On the recent strengthening of the relationship between ENSO and northeast monsoon rainfall over South Asia

The southeastern parts of India and Sri Lanka receive substantial rainfall from the northeast monsoon (NEM) during October through December. The interannual variability in NEM rainfall is known to be significantly influenced by the El-Niño/Southern Oscillation (ENSO). Unlike the southwest monsoon (SWM), the NEM rainfall is enhanced during the warm ENSO events, and vice versa. In the context of the recent weakening of the inverse relationship between Southwest Monsoon (SWM) and ENSO, we examine the secular variations in the positive relationship between ENSO and NEM rainfall over South Asia, showing that their relationship has strengthened over the recent years. Based on the analysis of GISST, IMD/CRU precipitation and NCEP/NCAR reanalysis data, we suggest that this secular variation of the relationship is due to epochal changes in the tropospheric circulation associated with ENSO over the region.     

The orthogonal structure of Monsoon rainfall variation over Sri Lanka

Summary The spatial organization of Monsoon rainfall over Sri Lanka is examined using Orthogonal Factor Analysis (OFA) on long-term mean monthly rainfall data. Three types of orthogonal structure of Monsoon regime in Sri Lanka have been identified. Interpretation of orthogonal factor scores revealed that a large amount of rainfall occurs from March to October in the southwestern parts of Sri Lanka, from December to February in the eastern parts, and in November in the northern and mid-western parts which are all represented by high positive factor scores. Orthogonal factor scores for the first three factors account for 93.6% of the total variance of mean monthly rainfall and clearly indicate that the southeast and northwest parts of the country with lowest rainfall, resulting from lack of Monsoons, are represented by negative factor scores. The three orthogonal factors identified different rainfall maxima in different time periods and, additionally, significant spatial differences between regions. Seasonal changes in the Monsoon wind system, ITCZ weather phenomena, and topography were the main factors which influence the spatial structure of Monsoon rainfall over Sri Lanka.With 4 Figures     

西南季风期间斯里兰卡降水的年代际、年际变化及其与热带印度洋的联系

本文主要研究了1979—2016期间斯里兰卡在西南季风期间降水的年代际、年际变化以及与印度洋海温的联系.首先用经验正交的方法分析了斯里兰卡以及周边地区降水的时空分布,发现前两个模态能够解释超过70%的方差.其中第一模态为均一模态,且其PC1以及斯里兰卡7 a滑动平均降水序列都有年代际变化,降水异常在2000年前后异常偏多和偏少.通过合成分析发现2000年之后降水的异常减少与热带西部、中部印度洋的暖海温异常有关.暖海温异常通过调整经向环流引起了斯里兰卡上空的下沉运动,抑制了降水.在第二模态中,负的信号出现在斯里兰卡大部分地区,只有在斯里兰卡北部海角很小地区出现了正的信号.PC2表现出了年际变化,且与热带东南印度洋海温异常有显著的关系.通过Gill-Matsuno响应,热带东南印度洋海温异常造成热带北印度洋上空的气旋性环流异常,引起了水汽的辐合,从而利于降水.     

Intraseasonal oscillation of the southwest monsoon over Sri Lanka and evaluation of its subseasonal forecast skill

斯里兰卡的雨季发生于5-9月间,主要受西南季风的控制.本文发现该地区的西南季风降水存在很强的次季节变率,主导周期为10-35天.降水的季节内变化与西传的异常气旋有关.进一步,利用S2S比较计划中欧洲中心的数值预报模式(ECMWF)提供的回报试验数据,评估了当今动力模式对斯里兰卡西南季风次季节变化的预报技巧.结果显示,对季风指数的预测技巧超过30天,而对降水指数的预测技巧大约两周,且模式的预报技巧具有明显的年际差异.分析表明,能否正确模拟出大尺度环流对热带对流的响应是影响斯里兰卡降水预测的重要因子.     

