Secular spring rainfall variability at local scale over Ethiopia: trend and associated dynamics

Spring rainfall secular variability is studied using observations, reanalysis, and model simulations. The joint coherent spatio-temporal secular variability of gridded monthly gauge rainfall over Ethiopia, ERA-Interim atmospheric variables and sea surface temperature (SST) from Hadley Centre Sea Ice and SST (HadISST) data set is extracted using multi-taper method singular value decomposition (MTM-SVD). The contemporaneous associations are further examined using partial Granger causality to determine presence of causal linkage between any of the climate variables. This analysis reveals that only the northwestern Indian Ocean secular SST anomaly has direct causal links with spring rainfall over Ethiopia and mean sea level pressure (MSLP) over Africa inspite of the strong secular covariance of spring rainfall, SST in parts of subtropical Pacific, Atlantic, Indian Ocean and MSLP. High secular rainfall variance and statistically significant linear trend show consistently that there is a massive decline in spring rain over southern Ethiopia. This happened concurrently with significant buildup of MSLP over East Africa, northeastern Africa including parts of the Arabian Peninsula, some parts of central Africa and SST warming over all ocean basins with the exception of the ENSO regions. The east-west pressure gradient in response to the Indian Ocean warming led to secular southeasterly winds over the Arabian Sea, easterly over central Africa and equatorial Atlantic. These flows weakened climatological northeasterly flow over the Arabian Sea and southwesterly flow over equatorial Atlantic and Congo basins which supply moisture into the eastern Africa regions in spring. The secular divergent flow at low level is concurrent with upper level convergence due to the easterly secular anomalous flow. The mechanisms through which the northwestern Indian Ocean secular SST anomaly modulates rainfall are further explored in the context of East Africa using a simplified atmospheric general circulation model (AGCM) coupled to mixed-layer oceanic model. The rainfall anomaly (with respect to control simulation), forced by the northwestern Indian Ocean secular SST anomaly and averaged over the 30-year period, exhibits prevalence of dry conditions over East and equatorial Africa in agreement with observation. The atmospheric response to secular SST warming anomaly led to divergent flow at low levels and subsidence at the upper troposphere over regions north of 5^° S on the continent and vice versa over the Indian Ocean. This surface difluence over East Africa, in addition to its role in suppressing convective activity, deprives the region of moisture supply from the Indian Ocean as well as the Atlantic and Congo basins.     

Modification of the southern African rainfall variability/ENSO relationship since the late 1960s

Analysis of 149 raingauge series (1946–1988) shows a weak positive correlation between late summer rainfalls (January–March) in tropical southern Africa and the Southern Oscillation Index (SOI). The correlation coefficients have been unstable since World War II. They were close to zero before 1970 and significant thereafter. Before 1970, southern African late summer rainfalls were more specifically correlated with regional patterns of sea surface temperature (SST), mainly over the southwestern Indian Ocean. After 1970, teleconnections with near global SST anomaly patterns, i.e. over the central Pacific and Indian oceans, dominate the regional connections. The increase in the sensitivity of the southern African rainfall to the global SO-related circulation anomalies is simultaneous with the correlation between SOI and more extensive SST anomalies, particularly over the southern Indian Ocean. This feature is part of longer term (decadal), global SST variability, as inferred from statistical analyses. Numerical experiments, using the Météo-France general circulation model ARPEGE-Climat, are performed to test the impact of the observed SST warming in the southern Indian and extratropical oceans during El Niño Southern Oscillation (ENSO) events on southern African rainfall. Simulated results show that ENSO events, which occurred in the relatively cold background of the pre-1970 period in the southern oceans, had a little effect on southern Africa climatic conditions and atmospheric circulation. By contrast, more recent ENSO events, with warmer SST over the southern oceans, lead to a climatic bipolar pattern between continental southern African and the western Indian Ocean, which is characterized by reduced (enhanced) deep convection and rainfall over the subcontinent (the western Indian Ocean). A weaker subtropical high-pressure belt in the southwestern Indian Ocean is also simulated, along with a reduced penetration of the moist southern Indian Ocean trade winds over the southern African plateau. These results are consistent with the strong droughts observed over all southern Africa during ENSO events since 1970.     

