The chaotic time series of Indian monsoon rainfall and its prediction

The time series of Indian summer monsoon rainfall for the period 1871–1989 has been analysed using the method of deterministic chaos. It is found that a strange attractor underlies the time series implying the existence of a prediction function. This function has been approximated by a second-degree polynomial, involving the rainfalls of the past seven years and the coefficients have been estimated by least squares fit. The interannual variations of actual and computed rainfalls have been presented for a comparative study.     

Homogeneous Indian Monsoon rainfall: Variability and prediction

The Indian summer monsoon rainfall is known to have considerable spatial variability, which imposes some limitations on the all-India mean widely used at present. To prepare a spatially coherent monsoon rainfall series for the largest possible area, fourteen subdivisions covering the northwestern and central parts of India (about 55% of the total area of the country), having similar rainfall characteristics and associations with regional/global circulation parameters are merged and their area-weighted means computed, to form monthly and seasonal Homogeneous Indian Monsoon (HIM) rainfall series for the period 1871–1990. This paper includes a listing of monthly and seasonal rainfall of HIM region. HIM rainfall series has been statistically analysed to understand its characteristics, variability and teleconnections for long-range prediction. HIM rainfall series isfound to be homogeneous, Gaussian distributed and free from persistence. The mean (R) rainfall is 757 mm (87% of annual) and standard deviation (S) 119 mm, with a Coefficient of Variation (CV) of 16%. There were 21 dry (K, -<R S) and 19 wet (R _i R + S) years during 1871–1990. There were clusters of frequent negative departures during 1899–1920 and 1965–1987 and positive departures during 1942–1961. The recent three decades show very high rainfall variability with 10 dry and 6 wet years. The decadal averages were alternatively positive and negative for three consecutive decades, viz., 1871–1900 (positive); 1901–1930 (negative); 1931–1960 (positive) and 1961–1990 (negative) respectively. Significant QBO and autocorrelation at 14th lag have been found in HIM rainfall series. To delineate the changes in the climatic regime of the Indian summer monsoon, sliding correlation coefficients (CCs) between HIM rainfall series and (i) Bombay msl pressure, (ii) Darwin msl pressure and (iii) Northern Hemisphere surface air temperature over the period 1871–1990 have been examined. The 31-year sliding CCs showed the systematic turning points of positive and negative CCs around the years, 1900 and 1940. In the light of other corroborative evidences, these turning points seem to delineate ‘meridional’ monsoon regime during 1871–1900 and 1940–1990 and ‘zonal’ monsoon regime during 1901–1940. The monsoon signal is particularly dominant in many regional and global circulation parameters, during 1951–1990. Using the teleconnections ofHIM series with 12 regional/global circulation parameters during the recent 36-year period 1951–86 regression models have been developed for long-range prediction. In the regression equations 3 to 4 parameters were entered, explaining upto 80% of the variance, depending upon the data period. The parameters that prominently enter the multiple regression equations are (i) Bombay msl pressure, (ii) April 500 mb Ridge at 75°E, (iii) NH temperature, (iv) Nouvelle minus Agalega msl pressure and (v) South American msl pressure. Eleven circulation parameters for the period 1951–80 were subjected to Principal Component Analysis (PCA) and the PC’s were used in the regression model to estimate HIM rainfall. The multiple regression with three PCs explain 72% of variance in HIM rainfall.     

Variation of coastal atmospheric boundary layer characteristics with convective activity along the west coast of India during the Arabian Sea Monsoon Experiment (ARMEX) 2002

This paper investigates the characteristic features of the coastal atmospheric boundary layer (CABL) along the west coast of India during the south-west monsoon (SWM) 2002. Extensive surface and upper-air findings were obtained during the same period from the Arabian Sea Monsoon Experiment (ARMEX; 15th June to 15th August 2002) 2002. The operational general circulation model (GCM) of the National Centre for Medium Range Weather Forecasting (NCMRWF) was used in this study to see the spatial variation of the CABL during two specific convective episodes that led to heavy rainfall along the west coast of India. The impact of a non-local closure (NLC) scheme employed in the NCMRWF GCM was carried out in simulating the CABL. The same episodes were also simulated using a similar parameterization scheme employed in the high resolution mesoscale modelling system (MM5). The diurnal variation of CABL is better represented from MM5 simulation. Comparing the MM5 simulation with that of the coarser grid NCMRWF GCM, we observed that the NCMRWF GCM underestimates the values of both latent heat flux (LHF) and the coastal atmospheric boundary layer height (CABLH). Results from MM5 therefore indicate that the best way to move forward in addressing the short-comings of coarse grid-scale GCMs is to provide a parameterization of the diurnal effects associated with convection processes.     

