Showing papers in "International Journal of Climatology in 1997"

Extension to the North Atlantic oscillation using early instrumental pressure observations from Gibraltar and south-west Iceland

Abstract: Early instrumental pressure measurements from Gibraltar and the Reykjavik area of Iceland have been used to extend to 1821 the homogeneous pressure series at the two locations In winter the two sites are located close to the centres of action that comprise the North Atlantic Oscillation (NAO). The extended 'winter half-year' record of the NAO enables recent changes in the record to be placed in the context of the period 1823-1996 The period since the early 1970s is the most prolonged positive phase of the oscillation and the late 1980s and early 1990s is the period with the highest values (strongest westerlies) The winter of 1995-1996 marked a dramatic switch in the index, with the change from 1994-1995 being the greatest change recorded from one year to the next since the series began in 1923. (The extended Gibraltar and Reykjavik monthly pressures and the NAO series can be found on the Climatic Research Unit home page. www.cru.uea-ac.uk.

Homogenization of swedish temperature data. part i: homogeneity test for linear trends

Abstract: A new test for the detection of linear trends of arbitrary length in normally distributed time series is developed With this test it is possible to detect and estimate gradual changes of the mean value in a candidate series compared with a homogeneous reference series The test is intended for studies of artificial relative trends in climatological time series, eg an increasing urban heat island effect The basic structure of the new test is similar to that of a widely used test for abrupt changes, the standard normal homogeneity test The test for abrupt changes is found to remain unaltered after an important generalization

The relationship of the el nino-southern oscillation to african rainfall

Abstract: This is a comprehensive study of the rainfall response over Africa to ENSO episodes in the Pacific. The harmonic method utilized by Ropelewski and Halpert is applied to 90 regionally averaged rainfall time series for the period 1901‐1990. The analysis was a composite of 20 episodes within this period. Seasons of maximum positive anomalies and maximum negative anomalies in the composite were identified. The method identifies 15 multiregion sectors where ENSO appears to modulate rainfall. The strongest signals are in eastern equatorial and south-eastern Africa. A continental-scale signal is also apparent. The magnitude, seasonal timing and duration, and consistency of the rainfall response to ENSO vary among the sectors and from episode to episode. The rainfall response is clearly seasonally specific. In general, the onset of the ENSO signal in rainfall commences far to the south and propagates latitudinally northward. For this reason, the equatorial regions are out-of-phase with the continental pattern. The ENSO mechanism is probably responsible for many of the well-established rainfall teleconnections over the continent, including the strong tendency for opposite anomalies in equatorial and southern Africa. There is a strong tendency for positive anomalies to occur during the first half of the ENSO cycle, negative during the second half. This corresponds to ‘cold’ and ‘warm’ phases in the adjacent Atlantic and Indian Oceans; continentally, rainfall tends to be enhanced during the cold phase, reduced during the warm phase. The northward propagation is most pronounced during the cold phase; a similar propagation and phase shift occurs at this time in the Atlantic. The rainfall anomalies of the warm phase are nearly constant in phase, as are the SST anomalies in the Indian Ocean. This suggests that, in general, the Atlantic Ocean controls rainfall during the cold phase, the Indian Ocean during the warm phase.

Rainfall variability over south-east Asia : Connections with Indian monsoon and enso extremes : New perspectives

Abstract: Seasonal and annual rainfall data for 135 stations for periods varying from 25 to 125 years are utilized to investigate and understand the interannual and short-term (decadal) climate variability over the South-east Asian domain Contemporaneous relations during the summer monsoon period (June to September) reveal that the rainfall variations over central India, north China, northern puts of Thailand, central parts of Brunet and Borneo and the Indonesian region east of 120 E vary in phase However, the rainfall variations over the regions surrounding the South China Sea, in particular the north-west Philippines, vary in the opposite phase. Possible dynamic causes for the spatial correlation structure obtained are discussed Based on the instrumental data available and on an objective criteria, regional rainfall anomaly time series for contiguous regions over Thailand, Malaysia, Singapore, Brunet, Indonesia and Philippines are prepared. Results reveal that although there are year-to-year random fluctuations, there are certain epochs of the above- and below-normal rainfall over each region These epochs are not forced by the El NiA±o La Nina frequencies Near the equatorial regions the epochs tend to last for about a decade, whereas over the tropical regions, away from the Equator, epochs last for about three decades There is no systematic climate change or trend in any of the series. Further, the impact of El NiA±o (La Nina) on the rainfall regimes is more severe during the below (above) normal epochs than during the above (below) normal epochs. Extreme drought flood situations tend to occur when the epochal behaviour and the El Nino La Nina events are phase-locked