Moisture transport between the South Atlantic Ocean and southern Africa: relationships with summer rainfall and associated dynamics

Moisture exchange between the South Atlantic and southern Africa is examined in this study through zonal moisture transport. Along the west coast of southern Africa, a multivariate analysis of the zonal flow of moisture computed from NCEP-DOE AMIP II Re-analyses reveals a primary mode of variability typical of variations in intensity and of the latitudinal migration of the circulation associated with the midlatitude westerlies and the South Atlantic anticyclone. In austral summer (January–February), this mode, referred to as the South Atlantic midlatitude mode, is found to be well correlated with rainfall over southern Africa (i.e. to the south of the upper lands surrounding the Congo basin). Its positive/negative phases are found to correspond with surface pressures changes over the South Atlantic region in austral summer when the South Atlantic anticyclone is shifted northward/southward respectively. Such changes are accompanied by dipole-like SST anomalies in the midlatitude South Atlantic Ocean, while simultaneous SST anomalies with a similar structure are also found over South Indian Ocean regions. In January–February, positive/negative events linked to the South Atlantic midlatitude mode are marked by meridional shifts (northward/southward) and weakening/strengthening of the ITCZ over the southern tropics, together with modulations in intensity (weakened/sustained) of the Angola low, which could act as a tropical source of moisture for Tropical Temperate Troughs (TTTs). In association with a strengthened/weakened zonal component of the southern extension of the African Easterly Jet (AEJ), this could modulate the meridional transfer of moisture south of 15°S to the advantage/detriment of Angolan coastal regions, where above/below rainfall are expected. Variations in the latitudinal position (northward/southward) of the South Atlantic anticyclone, and thus of the midlatitude westerlies, are also found to reduce/favour moisture advection towards southern Africa subtropics allowing the southern Indian trades to penetrate less/more over the subcontinent south of 25°S. This would create a situation where convection processes are inhibited/supported within the SICZ/TTTs region resulting in drier/wetter conditions locally for positive/negative events respectively.     

沈阳市城乡梯度湿地水体碳氮磷含量变化研究

湿地是人类生存和发展的重要环境之一,然而城镇化在带动经济发展的同时也严重影响了湿地生态系统。本文利用城乡梯度研究方法结合湿地分布特征,设置从沈阳城市中心到城市边缘的研究样带,沿着城镇密集区（浑河）—郊区（蒲河）—乡村（卧龙湖,仙子湖）梯度带选取典型湖泊和河流湿地样地为研究对象,系统分析城镇化对湿地水体碳氮磷含量的影响。研究发现湿地水体碳氮磷含量与城乡梯度具有相关性,总碳（TC）和可溶性有机碳（DOC）含量及SUVA₂₈₀值沿城镇密集区—郊区—乡村梯度带逐渐增加,即远离市中心的乡村卧龙湖湿地最高,分别为（120.68±2.34）mg/L,（41.56±6.27）mg/L和（0.35±0.10）L/(mg·m),显著高于位于四环以外乡村仙子湖湿地、流经沈阳市四环蒲河和穿越三环浑河湿地水体。湿地水体氮磷含量总体上沿着城乡梯度带呈降低趋势,但蒲河湿地水体总氮（TN）和总磷（TP）含量最高,分别为（5.35±0.19）mg/L和（1.45±0.07）mg/L,显著高于位于城镇密集区的浑河湿地水体。城镇化作用总体上增加了湿地水体铵态氮（NH₄⁺-N）和硝态氮（NO₃^--N）含量,其中,横穿城镇密集区的浑河湿地水体NH₄⁺-N含量最高,为（1.28±0.14）mg/L;NO₃^--N含量则是位于郊区的蒲河湿地水体最高,为(1.42±0.15) mg/L。研究结果表明城镇化改变了湿地生态系统水体碳氮磷含量,使水体DOC含量降低,小分子化合物增多,不利于DOC在水体中的累积;同时由于人类活动的加剧也使氮磷排放增加,使水体氮磷超标,导致流经城镇密集区部分的浑河和蒲河湿地NH₄⁺-N和NO₃^--N含量高于乡村。未来随着城镇化的不断发展,应严加控制和合理规划,防止城镇化导致的湿地水体污染和生态系统的破坏。     