Probabilistic multi-model ensemble prediction of Indian summer monsoon rainfall using general circulation models: A non-parametric approach

Probabilistic prediction has the ability to convey the intrinsic uncertainty of forecast that helps the decision makers to manage the climate risk more efficiently than deterministic forecasts. In recent times, probabilistic predictions obtained from the products from General Circulation Models (GCMs) have gained considerable attention. The probabilistic forecast can be generated in parametric (assuming Gaussian distribution) as well as non-parametric (counting method) ways. The present study deals with the non-parametric approach that requires no assumption about the form of the forecast distribution for the prediction of Indian summer monsoon rainfall (ISMR) based on the hindcast run of seven general circulation models from 1982 to 2008. Probabilistic prediction from each of the GCM products has been generated by non-parametric methods for tercile categories (viz. below normal (BN), near-normal (NN), and above normal (AN)) and evaluation of their skill is assessed against observed data. Five different types of PMME schemes have been used for combining probabilities from each GCM to improve the forecast skill as compared to the individual GCMs. These schemes are different in nature of assigning the weights for combining probabilities. After a rigorous analysis through Rank Probability Skill Score (RPSS) and relative operating characteristic (ROC) curve, the superiority of PMME has been established over climatological probability. It is also found that, the performances of PMME1 and PMME3 are better than all the other methods whereas PMME3 has showed more improvement over PMME1.     

The role of low-frequency intraseasonal oscillations in the anomalous Indian summer monsoon rainfall of 2002

We analyze the dynamical features and responsible factors of the low-frequency intraseasonal time scales which influenced the nature of onset, intensity and duration of active/break phases and withdrawal of the monsoon during the anomalous Indian summer monsoon of 2002 — the most severe drought recorded in recent times. During that season, persistent warm sea surface temperature anomalies over the equatorial Indian Ocean played a significant role in modulating the strength of the monsoon Hadley circulation. This in turn affected the onset and intense break spells especially the long break during the peak monsoon month of July. Strong low-frequency intraseasonal modulations with significant impact on the onset and active/break phases occurred in 2002 which were manifested as a good association between low-frequency intraseasonal oscillations and the onset and active/break spells. Further, SST anomalies over the equatorial Indo-Pacific region on low-frequency intraseasonal time scales were found to affect the equatorial eastward and thereby off-equatorial northward propagations of enhanced convection over the Indian region. These propagations in turn modulated the active/break cycle deciding the consequent severity of the 2002 drought.     

On the impacts of ENSO and Indian Ocean dipole events on sub-regional Indian summer monsoon rainfall

The relative impacts of the ENSO and Indian Ocean dipole (IOD) events on Indian summer (June–September) monsoon rainfall at sub-regional scales have been examined in this study. GISST datasets from 1958 to 1998, along with Willmott and Matsuura gridded rainfall data, all India summer monsoon rainfall data, and homogeneous and sub-regional Indian rainfall datasets were used. The spatial distribution of partial correlations between the IOD and summer rainfall over India indicates a significant impact on rainfall along the monsoon trough regions, parts of the southwest coastal regions of India, and also over Pakistan, Afghanistan, and Iran. ENSO events have a wider impact, although opposite in nature over the monsoon trough region to that of IOD events. The ENSO (IOD) index is negatively (positively) correlated (significant at the 95% confidence level from a two-tailed Student t-test) with summer monsoon rainfall over seven (four) of the eight homogeneous rainfall zones of India. During summer, ENSO events also cause drought over northern Sri Lanka, whereas the IOD events cause surplus rainfall in its south. On monthly scales, the ENSO and IOD events have significant impacts on many parts of India. In general, the magnitude of ENSO-related correlations is greater than those related to the IOD. The monthly-stratified IOD variability during each of the months from July to September has a significant impact on Indian summer monsoon rainfall variability over different parts of India, confirming that strong IOD events indeed affect the Indian summer monsoon.