Recent temperature variations in southern South America

Abstract: Results from a critical appraisal of surface mean air temperature in Chile and Argentina and extreme air temperature in Chile during the present century are presented. Observations were homogenized to produce a set of time series as reliable as possible. Linear trends computed for the period 1933–1992 resulted in warming rates from 1ċ3 to 2ċ0 °C 100 years−1; during the last three decades warming rates are twice as large. The generalized warming is not present around 41°S, where a cooling period from the 1950s to the 1970s prevails. Both positive and negative trends are due mostly to changes in minimum temperatures. The influence of El Niņo–Southern Oscillation on surface temperature along the Pacific South American coast from 18°S to 53°S was estimated and found to decrease southward. When its effect is extracted, warming trends become more uniform through time. In particular, the Southern Oscillation Index change around 1976 is felt in minimum temperatures at almost all stations, starting a period with higher values along the Chilean Pacific coast. Trend corrections for autocorrelation in the series introduce only small local changes. © 1997 by the Royal Meteorological Society.

An analysis of the enso signal in the tropical atlantic and western indian oceans

Abstract: This article examines the time-space evolution of the El Nino-Southern Oscillation (ENSO) signal in the tropical Atlantic and western Indian Oceans, using harmonic analysis. Composites of sea-surface temperatures (SSTs) and other variables are examined for a 24-month period beginning 6 months prior to the year of maximum warming in the Pacific (termed year 0). An ENSO signal is apparent in the Atlantic in six out of eight Pacific episodes and in the Indian Ocean in all eight episodes. Warming begins along the south-eastern Atlantic coast early in year 0, some months later elsewhere in the Atlantic and in the Indian Ocean. Maximum warming occurs in the Atlantic in October-December of year 0, but in the following January-March in the Indian Ocean. In these oceans a 'cold' phase occurs synchronously with the first half of the Pacific episode (July of year 1 to June of year 0, in the Rasmusson-Carpenter terminology), a 'warm' phase with the second half. Maximum cooling is 1 year prior to maximum warming in both oceans. In the Atlantic the cold phase occurs most consistently; in the Indian Ocean the warm phase occurs most consistently. There is a season-by-season reversal of SST anomalies and, to a lesser extent, pressure anomalies between the cold and warm phases. This is the basis for the biennial component of the ENSO signal. Our results indicate that the ENSO signal in African rainfall variability is a manifestation of ENSO's influence on SSTs in the Atlantic and Indian Oceans and, in turn, their influence on rainfall. The cold and warm phases correspond roughly to enhanced and reduced rainfall over the African continent, respectively. A similar reversal of rainfall anomalies is apparent season-by-season during these phases. The timing of the warming and cooling is relatively constant in the Indian Ocean. However, the onset of the warming and cooling in the south and equatorial Atlantic occurs progressively later from south to north, thus the signal 'propagates' northward. A similar propagation, synchronous in timing and latitude, is evident in ENSO induced rainfall anomalies over southern and equatorial Africa during the 'cold' phase but not during the 'warm' phase. This may be indicative of a switch in the control of rainfall from the Atlantic Ocean to the Indian Ocean during these phases. 1997 by the Royal Meteorological Society. Int. J. Climatol. 17, 345-375 (1997)

Trends in surface air temperature data over Turkey

Abstract: Eighteen stations are considered for the study of climatic records over Turkey in search for possible trends. Mean, unfiltered, seasonal and annual maximum and minimum temperatures are analysed using the Mann–Kendall rank statistic to demonstrate any existence of possible trends. The analyses indicate that the mean annual temperature records in Turkey have a warming trend over the 1939 to 1989 period, but a cooling trend from 1955 to 1989. These trends in mean annual temperatures, however, are not statistically significant. Comparatively greater warming effects have occurred in spring and winter minimum rather than the maximum temperature records. A regional increase in the mean minimum temperature around 1955 is attributed to the urban heat island effect. In general, general circulation models (GCM) predictions are consistent with the sign of the trends only in Turkish climate records during the entire 1939 to 1989 period © 1977 by the Royal Meteorological Society. Int. J. Climatol. 17: 511–520 (1997).