塔里木盆地夏季降水年代际的变化及联系的大尺度环流[*]

基于1961-2020年夏季塔里木盆地33站逐日降水数据和NCEP/NCAR大气环流再分析数据,分析了塔里木盆地夏季降水的年代际变化特征及影响环流。结果表明,塔里木盆地夏季降水在1986/1987年发生了由少到多的年代际突变,降水显著增加的区域主要位于盆地的西部和北部,主要由降水日数的增加贡献。不同年代际背景下,影响塔里木盆地夏季降水的环流配置存在一定差异。1961-1986年,影响盆地夏季降水的中亚副热带急流位置显著南移,中亚上空的异常气旋位于40°N附近,水汽源于阿拉伯海;1987-2020年,影响盆地夏季降水的中亚上空异常气旋位置位于40°N以南,水汽源于孟加拉湾和西北太平洋。东大西洋—西俄罗斯大气遥相关型在塔里木盆地夏季降水的年代际变化中扮演了重要角色。     

The effect of the subtropical jet on the rainfall over southern China in January 2008

ABSTRACT The third precipitation episode of China＇s great snowstorms of 2008 was analyzed using station observations and ECMWF six-hourly data. The variation of the shape of the upper-level subtropical jet played an important role in the rainfall over south- ern China. With the eastward movement of the trough, the jet shape changed from two straight jets to a tilting jet over China and then it moved southward. With these variations, the south-north movement of ascending flow and precipitation area over southern China occurred.     

Twentieth century simulation of the southern hemisphere climate in coupled models. Part 1: large scale circulation variability

The ability of five, global coupled climate models to simulate important atmospheric circulation characteristics in the Southern Hemisphere for the period 1960–1999 is assessed. The circulation features examined are the Southern Hemisphere annular mode (SAM), the semi-annual oscillation (SAO) and the quasi-stationary zonal wave 3 (ZW3). The models assessed are the National Center for Atmospheric Research Community Climate System Model Version 3 (CCSM3), the Commonwealth Scientific and Industrial Research Organisation Mark 3, the Geophysical Fluid Dynamics Laboratory Model, the Goddard Institute for Space Studies Model ER (GISS-ER) and the UK Meteorological Office Hadley Center Coupled Model Version 3. The simulations were compared to the NCAR–NCEP reanalyses. The models simulate a SAO which differs spatially from the observed over the Pacific and Indian oceans. The amplitudes are too high over the southern ocean and too low over the midlatitudes. These differences are attributed to a circumpolar trough which is too deep and extends too far north, and to the inability of the models to simulate the middle to high latitude temperature gradient. The SAM is well-represented spatially by most models but there are important differences which may influence the flow over the Pacific and in the region extending from the Ross to Weddell Seas. The observed trend towards positive polarity in the SAM is apparent in the ensemble averages of the GISS-ER and CCSM3 simulations, suggesting that the trend is due to external forcing by changes in the concentration of ozone and greenhouse gases. ZW3 is well-represented by the models but the observed trend towards positive phases of ZW3 is not apparent in the simulations suggesting that the observed trend may be due to natural variability, not external forcing.     