Seismotectonic studies of the coastal areas of India,Pakistan, Bangladesh,and Burma

Major geotectonic elements that are seismically active in the near-shore areas of the Indian subcontinent are the Mekran fault off the coast of Pakistan, the western part of the Narmada-Son lineament, the West Coast Fault off the west coast of India - a southward extension of the Cambay Rift, the Palghat Gap, the Godavari and Mahanadi grabens, transecting rather at an angle to the eastern coast of India and the Arakan-Yoma arcuate belt of Burma, which is a part of the global Alpine-Himalayan orogenic belt, continuing southwards into the Andaman-Nicobar island complex and the Java-Sumatra trench on the ocean floor of the advancing Indo-Australian Plate.The coastal belt exhibits varied degrees of seismicity from intensely seismic areas, like the Mekran coast off Pakistan, Kutch (India) and the Arakan-Yoma belt of Burma, with earthquake magnitudes of more than 8.0, while the intervening coastal areas of the Peninsular India are moderately seismic to aseismic. The remaining areas, namely, the major part of the coastal belt of Bay of Bengal in India and Bangladesh are broadly aseismic. However, the active Godavari graben and the eastern part of the coast of Bangladesh are frequented by low to moderate magnitude earthquakes. An extension of the active Arakan-Yoma belt in the Bay of Bengal in the form of the Andaman-Nicobar Island complex is highly seismic with a maximum earthquake magnitude of more than 8.0, while the Lakshadweep-Minicoy island complex, situated on the Chagos-Laccadive ridge is moderately seismic. This broad picture of coastal and marginal seismicity is corroborated by the geodynamics of the northern part of the Indo-Australian Plate.Observations along the coastal areas during historic and recent times, however, confirm the absence of significant tsunamis, though very mild tsunami surges have occasionally been observed along the coastal areas of the Bay of Bengal. No active volcanoes are known to exist in the coastal areas.Water reservoirs situated near the marginal areas of the Peninsular Shield exhibit moderate to intense seismic activities, viz. Ukai, Bhatsa, Koyna, Parambikulam, Sholayar, Idduki, and Kinnersani.     

Improving the performance of precipitation outputs from Global Climate Models to predict monthly and seasonal rainfall over the Indian subcontinent

Skilful prediction of the monthly and seasonal summer monsoon rainfall over India at a smaller spatial scale is a major challenge for the scientific community. The present study is aimed at achieving this objective by hybridising two mathematical techniques, namely synthetic superensemble (SSE) and supervised principal component regression (SPCR) on six state-of-the art Global Climate Models (GCMs). The performance of the mathematical model is evaluated using correlation analysis, the root mean square error, and the Nash–Sutcliffe efficiency index. Results feature reasonable improvement over central India, which is a zone of maximum rainfall activity in the summer monsoon season. The study also highlights improvement in the monthly prediction of rainfall over raw GCMs (15–20% improvement) with exceptional improvement in July. The developed model is also examined for anomalous years of monsoon and it is found that the model is able to capture the signs of anomalies over different gridpoints of the Indian domain.     

Relationship between the southern oscillation/El Nino and rainfall in some tropical and midlatitude regions

Association of ENSO (El Nino/southern oscillation) events with annual rainfalls in S. America, Africa, India, Australia and New Zealand was found to be poor except for India and Victoria. For these two regions the relationships were one-sided i.e. most of the ENSO were associated with droughts; but many droughts were not associated with ENSO. for some rainfall series, the relationships are significant and prominent periodicities existed in the long periodicity region.     

公元1180和1181年浙江金华地区梅汛期降水的重建

利用古代日记资料发掘历史气候信息是重建古气候的重要方法。根据南宋吕祖谦的《庚子辛丑日记》记录的金华(婺州)公元1180年和1181年5—6月份逐日天气记录,利用金华1954—2012年实测月雨量和雨日资料的统计规律,重建了1180年和1181年金华地区5—6月份梅汛期雨量,并结合《日记》反映的天气过程,确定了这2年金华的出入梅时间,结果表明1180年和1181年入梅较现代偏早,1180年5—6月份雨量比现代偏小,梅雨期略偏短,1181年则雨量偏大,梅雨期明显偏长。上述结论对这2年历史文献记载的旱涝灾情提供了直观的有力佐证。     

A study of the empirical functions of the height fields over India and their relation with the rainfall