Precipitation in the british isles: an analysis of area‐average data updated to 1995

Abstract: This paper updates precipitation series forEngland and Wales, Scotland and Northern Ireland for 1990 and 1995 A new precipitation series for the whole of Ireland for 1840 to 1995 is developed. Redent changes in seasonal precipitation totals in all three regions are discussed. In all regions, recent precipitation in winter (DJF) has increased and summer (JJA) precipitation has decreased. This trend is most evident in Scotland, where the November to April total for 1986 to 1995 is 30 per cent more than corresponding values in 1951 to 1980. For England and Wales the 1995 summer was the driest in the record, with 7 mm less rainfall than in 1976, the previous driest summer. Over Scotland and Ireland the 1995 summer was also dry but not as exceptional.

Mapping regional air temperature fields using satellite-derived surface skin temperatures

Abstract: Screen air temperature is an important climatological variable and accurate mapping of its spatial and temporal distribution is of great interest for various scientific disciplines. The low spatial density of meteorological stations, however, results in relatively large errors during data interpolation and makes it difficult to retrieve the spatial pattern of the temperature field. Errors of the order of 1 to 3 K are mentioned in the literature. The current study investigates the possibilities of mapping and monitoring the spatial distribution of daily maximum air temperatures with the help of time series of NOAA-AVHRR images. The study has been performed for the Mediterranean region of Andalusia in southern Spain. Data analysis included 31 meteorological stations and 148 AVHRR images from the year 1992. Regression analysis between the daily maximum air temperature (Tmax) and the mean surface skin temperature (Ts) retrieved for 11 km2image windows centred over each station, suggests that Tmaxis strongly linked to Tsin the given environment (mean R2=0·823) and that for individual stations Tmaxcan be retrieved from Tswith a mean error of about 2 K. The spatial representativity of the station measurements as well as the influence of altitude and land use on the results are discussed. Finally, the possibilities of retrieving the spatial pattern of Tmaxhave been evaluated through a cross-validation approach. In this analysis Tmaxhas been predicted for each station and for all days of available image data based on a regression model retrieved from all other stations. Again the results indicate that we are able to reproduce the daily distribution of maximum air temperatures with a mean error of the order of 2 to 2·5 K, using satellite-retrieved surface skin temperatures. In addition, the method allows for the detection of stations with a low spatial representativity or a pronounced measurement bias. Future research will aim at the inclusion of further physiographic data, the grouping of stations according to site-specific characteristics and an analysis according to seasons. © 1997 Royal Meteorological Society.

A principal component and long‐term trend analysis of daily precipitation in Switzerland

Abstract: Daily precipitation patterns over Switzerland are investigated by rotated and unrotated principal component analysis for the periods 1901–1990 (with 113 continuously operating rain-gauge sites) and 1961–1990 (with 304 sites). Empirical orthogonal functions are utilized to homogenize the precipitation series and to optimally transform the 113 series of the long-term record into a few variables. Several statistically significant linear trends are detected. This includes, in particular, a wintertime increase in precipitation by up to 30 per cent per 100 years in the western and northern parts of Switzerland. This trend is statistically significant at the 90 per cent level. In most parts of southeastern Switzerland, winter precipitation increased as well but by a smaller rate and at a slightly lower statistical significance level. For the period 1961–1990, precipitation amounts increased in most subregions and all seasons with the exception of summer. A statistical model is formulated in order to isolate the contributions to the observed precipitation trends that are related to changes in either the frequency or the precipitation activity of Alpine weather classes. It is demonstrated that the observed precipitation trends cannot be interpreted based upon changing frequencies of weather classes (the observed changes would in fact lead to drier wintertime conditions), but are rather dominated by the tendency of most rain-producing weather types to produce more rain. Some physical mechanisms that could account for this unexpected result are also discussed. © 1997 by the Royal Meteorological Society. Int. J. Climatol., 17: 1333–1356 (1997) (No. of Figs: 9. No. of Tables: 5. No. of References: 69)