Decadal scale oscillations and trend in the Indian monsoon rainfall

The emerging need for extended climate prediction requires a consideration of the relative roles of climate change and low-frequency natural variability on decadal scale. Addressing this issue, this study has shown that the variability of the Indian monsoon rainfall (IMR) consists of three decadal scale oscillations and a nonlinear trend during 1901–2004. The space–time structures of the decadal oscillations are described. The IMR decadal oscillations are shown to be associated with Atlantic Multidecadal Oscillation (AMO), Atlantic tripole oscillation and Pacific Decadal Oscillation (PDO). The sea surface temperatures (SSTs) of the North Pacific and North Atlantic Oceans are also resolved as nonlinear decadal oscillations. The SST AMO mode has high positive correlation with IMR while the SST tripole mode and SST PDO have negative correlation. The trend in IMR increases during the first half of the period and decreases during the second half. The IMR trend is modified when combined with the three decadal oscillations.     

Seasonality and time scales in the relationship between global SST and African rainfall

The main goal of this study is to determine the oceanic regions corresponding to variability in African rainfall and seasonal differences in the atmospheric teleconnections. Canonical correlation analysis (CCA) has been applied in order to extract the dominant patterns of linear covariability. An ensemble of six simulations with the global atmospheric general circulation model ECHAM4, forced with observed sea surface temperatures (SSTs) and sea ice boundary variability, is used in order to focus on the SST-related part of African rainfall variability. Our main finding is that the boreal summer rainfall (June–September mean) over Africa is more affected by SST changes than in boreal winter (December–March mean). In winter, there is a highly significant link between tropical African rainfall and Indian Ocean and eastern tropical Pacific SST anomalies, which is closely related to El Niño-Southern Oscillation (ENSO). However, long-term changes are found to be associated with SST changes in the Indian and tropical Atlantic Oceans, thus, showing that the tropical Atlantic plays a critical role in determining the position of the intertropical convergence zone (ITCZ). Since ENSO is less in summer, the tropical Pacific and the Indian Oceans are less important for African rainfall. The African summer monsoon is strongly influenced by SST variations in the Gulf of Guinea, with a response of opposite sign over the Sahelian zone and the Guinean coast region. SST changes in the subtropical and extratropical oceans mostly take place on decadal time scales and are responsible for low-frequency rainfall fluctuations over West Africa. The modelled teleconnections are highly consistent with the observations. The agreement for most of the teleconnection patterns is remarkable and suggests that the modelled rainfall anomalies serve as suitable predictors for the observed changes.     

中国南方夏季低频雨型特征及其年代际变化研究

为了揭示南方低频降水的季节内及年代际变化规律,利用1961—2011年夏季中国南方逐日降水资料和大气环流再分析资料,运用经验正交函数分解(EOF)与聚类分析相结合方法,将南方地区10 d以上的低频尺度降水划分为5个异常雨型.分析表明,各雨型具有较好的持续性,与同期的环流场有很好的对应关系,彼此间水汽异常输送特征有明显差别.通过对雨型的统计特征进行分析,指出不同于以往的季节平均雨型,低频雨型可在季节内交替出现,反映了大尺度异常降水的低频演变特征.进一步研究发现,近50年来各低频雨型的多项统计特征存在显著的年代际变化:20世纪80—90年代各多雨型均相继出现峰值,进入21世纪偏旱型的各项统计特征存在明显上升趋势,而大部分低频多雨型出现次数减少,这可能预示着中国南方地区正在进入一个季节内尺度降水减少的阶段.     

