越大。流域年际变化同样存在空间分异,整体上呈现降低的趋势。[结论]将TRMM 3B42降水数据应用于气象站点稀疏的金沙江流域的多年平均降雨侵蚀力估算是可行的。但是各个站点估算结果的一致性高低程度不同,且某些年份的适用性程度受极端气候的影响。     

Potential predictability of crop yield using an ensemble climate forecast by a regional circulation model

Global/Regional Circulation Models (GCM/RCM) predict the interannual climate variability better than the absolute values of meteorological variables. Statistical bias-correction methods increase the quality of daily model predictions of incoming solar radiation, maximum and minimum temperatures and rainfall frequency and amount. However, when bias-corrected forecasts/hindcasts are used by dynamic crop models, timing of dry-spell occurrences generate the largest uncertainty during the linking process. In this study, we used 20 ensemble members of an 18-year period provided by the Florida State University/Center for Ocean-Atmospheric Prediction Studies (FSU/COAPS) regional spectral model coupled to the National Center for Atmospheric Research Community Land Model (CLM2). The daily seasonal-climate hindcast was bias-corrected and used as input to the CERES-Maize model, thus producing 20 crop yield ensemble members. Using observed weather data for the same period, a time series of simulated crop yields was produced. Finally, principal component (PC) regression analysis was used to predict this time series using the crop yield ensemble members as predictors. Between 13.7 and 28.8% of the simulated corn yield interannual variability was explained using only one principal component (p < 0.05), and estimated yields were in the correct tercile by margins of 16.7 to 38.2% beyond chance. Predictability of simulated corn yields using principal components was improved relative to the use of bias-corrected daily hindcasts. Bias-correcting all meteorological variables used by the crop model increased predictability skills compared with use of raw hindcasts, individual bias-correction of rainfall, and climatological values.     

河南省降雨侵蚀力时空变异特征

利用河南省119个气象台站自建站始至2003年的逐日降雨量资料,计算其所代表县(市)的逐年与平均降雨侵蚀力,并利用GIS等工具分析河南省降雨侵蚀力的时空变异特征。结果表明:河南省多年年均降雨侵蚀力总体趋势是由北向南递增,最大值出现在南部的新县、鸡公山、商城与桐柏、平舆,其值均超过7 000 MJ.mm/(hm2.h.a);河南省多年年均降雨侵蚀力排序结果可分为5个区域等级,基本与等值线一致;河南省各地的降雨侵蚀力在不同年份变异较大,FFT(快速傅立叶变换)表明无明显的年际周期性规律;河南省降雨侵蚀力的年内变化趋势表现为单峰型,侵蚀主要发生在7—9月份,集中度在北部区域均超过60%;河南省降雨侵蚀力与年侵蚀性降雨量或年侵蚀性降雨量和逐日侵蚀性降雨量之间存在极显著的线性相关性,相关系数分别为0.967 2和0.994 2。     

谢儿渠小流域天然降水利用研究

本文以谢儿渠流域１９８３～１９９５年１３年的降水实测资料为基础，分析了流域降水在时间上的分布特点和一次降水量特点以及拦蓄利用的可能性，预测了可利用的降水量、径流量以及可发展窖灌农业的潜力．提出了流域利用水窖蓄水发展窖灌农业的方案。发展窖灌农业是雨育农业区小流域水土保持综合治理的一项重要内容，是农业种植从粗放经营向集约化经营转变的一种方式，是实现小流域经济持续发展的关键。