Oscillatory behavior of monsoon rainfall over Sri Lanka during the late 19th and 20th centuries and its relationships to SSTs in the Indian Ocean and ENSO

Summary In this study, we have analyzed the temporal oscillations of precipitation in meso-scale zones of Sri Lanka to examine potential existence of periodic oscillatory behavior in rainfall. Only a few statistically significant cycles were identified: a 3.5-year cycle in most of central Sri Lanka during the January–March rainfall regime and a cycle of the same length in southwestern Sri Lanka during the October regime. A 2.1-year cycle marks the northeastern parts of Sri Lanka during the December/April contrast rainfall regime. This cycle is shown to be strongly related to Quasi-Biennial Oscillation. October and November rainfall are found to be coupled with ENSO fluctuations, and on average, more than 900 mm more rainfall is observed per month over all stations during El Ni?o than during La Ni?a years. Analysis of relationships between the observed meso-scale rainfall regions and the Sea Surface Temperatures (SSTs) in the Indian Ocean north of the Equator showed that the northern Indian Ocean can be divided into three particular regions based on similarity in the SST fluctuations: (a) a region with cool upwelling water, (b) non upwelling water, and (c) the Indian Ocean Warm Pool. We found that there are no statistically significant relationships between the observed SST regions in the Indian Ocean and the meso-scale precipitation patterns in Sri Lanka.     

斯里兰卡南部西向沿岸流在西南季风期间的动力学研究

北半球夏季,北印度洋环流主要受到西南季风流控制,将热带印度洋水体从西向东进行跨海盆输运,然而在斯里兰卡南部沿岸存在一支与西南季风流方向相反的西向沿岸流,即南斯里兰卡沿岸流（SSLCC）.本文主要利用ECCO2资料进行南斯里兰卡沿岸流的动力学特征研究.结果表明,SSLCC的形成和孟加拉湾局地环流密切相关.当斯里兰卡穹顶区（SLD）环流偏强时,斯里兰卡南部形成局地气旋式涡旋,斯里兰卡东部沿岸流在SLD西部向南流动,随着气旋式涡旋北部转向西流形成强的SSLCC.相比之下,SLD较弱时,沿岸流仅存在斯里兰卡东部沿岸,斯里兰卡东部沿岸流无法向西转向,SSLCC和西南季风流一起向东流动,其可能的主要原因是局地风应力对SLD产生的强度影响.研究还表明,SLD强度对SSLCC流向和强度有着重要影响.     

Factors controlling January�CApril rainfall over southern India and Sri Lanka

Most of the annual rainfall over India occurs during the Southwest (June?CSeptember) and Northeast (October?CDecember) monsoon periods. In March 2008, however, Southern peninsular India and Sri Lanka received the largest rainfall anomaly on record since 1979, with amplitude comparable to summer-monsoon interannual anomalies. This anomalous rainfall appeared to be modulated at intraseasonal timescale by the Madden Julian Oscillation, and was synchronous with a decaying La Ni?a event in the Pacific Ocean. Was this a coincidence or indicative of a teleconnection pattern? In this paper, we explore factors controlling rainfall over southern India and Sri Lanka between January and April, i.e. outside of the southwest and northeast monsoons. This period accounts for 20% of annual precipitation over Sri Lanka and 10% over the southern Indian states of Kerala and Tamil Nadu. Interannual variability is strong (about 40% of the January?CApril climatology). Intraseasonal rainfall anomalies over southern India and Sri Lanka are significantly associated with equatorial eastward propagation, characteristic of the Madden Julian Oscillation. At the interannual timescale, we find a clear connection with El Ni?o-Southern Oscillation (ENSO); with El Ni?os being associated with decreased rainfall (correlation of ?0.46 significant at the 98% level). There is also a significant link with local SST anomalies over the Indian Ocean, and in particular with the inter-hemispheric sea surface temperature (SST) gradient over the Indian Ocean (with colder SST south of the equator being conducive to more rainfall, correlation of 0.55 significant at the 99% level). La Ni?as/cold SSTs south of the equator tend to have a larger impact than El Ni?os. We discuss two possible mechanisms that could explain these statistical relationships: (1) subsidence over southern India remotely forced by Pacific SST anomalies; (2) impact of ENSO-forced regional Indian Ocean SST anomalies on convection. However, the length of the observational record does not allow distinguishing between these two mechanisms in a statistically significant manner.     