Homogenization of swedish temperature data. part ii: homogenized gridded air temperature compared with a subset of global gridded air temperature since 1861

Abstract: Homogeneity tests of long seasonal temperature series from Sweden, Denmark, Finland, and Norway indicate that homogeneous series are rare and that an abrupt change of the relative mean level is a much more common type of nonhomogeneity than a gradual change. Furthermore, negative shifts were 20% more common than positive shifts. Homogenized temperature anomaly series that were constructed for six 5 degrees latitude x 5 degrees longitude grid boxes indicate that the temporal pattern of temperture changes has been similar in different parts of Sweden since 1861. The annual mean temperature over Sweden was found to have increased by 0.68 degrees C from the period 1861-1890 to 1965-1994. The corresponding changes for the seasons were: +0.18 degrees C (winter), +1.40 (spring), +0.42 (summer) and +0.60 (autumn). A direct comparson shows that non-homogeneities in the temperature series from individual grid boxes in a global data set can be as large as the total changes observed. We estimate that a 95 per cent confidence interval for the error, due to non-homogeneous long station records, in estimates of hemispheric temperature changes over land regions since the period 1861-1890 is +/-0.1 degrees C for the Northern Hemisphere and the globe and +/-0.25 degrees C for the Southern Hemisphere. For a region consisting of about five grid boxes, this error is +/-0.5 degrees C. The large non-homogeneities in individual grid-box series in the global data set is partly a consequence of the fact that homogeneous climate data are not always easily available for the open research community. We urge that efforts are made to improve this situation.

Relief and climate in South Asia: the influence of the western ghats on the current climate pattern of peninsular India

Abstract: The 1500-km-long Western Ghats mountain barrier of peninsular India interacts with the southwest monsoon in a manner which bears heavily on the exceptionally varied climate pattern of the Deccan. Karnataka Province alone concentrates five of the six major climate types of the entire Indian Union. This review explores the interactions between atmospheric structure over South Asia and relief and discusses the efficiency with which the passive margin uplift, second only to the Himalayan barrier, acts as a climatic gatekeeper to the subcontinent. Particular attention is given to rainfall patterns and regimes. These are revealed by a variety of statistical classification and mapping techniques, and the analysis is guided by the steep environmental gradient observed on the immediate backslope of the Ghats, where annual totals can drop from 6000 to 600 mm in ca. 80 km. This is strongly reflected in the landform, soil, vegetation and cropping patterns and raises the question of the relationship between the uplift history of the mountain barrier at geological time-scales, the history of the South Asian monsoon circulation and the stability and diversity of the climatic pattern as seen today. The tightly arranged suite of bioclimatic regions also provides a unique geographical backdrop to the agricultural diversity of South India, rarely found on such a scale in other monsoon contexts of the Tropics. © 1997 The Royal Meteorological Society.

Interannual variability of south‐eastern african summer rainfall. part 1: relationships with air–sea interaction processes

Abstract: This paper investigates the role that air‐sea interaction processes may play in interannual variability of south-eastern African summer rainfall. The principal spatial modes of south-eastern African summer rainfall are first identified using principal component analysis. Four modes are retained. The most important mode of variability is found to represent rainfall variability over most of the domain, particularly in the regions to the south. The influence of ENSO (as measured by the SOI) on summer rainfall is investigated in detail for different SOI leads. The relationship is such that during the summer following the onset of an ENSO event, south-eastern Africa tends to experience dry conditions. Strongest relationships are found with the SOI leading rainfall by about 3 to 6 months. A second index, the Brandon‐Marion Index (BMI) which is indicative of changes in the pressure field over the Indian Ocean correlates with rainfall better than the SOI. Strongest correlations are found when this index leads rainfall by about 1 to 3 months. More importantly, a partial correlation analysis reveals that the BMI influences rainfall independently of ENSO. Both the SOI and the BMI are potential predictors of summer rainfall. An investigation of rainfall associations with global SST anomalies reveals areas in the tropical Indian and Pacific Oceans that are linked with rainfall changes over the subcontinent. The relationship is such that warm anomalies tend to be followed by dry conditions over much of south-eastern Africa. Strongest relationships are found when SSTs lead the rainfall season by about 1 to 3 months. Well-defined atmospheric anomalies are identified during dry south-eastern African summers. These include, amongst others, anomalously warm tropospheric temperatures and marked low-level cyclonic circulation anomalies over the central Indian Ocean, which generate abnormally weak easterly winds along much of the south-eastern coast of Africa. These perturbations to the low-level flow divert moisture from the continent and result in precipitation decreases. An important and related finding is the fact that the SST‐rainfall link over the Indian Ocean remains strong after the ENSO effects have been removed, suggesting that the atmospheric circulation anomalies observed over south-eastern Africa during dry summers, are linked mainly to SST anomalies over the Indian Ocean. This hypothesis will be tested in a companion paper through a series of GCM simulations.