Effects of convective scale downdrafts on the rainfall simulation in NCAR-CAM3

In this work, an attempt is made to systematically evaluate the effect of convective scale downdrafts on the model-simulated rainfall, in both aqua- and actual-planet frameworks, using the NCAR CAM3. From the aqua-planet simulations, it was found that there is a reduction in the total rainfall (TRF) with increase in the intensity of downdrafts, which is primarily attributed to the reduction in the deep convective component (DRF). However, with stronger downdrafts, the shallow convective and the large-scale components (SRF and LRF, respectively) are found to increase. The reduction in DRF is due to the increased evaporation of convective precipitation within the downdrafts. It is found that, with intense downdrafts, there is an increase in relative humidity throughout the troposphere, due to the combined effect of both moisture and temperature. There is an overall increase in specific humidity of the atmosphere with stronger downdrafts, excepting at around the 900-hPa level. In addition, there is a reduction in temperature throughout the troposphere, primarily due to the reduction in the overall temperature tendency due to moist processes and that due to the radiative processes. The changes in the radiative forcing are found to be primarily due to a significant increase in the low cloud fraction with strong downdrafts. In the actual-planet framework, it is seen that, with strong convective downdrafts, there is a reduction in TRF and DRF and a corresponding increase in SRF and LRF, similar to the results obtained from the aqua-planet simulations. The vertical structures of the thermodynamic variables (RH, q, and T) show similar sensitivity to the downdraft intensity as that seen in the aqua-planet framework. Sensitivity of frequency and intensity of model-simulated rainfall to the downdraft intensity was also analyzed, and it was seen that there were significant differences in the frequency distribution of rainfall. It was seen that, with an increase in downdraft intensity, there is an increase in the frequency of light rain (1–10?mm/day) for TRF with a corresponding reduction in all other rainfall bins. A similar behavior was seen for the DRF as well, while the SRF and LRF components showed an increase in rainfall accumulation in all the bins. In addition, the impact of convective downdrafts on the mean spatial pattern of rainfall is also analyzed, for the DJF and JJA periods (boreal winter and summer, respectively). For the DJF period, with strong downdrafts, it was seen that grossly over the whole domain, there were a reduction in DRF and an increase in SRF and LRF. In contrast, during JJA, although a major part of the domain showed a reduction in DRF, there were regions like western Arabian Sea and the Somali coast with increase in DRF with intense downdrafts. The SRF and LRF components, however, show a spatially homogeneous increase over almost the entire domain with increase in downdraft intensity.     

Monsoon rainfall and southern oscillation responses in the pressures over the Northern Indian Ocean

The pressure variations over the North Indian Ocean during the summer monsoon season have been exam-ined using the monthly data from June to September for the period 1961 to 1968. It is found that these varia-tions can be described by two significant eigenvectors (EV1 and EV2) which together account for 53% of the total variance.The first eigenvector (EV1) represents in phase variation over both, the Arabian Sea and the Bay of Bengal with higher variations over the northern side of the area. The second eigenvector (EV2) depicts the out-of-phase variation between the pressure anomalies over the north and the south of 15°N latitude with two areas of pronounced variation, viz., the head Bay of Bengal and the equatorial region near 65°E longitude.The coefficients of EV1 show significant association with rainfall of West Coast and Central India for the concurrent months. These coefficients also show significant association with the pressure and temperature indices of the Southern Oscillation. The coefficients of EV2 show significant association with the monsoon rainfall of south peninsular India.     

攀西地区南部年降雨量与海温指数的典型相关分析

本文通过典型相关分析研究了海温指数与攀西地区南部年降雨量的相关关系,结果表明年降雨量趋势与NINO3呈较大正相关,两者有密切的联系.     