Application of the Ricardian Technique to Estimate the Impact of Climate Change on Smallholder Farming in Sri Lanka

This study applies the Ricardian technique to estimate the effect of climate change on the smallholder agriculture sector in Sri Lanka. The main contribution of the paper is the use of household-level data to analyze long-term climate impacts on farm profitability. Household-level data allows us to control for a host of factors such as human and physical capital available to farmers as well as adaptation mechanisms at the farm level. We find that non-climate variables explain about half the variation in net revenues. However, our results suggest that climate change will have a significant impact on smallholder profitability. In particular, reductions in precipitation during key agricultural months can be devastating. At the national level, a change in net revenues of between −23% and +22% is likely depending on the climate change scenario simulated. These impacts will vary considerably across geographic areas from losses of 67% to gains that more than double current net revenues. The largest adverse impacts are anticipated in the dry zones of the North Central region and the dry zones of the South Eastern regions of Sri Lanka. On the other hand, the intermediate and wet zones are likely to benefit, mostly due to the predicted increase in rainfall.     

Persistence in rainfall occurrence over Tropical south-east Asia and equatorial Pacific

Summary Daily rainfall observations during the principal rainy seasons over a large part of Tropical Asia and the equatorial Pacific are analysed for persistence by fitting Markov chains of various order. Daily rainfall data of 98 stations from India, Sri Lanka and Thailand falling in the monsoonal regime and 9 stations in the non-monsoonal regime of the equatorial Pacific are examined.The appropriate order of Markov chain is determined by analyzing wet and dry spell length characteristics and by applying the Schwarz Baysian Criterion to the arbitrary sequences of 5-day length. Markov chains of order greater than 1 are found to characterize the persistence in rainfall over India and to some extent over wet zones of Sri Lanka and central equatorial Pacific. Simple Markov chains are suggested for Thailand, the dry zone of Sri Lanka and the stations of central equatorial Pacific lying some what away from the equator.With 5 Figures     

An objective classification of homogeneous rainfall regimes in Sri Lanka

Summary Homogeneous rainfall regimes in Sri Lanka have been devised through multivariate statistical techniques. The nature of previous subjective classifications are reviewed and the need for an objective classification is stressed. Each of the major classifications of rainfall regimes, micro, meso and macro, throw new light on our understanding of the climates of Sri Lanka. New boundaries between dry/wet and dry/arid zones have emerged and are compared with previous examples. Three rainfall dominance regimes are also identified. Finally a brief account of the micro scale rainfall regimes is given.With 5 Figures     

Rainfall Variations of Sri Lanka, Part 2: regional gluctuations

Summary Based on the regional division of another paper [39], the rainfall variations of Sri Lanka have been investigated for the respective regions by power spectrum analysis and filtering methods, making use of data for the period from 1881 to 1980. The 3–4 year periodicity was observed over the entire island, but other cycles differ from region to region. The 13–16 months oscillation arises in Regions A, D and E, which roughly correspond to the Wet Zone and Dry Zone. The 10 and 2 year oscillations emerge in Regions A, B and C, where the southwest monsoon dominates the fluctuation patterns. In particular, it was confirmed that the quasi-biennial oscillation is not only in Sri Lanka, but also in other low latitude countries. The quasi-five year oscillation is noticed in Regions D and E, where the northeast monsoon influences on the fluctuation patterns. Irregularities in amplitude and in phase changes were noticed in their longer period fluctuation.

---

Variationen des Niederschlags in Sri Lanka. Teil 2: Regionale Fluktuationen

Zusammenfassung Aufgrund einer früher vorgenommenen regionalen Einteilung [39] wurden die Niederschlagsvariationen in Sri Lanka für die einzelnen Regionen mit Spektralanalyse und Filterungsmethoden unter Verwendung von Beobachtungsdaten aus der Periode 1881–1980 untersucht. Die 3- bis 4jährige Periodizität wurde auf der ganzen Insel festgestellt; aber andere Zyklen unterscheiden sich von Region zu Region. Eine 13- bis 16monatige Oszillation zeigt sich in den Regionen A, D und E, die ungefähr der Feuchtzone und der Trockenzone entsprechen. 10- bis 2jährige Oszillationen treten in den Regionen A, B und C auf, wo der Südwestmonsun die Fluktuationsformen beherrscht. Im besonderen wurde festgestellt, daß die quasi-zweijährige Oszillation nicht nur in Sri Lanka, sondern auch in anderen Ländern niedriger Breiten vorkommt. Eine quasifünfjährige Oszillation wurde in den Regionen D und E festgestellt, wo der Nordostmonsun die Fluktuationsformen beeinflußt. Unregelmäßigkeiten in der Amplitude und in Phasenänderungen wurden in ihrer längerperiodischen Fluktuation festgestellt.