Association of the Indian summer monsoon with the northern hemisphere mid-latitude circulation

Abstract: The association between the mid‐latitude circulation and rainfall over the Indian region on an intraseasonal time‐scale is investigated by considering 11 years (1974‐1984) of Northern Hemisphere 500 hPa geopotential heights and rainfall data for the Indian summer monsoon months June through to September. On the basis of extensive correlation analysis between the geopotential heights and rainfall, it is seen that three regions over the mid‐latitudes, the Manchurian region, the Algerian region and the Caspian sea region show positive correlation with rainfall over India, with higher values north of 20N latitude. Lead and lag correlations between the heights at the locations identified above and rainfall over India reveals that some common element of low‐frequency variability is influencing the mid‐latitude circulation and Indian rainfall. On the interannual scale the connections between the winter‐time low‐frequency patterns (the Pacific/North Atlantic, the West Pacific Oscillation, the North Atlantic Oscillation and the Eurasian) and Indian summer monsoon rainfall (ISMR) are investigated. Only the West Pacific Oscillation pattern shows a significant relationship with the ISMR. Further, the interannual and the decadal variability is examined by using the Northern Hemisphere zonal index data for the period 1900‐1993. Results reveal that the decadal‐scale variability of the ISMR and the circulation features of the Northern Hemisphere are connected. # 1997 by the Royal Meteorological Society. Int. J. Climatol., 17: 1055‐1067 (1997)

Homogenization of Swedish temperature data. Part III: the long temperature records from Uppsala and Stockholm

Abstract: Homogenization of Swedish temperature data Part III : The long temperature records from Uppsala and Stockholm

Non‐stationarity in daily precipitation series: implications for gcm down‐scaling using atmospheric circulation indices

Abstract: The paper addresses three issues relating to the design and calibration of precipitation models for down-scaling General Circulation Model (GCM) output. First, the significance of non-stationary daily circulation-pattern–precipitation relationships will be described for two UK sites, each with records beginning in 1881. Second, the cause(s) of the observed non-stationarity will be investigated using lengthy air flow, synoptic, temperature, and standardized teleconnection indices. Third, the prospects for incorporating these factors within down-scaling methods will be discussed with reference to several alternative techniques. Finally, a multivariate approach incorporating teleconnection and airflow indices within precipitation models is advocated. © 1997 by the Royal Meteorological Society. Int. J. Climatol. 17, 439–454 (1997).

Simulated changes in extreme rainfall over southern africa

Abstract: A general circulation model simulation is used to investigate possible changes in rainfall over southern Africa resulting from a doubling of atmospheric carbon dioxide. Simulated increases in rainfall intensity are found to be a spatially coherent and an apparently less regionally dependent signal of climatic change than changes in annual means or number of rain-days. Accordingly, increases in both the frequency and intensity of extreme daily rainfall events are simulated throughout most of the subcontinent. Simulated increases in the intensity of the lowest frequency floods are shown to be particularly severe, suggesting that greenhouse-relate d climatic change may be most detectable through an increase in extreme flood events rather than changes in long-term means. Similar results are evident when changes in the frequency and intensity of prolonged rainfall events, measured over a period of five consecutive days, are analysed. All results are qualitatively similar to those for the Australian region, except that the model's sensitivity to sharp changes in topography over southern Africa is highlighted. © 1997 by the Royal Meteorological Society.