攀西地区南部年降雨量与海温指数的典型相关分析

本文通过典型相关分析研究了海温指数与攀西地区南部年降雨量的相关关系,结果表明年降雨量趋势与NINO3呈较大正相关,两者有密切的联系。     

地表热通量对陕南强降水的影响

基于北京“7.21”特大暴雨个例,设计了一种考虑地形不确定性对降水影响的集合预报方案,在对该方案进行初步评估的基础上,重点通过计算相关系数、扰动总能量和尺度分解,对包括地形扰动方案在内的4种集合预报方案（初值、多物理、地形、初值-多物理混合）中离散度演变的异同性进行了分析。结果表明：（1）考虑模式地形不确定性的扰动方案,在不影响集合平均降水预报质量的基础上,对集合降水预报的离散度和概率预报略有正贡献。（2）离散度空间结构的演变与天气形势的演变密切相关。不同扰动方案产生的离散度在初始时刻的空间分布各不相同,但随模式向前积分其离散度的相似度快速增大,其中0-6 h内增长速度最快,离散度场之间的相关系数可以超过0.6。混合方案与单一扰动方案相比,对离散度空间结构的贡献不大。（3）虽然不同方案的离散度空间结构相似,但其幅度却存在明显的差异,如地形扰动方案的离散度幅度明显小于初值扰动和物理过程扰动方案。混合方案可以增加原单一扰动方案的离散度振幅,但这种增加在高层明显,而在近地面层并不明显,因而增加降水和其他近地面大气变量的离散度要比增加上层大气变量的离散度更困难。（4）尺度分离的结果表明,随着空间尺度的增大和积分时间的延长,不同扰动方法产生的离散度结构会逐渐变得相似,但在积分早期（<12 h）和较小的空间尺度（<448 km）上离散度结构的差异仍明显,并且在较小的空间尺度（<448 km）上,不同扰动方法产生的离散度幅度有明显的差异。所以对于小空间尺度系统或甚短期预报,选择扰动方案比大尺度和较长期的预报更重要。以上研究可为集合预报如何合理采用不同扰动方案或不同方案的组合提供科学依据。     

Correction to: The contributions of fronts,lows and thunderstorms to southern Australian rainfall

Climate Dynamics - In the original published version of the paper, the figures reported in Sect.&nbsp;4 relating the proportion of rainfall in southern Australia that is due to each of the...     

Effects of the southern oscillation on the probability for climatic categories of monthly rainfall, in a semi-arid region in the southern mid-latitudes

In many regions of the world, planning agricultural and water management activities is usually done based on probabilities for monthly rainfall, taking on values on specified intervals of values. These intervals of monthly rainfall amounts are commonly grouped into three categories: drought, normal rainfall, and abundant rainfall. Changes in the probabilities for occurrence of monthly rainfall amounts within these climatic rainfall categories will influence the decisions farmers and water managers will take (for example, crops to cultivate, flood preparedness, and operations of water reservoirs). This research explores the changes produced by the SO (Southern Oscillation) on the probability that the areal average of monthly rainfall (AAvMR) takes on values belonging to specified climatic rainfall categories. The semi-arid region under study is a major agricultural region in central Argentina; weather effects on agriculture in this region influence the world market of several crops. The evolution of the Southern Oscillation was divided into three phases: LSOI (low Southern Oscillation index phase, that includes ENSO events), NSOI (neutral SOI phase), and HSOI (high SOI phase that includes La Niña–SO events). The following are the criteria defining the three phases of the SO: (1) low SOI (ENSO), where the five-month moving average of the SO index, SOI, is less than −0.5 standard deviation during at least five consecutive months, and is equal to or less than −1 standard deviation during at least one month; (2) high SOI (La Niña–SO), where the SOI is greater than 0.5 standard deviation during at least five consecutive months, and is equal to or greater than 1 standard deviation during at least one month; and (3) neutral SOI (transition between extremes), where the SOI does not correspond to low SOI nor to high SOI. It was found that the Southern Oscillation influences the probability distribution of monthly rainfall only in four months of the year. Findings show that monthly rainfall has a complex response to the evolution of the SO. The response is not restricted to higher probability for occurrence of abundant rainfall or drought categories during low SOI (ENSO) or high SOI (La Niña–SO) episodes, respectively. The LSOI (ENSO) phase influences the AAvMR in several ways: depending on the month, it increases or decreases the probability of the abundant rainfall category. LSOI (ENSO) also increases or decreases, depending on the month, the probability of the normal rainfall category. It also decreases the probability that AAvMR takes on values in the drought category. A similar kind of complex response of monthly rainfall amounts occurs when the active phase is the HSOI (La Niña–SO). The responses are: (1) the probability of the category `drought' increases only in three months of the year, (2) increase or decrease of the probability of the normal rainfall category, depending on the month, and (3) decrease of the probability of the abundant rainfall category. Finally, the effects of NSOI (neutral phase of the SO) are not negligible. Depending on the month, NSOI episodes increase or decrease the probability of drought, or abundant rainfall, or normal rainfall categories.