---

---

With 4 Figures     

Impacts of absorbing aerosols on South Asian rainfall

Anthropogenic aerosols in the lower troposphere increase the absorption and scattering of solar radiation by air and clouds, causing a warmer atmosphere and a cooler surface. It is suspected that these effects contribute to slow down the hydrological cycle. We conducted a series of numerical experiments using a limited area atmospheric model to understand the impacts of aerosol radiative forcing on the rainfall process. Experiments with different radiative conditions under an idealized setting revealed that increasing atmospheric forcing and decreasing surface forcing of radiation causes reductions in rainfall. There was no relationship of top of the atmosphere forcing to the rainfall yield. The model was then used to simulate a domain covering southern part of Sri Lanka, over for the period from November 2002 to July 2003. For a given radiative forcing, instances with lower rainfall yields showed larger fractional reductions in rainfall. The trends in seasonal rainfall observed over the site in past 30 years in a different study confirms this finding. We conclude that the negative impact of increase of anthropogenic aerosols on rainfall would be more severe on regions and seasons with lower rainfall yields. The consequences of this problem on the industries that critically depend on well-distributed rainfall like non-irrigated agriculture and on the general livelihood of societies in low-rain areas can be serious.     

夏季印度季风环流系统与我国长江上游区域性持续暴雨的关系

本文对夏季6—8月我国长江上游地区56次区域性持续暴雨过程进行了统计分析。指出它们与夏季印度季风环流系统之间存在着密切的联系。给出暴雨期间印度—孟加拉湾地区的温压场和有关气象要素分布的平均形势,计算了暴雨日雨量与其前24h气象要素的相关系数,从而说明长江上游暴雨与印度季风环流系统的相关关系,并指出主要的季风环流影响系统和关键区。     

Relationship between North Pacific atmospheric pressure variations and Indian Monsoon rainfall

Summary Analysis of mean sea level pressure (1925 to 1988) over the North Pacific Ocean (NPP) for the winter period (November to March) revealed a significant correlation with Indian Monsoon rainfall during the later period. Its correlation coefficients (CC) for different periods (during 1951–1988) are significant at the 1% to 5% levels. The temporal stability of these CCs is examined using 11, 21 and 31 year sliding windows. NPP is seen to play an important role in the regression models as revealed by the relative significance of its partial regression coefficients. The regression models developed are seen to perform well for the independent period.With 5 Figures     

Pacific Ocean sea-surface temperature variability and predictability of rainfall in the early and late parts of the Indian summer monsoon season

For central India and its west coast, rainfall in the early (15 May–20 June) and late (15 September–20 October) monsoon season correlates with Pacific Ocean sea-surface temperature (SST) anomalies in the preceding month (April and August, respectively) sufficiently well, that those SST anomalies can be used to predict such rainfall. The patterns of SST anomalies that correlate best include the equatorial region near the dateline, and for the early monsoon season (especially since ~1980), a band of opposite correlation stretching from near the equator at 120°E to ~25°N at the dateline. Such correlations for both early and late monsoon rainfall and for both regions approach, if not exceed, 0.5. Although correlations between All India Summer Monsoon Rainfall and typical indices for the El Ni?o-Southern Oscillation (ENSO) commonly are stronger for the period before than since 1980, these correlations with early and late monsoon seasons suggest that ENSO continues to affect the monsoon in these seasons. We exploit these patterns to assess predictability, and we find that SSTs averages in specified regions of the Pacific Ocean in April (August) offer predictors that can forecast rainfall amounts in the early (late) monsoon season period with a ~25% improvement in skill relative to climatology. The same predictors offer somewhat less skill (~20% better than climatology) for predicting the number of days in these periods with rainfall greater than 2.5?mm. These results demonstrate that although the correlation of ENSO indices with All India Rainfall has decreased during the past few decades, the connections with ENSO in the early and late parts have not declined; that for the early monsoon season, in fact, has grown stronger in recent decades.     