Multiscale detection of abrupt climate changes: application to River Nile flood levels

Abstract: The historical flood-level time series of the River Nile (AD 622–1470) is chosen to identify abrupt climate changes by applying global and local analysis techniques: the Mann–Kendall test and a non-hierarchical cluster analysis method to improve the Mann–Kendall test; a multiscale moving t-test with correction to the degree of freedom and an antisymmetric wavelet transform. The global estimates show three distinct epochs, AD 622–1078, 1079–1325 and 1326–1470, coinciding with larger scale climate changes: a relatively cool age, the Little Climatic Optimum of the Middle Ages, and an interim period before the Little Ice Age. The local estimates reveal the following results. The reference time of abrupt changes can be clearly identified, the associated time-scale coincides with the persistent anomaly period, and the maximum absolute t-value is statistically significant. There are about eight almost synchronous abrupt changes in the minimum and maximum River Nile flood levels, many of them are associated with 35–45 year persistence time-scales. An association of these short time-scales with those of interdecadal variability reported for the mid- and high-latitude sea-surface temperature of the North Atlantic is suggested, although information on phase coherence is not available. ©1997 by the Royal Meteorological Society. Int. J Climatol., 17: 1301–1315 (1997) (No. of Figures: 9 No. of Tables: 1 No. of References: 40)

Comparison of circulation classification schemes for predicting temperature and precipitation in the netherlands

Abstract: Three existing classification schemes of daily circulation patterns are considered: (i) the subjective Grosswetterlagen; (ii) an objective scheme, the Jenkinson classification, which produces weather types similar to the subjective Lamb classification for the British Isles; and (iii) the objective P27 classification scheme developed at the Royal Netherlands Meteorological Institute. The comparison between these schemes is based on the mean-squared-error skill score (percentage of explained variance) and the correlation coefficient between the observed and predicted values. The relationship between these performance measures is examined. For the Grosswetterlagen and the P27 classification the class averages of the daily temperatures at De Bilt explain about 40 per cent of the variance of the daily temperatures at that location in the period 1949–1993. The corresponding figure for the Jenkinson classification is 29 per cent. The Grosswetterlagen perform less well for the daily temperatures prior to 1949, when upper-air data were not available to define the daily weather types. For daily precipitation characteristics the three classification schemes perform almost equally well. The class averages of the daily amounts explain 13 (summer season) to 20 per cent (winter season) of the variance of the local precipitation at De Bilt and about 25 per cent of the variance of the area-average precipitation over The Netherlands. A somewhat higher skill is achieved for precipitation occurrence. There is some evidence of systematic changes in the relationships between rainfall characteristics and weather types. The class averages of the daily values have been used to estimate monthly values. Especially for the precipitation characteristics the estimates of the monthly values have higher skill than those of the daily values. Nearly 60 per cent of the variance of the monthly area-average rainfall over The Netherlands is explained. Almost the same skill is achieved with a regression on the monthly number of cyclonic and anticyclonic days or a regression on the monthly mean vorticity index in the Jenkinson classification. © 1997 by the Royal Meteorological Society. Int. J. Climatol., 17: 875–889 (1997) (No. of Figures: 2. No. of Tables: 8. No. of References: 22.)

Prediction of Droughts in North‐East Brazil: Role of ENSO and Use of Periodicities

Abstract: The relationship between El Nino and droughts in north-east Brazil is poor. From 46 El Nino events (strong and moderate) during 1849-1992, only 21 (ca. 46 per cent) were associated with droughts at Fortaleza, Ceara (north-east Brazil) and only 26 (ca. 57 per cent) with negative deviations (of any size). Thus, forecasts of droughts based on the appearance of El Nino alone would be wrong half the time. Instead, predictions based on significant periodicities (ca. 13 and ca. 26 years) give reasonably good results. Data up to 1978 could predict the drought that occurred during 1979-1983, although the severe drought of 1983 was attributed to a strong El Nino. Mild droughts during 1991-1993 were also foreseen. In future, severe droughts are predicted for 2000-2010 in north-east Brazil. However, the prediction skill is not very good and predictions should be checked by analysis of as much updated data as possible. Predictions beyond about 5-10 years are not reliable. The significant periodicities in North-east Brazil rainfall series might be related to periodicities in Atlantic parameters (sea-surface temperatures, etc.) that are well correlated with north-east Brazil rainfall.