The empirical orthogonal functions have been obtained for the individual summer monsoon (June through September) months using the grid point values of monthly 700 mb geopotential heights over Indian region. The data for 21 summer monsoon months for the years 1958 to 1978 have been used in the present computation. The major variance reduction is due to the first three dominant functions accounting over 80% of the total variance in each month. The variance reduction only due to the first function ranges from 45 to 65%. The first function has in-pbase oscillation throughout the area indicating that the area under study is homogeneous and the centre of the oscillation lies over northwest India. The amplitudes of the first function also show generally quasipers stence in their sign within a season. The second function has two centres of action over the region of monsoon trough which are in phase. The third function has also two centres oriented in the east-west direction but they are in the opposite phase. Fairly large values of correlation coefficients between the patterns of the different monsoon months suggest that the patterns for these months corresponding to the first and the second functions respectively are quite similar. The patterns for these months also evolve with time in a related way. The spectrum analysis to the time series of amplitudes indicates the presence of the quasi-periodicity of 3 years during these monsoon months. The amplitudes corresponding to the dominant functions are found to be significantly related with the rainfall of central and western parts of India     

Indian summer monsoon rainfall (ISMR) forecasting using time series data: A fuzzy-entropy-neuro based expert system

This study presents a model to forecast the Indian summer monsoon rainfall(ISMR)(June-September)based on monthly and seasonal time scales. The ISMR time series data sets are classified into two parts for modeling purposes, viz.,(1) training data set(1871-1960), and(2) testing data set(1961-2014).Statistical analyzes reflect the dynamic nature of the ISMR, which couldn't be predicted efficiently by statistical and mathematical based models. Therefore, this study suggests the usage of three techniques,viz., fuzzy set, entropy and artificial neural network(ANN). Based on these techniques, a novel ISMR time series forecasting model is designed to deal with the dynamic nature of the ISMR. This model is verified and validated with training and testing data sets. Various statistical analyzes and comparison studies demonstrate the effectiveness of the proposed model.     

Ocean-atmosphere interaction and synoptic weather conditions in association with the two contrasting phases of monsoon during BOBMEX-1999

Surface meteorological parameters acquired during the field phase experiment, BOBMEX-99, for the stationary periods (SP I and II) of the ship ORV Sagar Kanya over Bay of Bengal have been analysed. Active and weak monsoon conditions were observed during the first and the second phase of the experiment respectively over India as well as over the stationary ship location. The phase mean sea surface temperature (SST) is found to be the same in both the phases, however large differences have been observed in the phase mean values of wind speed, mean sea level pressure, latent heat and momentum fluxes at air-sea interface. Synoptic scale monsoon disturbances formed only during the period of strong north-south pressure gradient over the Bay region. Events of prominent fall in SST and in the upper 15 m ocean layer mean temperature and salinity values during typical rainfall events are cited. The impact of monsoon disturbances on ocean-atmosphere interface transfer processes has been investigated.     

流动沙丘水分深层入渗量与降雨的关系

为分析流动沙丘水分深层入渗与降雨的关系,采用YWB-01型水分深层渗漏记录仪,运用土壤水动力学原理,研究了毛乌素沙地流动沙丘降雨入渗水分传递过程及地表以下150 cm深度入渗补给特征.结果表明:流动沙丘深层土壤水分来源主要为降雨入渗,2010年入渗补给量为141.4 mm,2011年为355.8 mm,分别占同期降雨量的55.1%和68.2%.深层入渗主要补给期为5—11月,期间补给量占全年总量的95%.当土壤水分条件保持在田间持水量范围之内时,降雨过程对土壤150 cm深度的补给作用主要受降雨量和降雨强度的影响,入渗量与降雨量呈极显著正相关(P<0.01),降雨通常在30~48 h后渗透补给到150 cm深层土壤,入渗补给速率在前3~5 h呈快速上升趋势,3~8 h左右达到最大值,此后缓慢下降,整个入渗补给过程可以持续96 h以上.     

东绒布冰芯净积累量与印度夏季风降水的关系

珠穆朗玛峰东侧东绒布冰川积累区的降水主要是由印度夏季风环流带来的.根据东绒布冰芯记录恢复的净积累量与印度中北部和印度半岛东部地区的夏季风降水量具有基本一致的周期, 三者有着较好的相关性, 因此可以通过冰芯净积累量来反映更长时间序列印度某些地区夏季风降水量的变化趋势.东绒布冰芯净积累量的相对变化幅度大于印度夏季风降水量的变化, 表明高海拔地区的降水比低海拔地区具有更高的敏感性.     