A climatological perspective of water vapor at the UTLS region over different global monsoon regions: observations inferred from the Aura-MLS and reanalysis data

The Aura-MLS observations of eight years from 2004 to 2011 have been utilized to understand the hydration and the dehydration mechanism over the northern and the southern hemispheric monsoon (NH and SH) regions. The monsoon regions considered are the Asian Summer Monsoon, East Asian Summer Monsoon, Arizona Monsoon (AM), North African Monsoon, South American Monsoon and the Australian Monsoon. The annual cycle of water vapor as expected shows maxima over the NH during June–August and during December–February over the SH. The time taken by the air parcels over the NH monsoon regions is found to be different compared to that over the SH monsoon regions. The analysis shows the concentration of water vapor in the upper troposphere and the lower stratosphere (UTLS) has not changed over these eight years in both the hemispheres during their respective monsoon seasons. The present analysis show different processes viz., direct overshooting convection, horizontal advection, temperature and cirrus clouds in influencing the distribution of water vapor to the UTLS over these different monsoon regions. Analysis of the UTLS water vapor with temperature and ice water content shows that the AM is hydrating the stratosphere compared to all the other monsoon regions where the water vapor is getting dehydrated. Thus it is envisaged that the present results will have important implications in understanding the exchange processes across the tropopause over the different monsoon regions and its role in stratosphere chemistry.     

Equitable resilience in flood prone urban areas in Sri Lanka: A case study in Colombo Divisional Secretariat Division

Amid increasing flood incidences and damages in many parts of the world, the fundamental question arises as to the extent to which poor and marginalized residents can manage disasters by receiving equitable, fair and just support. This paper seeks to examine this question by focusing on a poor and vulnerable area of Colombo, the capital of Sri Lanka. Administratively this area is called Colombo Divisional Secretariat Division (DSD). Here mainly low-income residents live in congested housing conditions with narrow streets and poor drainage management. For years, this Division was regarded as one of the most flood vulnerable areas of Sri Lanka. To understand the resaons, we conducted our field research in this area and interviewed DSD officials and local people with three key equitable resilience dimensions in mind: distributive, procedural and contextual equities. We found that the intensity and frequency of rainfalls had increased in the area, but the residents had not received any flood protection support from the government due largely to some legal and socio-political complications. Many expressed their fear of the next flood incident. As these residents were without legal land ownership the government did not pay much attention to their needs. We also found that a low education level and a lack of political representation led to the marginalization of people in this area. Using information we collected at the Colombo DSD office and other relevant government agencies, we then examine a set of factors that are relevant to income and poverty level, population density, quality of housing, education, infrastructure and participatory decision making. The results show that flood loss and damage risks were heightened by such social vulnerability factors as low income, an unaffordability of flood resilient houses and an absence of policy implementations for flood resilient infrastructure. We also found that a lack of community leadership led to poor participation in decision making. This paper then highlights the important area for mainstreaming equitable community resilience actions.     

斯里兰卡穹顶区的形成和演变机制分析

通过SODA再分析资料和AVISO观测资料研究了斯里兰卡穹顶区（SLD）的迁移和消散机制.斯里兰卡穹顶区是孟加拉湾西南部的一个气旋涡旋,主要出现在西南季风（5-9月）期间,与西南季风海流侵入孟加拉湾同时存在.正风应力旋度引起的Ekman抽吸是形成SLD的主要原因.回归分析结果表明SLD区域的风应力旋度与Ekman抽吸存在较强的正相关（r²=0.93,p>0.5）.此外,结果表明SLD在发展过程中的移动主要受正风应力旋度移动的影响,SLD的消减与该正风应力旋度减弱和西传的暖Rossby波有关,而冷Rossby波的传播有益于SLD的发展.在SLD消减时期,孟加拉湾涡旋（BBD）独立发展并进一步与SLD融合,回归分析发现BBD区域的Ekman抽吸与当地风应力旋度的关系密切（r²=0.76,p>0.5）,这表明了BBD在形成阶段由局地的风应力主导.9月之后,风应力旋度减弱,BBD和SLD开始了合并过程.动力方面,EKE分析显示SLD衰退的同时,BBD的EKE大幅增加;热力方面,10-11月时,由Ekman抽吸引起的SLD和BBD次表层冷水汇合,清晰地表明了二者之间的热动力学联系.     