Some potential forcing mechanisms of the year‐to‐year variability of the tropical convection and its intraseasonal (25–70‐day) variability

Abstract: Three potential forcing mechanisms for the convective signal of the Madden–Julian Oscillation (MJO) are investigated.Outgoing longwave radiation (OLR) is used to indicate convection, and data from 1974 to 1994 are subjected to a 25–70-daybandpass filter to represent the MJO. The focus of the paper is on interannual variability and the three mechanisms diagnosedare: tropical sea-surface temperature (SST); the phase of El Nin˜o–Southern Oscillation (ENSO); and atmospheric precipitablewater (W). The respective roles of these potential forcings are also considered for the year-to-year variability of the seasonalanomalies of mean convection.Observed seasonal anomalies of both mean convection and MJO activity over the equatorial Pacific from 160  E to thecoast of South America are highly correlated to collocated anomalies of SST and the phase of ENSO. During El Nin˜o wintersthe eastward-propagating convective MJO events penetrate farther than normal into the central Pacific before curvingpoleward into the Southern Hemisphere. Over the maritime continent, moist convection is significantly suppressed during ElNin˜o. A corresponding signal in the MJO activity is confined to the Philippine Sea and North Australia during boreal winter,and to the Bay of Bengal in autumn. In contrast, there is no uniform MJO response over the eastern Indian Ocean when thelatest five ENSO warm events (1976–1977, 1982–1983, 1986–1987, 1991–1992 and 1993–1994) are considered.On a seasonal time-scale, W over the Indian and Pacific Oceans is closely related to convection anomalies, as well as tothe variance of the high-frequency convective perturbations. In contrast, convective activity of the MJO shows no relation tovariations in the amount of W. It is concluded that the interannual variability of the convective MJO signal over the IndianOcean and western Pacific Warm Pool is largely determined by the frequency and duration of ‘weak-signal’ episodes ratherthan by modulations of its amplitude. # 1997 Royal Meteorological Society. Int. J. Climatol., Vol. 17, 1513–1534 (1997)(No. of Figures: 12. No. of Tables: 2. No. of References: 66)

North Atlantic and European seasonal predictability using an ensemble of multidecadal atmospheric GCM simulations

Abstract: An initial study of seasonal predictability for the North Atlantic and European regions is presented, using an ensemble of six integrations of the Hadley Centre atmospheric climate model (HADAM1) for the period 1949‐1993. The model was forced by the Hadley Centre’s observed global sea-ice and sea-surface temperature data set (GISST). The model reproduces with considerable skill the patterns of many of the main modes of mean sea-level pressure (MSLP) variance over this region, including the North Atlantic Oscillation (NAO). For MSLP, the most predictable modelled seasons over the North Atlantic‐European sector as a whole are spring and winter (although random variability is still substantial), and for the NAO specifically, the model shows significant skill from winter to spring. At this time of year the observed NAO is related to SST patterns mainly in the North Atlantic, and in years of extreme Atlantic SST anomalies, reasonably skilful statistical simulations of the sign of the NAO anomaly can be made. For the UK, the highest simulation skill is for temperature, which we hypothesize is related directly to local SST anomalies. The effects of ENSO have also been investigated. Over the North Atlantic, the model produces a strong 500 hPa height

Tropical–extratropical cloudbands and Australian rainfall: I. climatology

Abstract: This paper examines the contribution of tropical–extratropical cloudbands, and of interactions between tropical cloud and mid-latitude systems, to cool season (April–October) rainfall in agriculturally marginal areas of Australia. A following paper describes inter- and intra-annual variability of these features. A classification scheme for these tropical influences based on GMS satellite imagery is described, and used to compile a 15-year archive of events. It is shown that cloudbands extending from the tropical oceans bordering Australia (‘Oceanic’ Cloudbands) are most frequent and influential between April and July, but decrease sharply after August, at which time bands originating over the continental interior (‘Continental’ Cloudbands) increase. The contribution of these systems to rainfall at stations representing agriculturally marginal areas is assessed. Oceanic Cloudbands originating west of 120°E contribute 70–90 per cent of cool-season rain in north-western Australia, with the contribution decreasing to the south and east. North-eastern Australia receives a significant portion of its rain from Cloudbands originating east of 120°E. Tropical–mid-latitude interactions are more important over eastern than western Australia, and produce some 30–40 per cent of rain over much of inland eastern Australia. The overall tropical influence (Cloudbands plus interactions) on rainfall is least in South Australia and western Victoria, but still amounts to some 35–40 per cent of cool-season rain in those areas. The proportion of events producing significant rainfall (>10 mm) is also examined: almost two-thirds of the Oceanic Cloudbands to affect western Australia produce significant rain, and about half of those affecting eastern Australia. © 1997 by the Royal Meteorological Society. Int. J. Climatol., 17: 807–829 (1997) (No. of Figures: 10. No. of Tables: 4. No. of References: 33.)