上新世暖期和未来增暖背景下亚非夏季风降水变化的模拟对比研究

亚非夏季风降水对当地农业、水资源管理、粮食安全以及生态系统等均有广泛影响, 其对全球变暖的响应特征是一个重要的科学问题。本研究基于上新世模式比较计划(PlioMIP2)的15个上新世暖期模式模拟数据和第六次国际耦合模式比较计划(CMIP6)的31个高排放情景SSP5-8.5模式模拟数据, 对比分析了上新世暖期和未来增暖背景下亚非夏季风降水变化特征的差异及其成因机制。结果表明: 在这两类气候增暖背景下, 亚非夏季风降水都呈增加的趋势, 然而对应于全球平均1 ℃升温, 季风降水在上新世暖期的强度增幅(0.24 mm/day/℃)明显大于其在未来增暖情景的强度增幅(0.17 mm/day/℃), 前者降水异常值约为后者的1.4倍。这种差异主要源于, 对应于全球平均1 ℃升温, 上新世暖期中高纬增温幅度显著强于未来增暖情景, 而低纬度较小。上新世暖期低纬和北半球中高纬度之间的经向温度梯度减弱幅度更大, 有利于亚非夏季风环流显著增强, 从而导致亚非夏季风降水强度显著强于未来增暖时期。本研究表明, 气候增暖背景下低纬与中高纬度之间的经向温度梯度变化对亚非夏季风具有显著影响, 准确衡量暖期南-北半球间经向温度梯度的变化对预估区域季风和水循环变化至关重要。

     

降雨条件下坡面径流和入渗耦合的数值模拟

降雨条件下坡面径流和降雨入渗的模拟互为条件,必须采用耦合分析的方法才能较好地解决这一问题。从坡面径流和降雨入渗控制方程着手,建立了用有限元方法求解该问题的耦合方程,并编制了相应的计算程序。为了加快迭代收敛的速度,还对地表饱和的判断提出了一种近似处理的方法。算例表明,所提出的方法可行,能较好地反映实际现象。该方法可为边（滑）坡稳定、水土流失、山地洪水分析等提供计算依据。     

泥沙输移与坡面降雨和径流能量的关系

从导致土壤侵蚀的降雨和径流能量出发,提出了基于物理学原理的降雨能和径流能的概念,并采用人工模拟降雨实验,分析了泥沙输移与降雨和径流能的关系,结果显示:坡面薄层径流泥沙剥蚀量随着地表坡度、降雨能和径流能的增加而增加,雨滴击溅作用下泥沙剥蚀量远远大于无雨滴击溅作用时泥沙剥蚀量;薄层水流泥沙浓度随着坡度和降雨能的增加而增加,当坡度和降雨能一定时,泥沙浓度随着径流能的增加而减小;降雨扰动系数与降雨和径流能的比值按照对数关系增长,在相同坡度下,当降雨能一定时,降雨扰动系数随着径流能的增加而减小,当径流能一定时,降雨扰动系数随着降雨能的增加而增加。降雨能是导致泥沙剥离的本质,径流能是泥沙搬运的动力。     

降雨作用下砾石土斜坡坡面形态及临界特性

砾石土因其级配宽、不均匀系数大、透水性强等特点在工程防渗中得到广泛的应用,其强度和力学性质受粗细粒含量和粒间咬合程度、黏接状态等影响。通过人工降雨原型试验、模型试验、室内试验及理论分析,研究松散砾石土斜坡在降雨作用下坡面土体的形态特征和土体性状的变化以及斜坡稳定性和该过程中可能出现的临界状态,探索坡面松散砾石土触变液化的过程和机理。试验结果表明,砾石土斜坡在降雨过程中,坡面土体形态、坡面径流泥沙含量具有阶段性特性;各典型现象土体含水量分布具有区段性;土体强度和斜坡稳定性随着土体含水量的增加均存在明显的临界特征。     

Influence of southern oscillation and SSTs over Nino-3.4 region on the winter monsoon rainfall over coastal Andhra Pradesh

An analysis of the mean monthly data of 124 years reveals that the relationship between the Southern Oscillation Index in September and the winter monsoon rainfall (WMR) over Coastal Andhra Pradesh (CAP) is variable and non-stationary. In the recent four decades, however, SOI (Sept) is negatively and significantly correlated with CAP WMR. A similar analysis is performed using 50 years of mean monthly SSTs over Nino-3.4 region in August and September and CAP WMR to detect a possible relationship and there is a striking positive relation between them. In both of the above cases, the September signal is more significant in the recent four decades than for the other months and seasons for probable prediction of CAP WMR. Finally, to examine the influence of SO on the winter monsoon rainfall, a non-parametric test “Mann-Whitney Rank Statistics” test has been applied to the rainfall associated with extreme positive and negative SOI events