Hiatus of wetland methane emissions associated with recent La Niña episodes in the Asian monsoon region

The variability of methane emissions from wetlands in the tropics and northern temperate regions can explain more than 70% of the interannual variation in global wetland methane emissions, which are largely driven by climate variability. We use climate reanalysis, remote sensing wetland area dataset and simulations from 11 land models contributing to Global Methane Budget to investigate the interannual variation and anomalies of wetland methane emissions in the Asian Monsoon region. Methane emissions in this region steadily increased over 2000–2012. However, abnormally low methane emissions were found in equatorial fully humid (Af), warm temperate winter dry (Cw), and warm temperate fully humid (Cf) Asian Monsoon climate sub-regions in 2008, 2009 and 2011, respectively. These spatially-shifting low emissions occurred simultaneously with observed wetland area shrinkage due to abnormally low precipitation. Interannual variability of wetland methane emissions in Asian Monsoon region are primarily driven by South Asian Monsoon system. However, the abnormally low emissions are related to strong La Niña events, and its accompanying effect of weakened East Asian Monsoon system and eastward Western Pacific subtropical high, which drives the shifting pattern of rainfall, and thus the spatial pattern of methane emission anomalies.     

AGCM simulations of intraseasonal variability associated with the Asian summer monsoon

The intraseasonal variability associated with the Asian summer monsoon as simulated by a number of atmospheric general circulation models (AGCMs) are analyzed and assessed against observations. The model data comes from the Monsoon GCM Intercomparison project initiated by the CLIVAR/Asian–Australian Monsoon Panel. Ten GCM groups, i.e., the Center for Ocean–Land–Atmosphere Studies (COLA), Institute of Numerical Mathematics (DNM), Goddard Space Flight Center (GSFC), Geophysical Fluid Dynamics Laboratory (GFDL), Institute of Atmospheric Physics (IAP), Indian Institute of Tropical Meteorology (IITM), Meteorological Research Institute (MRI), National Center for Atmospheric Research (NCAR), Seoul National University (SNU), and the State University of New York (SUNY), participated in the intraseasonal component of the project. Each performed a set of 10 ensemble simulations for 1 September 1996–31 August 1998 using the same observed weekly SST values but with different initial conditions. The focus is on the spatial and seasonal variations associated with intraseasonal variability (ISV) of rainfall, the structure of each model's principal mode of spatial-temporal variation of rainfall [i.e. their depiction of the Intraseasonal Oscillation (ISO)], the teleconnection patterns associated with each model's ISO, and the implications of the models' ISV on seasonal monsoon predictability. The results show that several of the models exhibit ISV levels at or above that found in observations with spatial patterns of ISV that resemble the observed pattern. This includes a number of rather detailed features, including the relative distribution of variability between ocean and land regions. In terms of the area-averaged variance, it is found that the fidelity of a model to represent NH summer versus winter ISV appears to be strongly linked. In addition, most models' ISO patterns do exhibit some form of northeastward propagation. However, the model ISO patterns are typically less coherent, lack sufficient eastward propagation, and have smaller zonal and meridional spatial scales than the observed patterns, and are often limited to one side or the other of the maritime continent. The most pervasive and problematic feature of the models' depiction of ISV and/or their ISO patterns is the overall lack of variability in the equatorial Indian Ocean. In some cases, this characteristic appears to result from some models forming double convergence zones about the equator rather than one region of strong convergence on the equator. This shortcoming results in a poor representation of the local rainfall pattern and also significantly influences the models' representations of the global-scale teleconnection patterns associated with the ISO. Finally, analysis of the model ensemble shows a positive relationship between the strength of a model's ISV of rainfall and its intra-ensemble variability of seasonal monsoon rainfall. The implications of this latter relation are discussed in the context of seasonal monsoon predictability.