Surface Pressure Patterns During The Year Over Southern South America

Abstract: Principal component analysis was carried out on daily sea-level pressure maps for the period 1972–1983 in southern South America to determine the main synoptic types. Each month in the whole period was studied in order to analyse the variability through the annual cycle. The results show that the first six principal components (PC) explain more than 90 per cent of the total variance and they give approximately the same patterns for all the months. For each pattern the most important difference observed in the features through the annual cycle was the shift towards high latitudes in the synoptic systems during the summer months. It means that there are no remarkable seasonal contrasts apart from the latitudinal position and differences in the variances accounted for by the patterns. Half or more than half of the whole variance was explained by the first PC, the spatial pattern of which is quite similar to the feature of the monthly mean pressure field. This synoptic situation represents the major contribution to the zonal flow for surface circulation at mid-latitudes of the area studied. The remaining PC patterns have been recognized as features corresponding to the most frequent baroclinic perturbations to the basic flow. Furthermore, these synoptic situations account for the principal meridional flow component. In brief, the present results are found to be in general agreement with previous knowledge of southern South America. © 1997 by the Royal Meteorological Society.

Downscaling large-scale circulation to local winter rainfall in north-eastern Mexico

Abstract: The large-scale atmospheric controls on winter rainfall variability for north‒eastern Mexico and south‒eastern Texas are diagnosed based on 8 years (1985–1993) of daily data from the Goddard Space Flight Center four‒dimensional data assimilation scheme. Downscaling techniques based on artificial neural nets are used to derive transfer functions from the large‒scale circulation to local precipitation. Daily rainfall amounts are predicted from synoptic sea‒level pressure, 500 hPa heights, and the 1000–500 hPa thickness, and summed over the cool season (NDJF). The monthly rainfall totals are also predicted independently from other indices of the large‒scale circulation, such as 1000–500 hPa thickness, the Pacific/North American (PNA) pattern, and a standardized Southern Oscillation Index (SOI) derived from the Tahiti minus Darwin sea‒level pressure difference. Time‒lagged component scores from a rotated principal component analysis of sea‒level pressure, 500 hPa heights, and 1000–500 hPa thickness serve as input to a neural net that produces time‒series of daily rainfall amounts for 20 grid‒points in the study area. The correlation between the observed and predicted rainfall is greater than 0·7 in the coastal plains of the Gulf of Mexico and less than 0·7 over the Sierra Madre and the Gulf, suggesting an increase in the importance of local rainfall processes in the last two regions. The analysis shows a systematic relationship between the performance of the net and physiography, which is confirmed by the consistency of the patterns of spatial correlation, mean absolute error, and root‒mean‒square error at different time‒scales. The net captures the phase and amplitude of most of the rainfall events, reflecting the influence of the large‒scale circulation. However, interannual fluctuations in rainfall associated with persistent El Nino conditions are partly bypassed by the net, suggesting that more information is needed to predict extreme events. The PNA pattern does not seem to be associated with local rainfall in north‒eastern Mexico and south‒eastern Texas during the period analysed. © 1997 The Royal Meteorological Society.

Persistence Characteristics of Australian Rainfall Anomalies

Abstract: Using 79 years (1913‐1991) of Australian monthly precipitation data we examined the nature of the persistence of rainfall anomalies. Analyses were performed for four climate regions covering the country, as well as for the entire Australian continent. We show that rainfall over these regions has high temporal variability and that annual rainfall amounts over all five sectors vary in phase and are, with the exception of the north-west region, significantly correlated with the Southern Oscillation Index (SOI). These relationships were particularly strong during the spring season. It is demonstrated that Australian rainfall exhibits statistically significant persistence on monthly, seasonal, and (to a limited extent) annual time-scales, up to lags of 3 months and one season and 1 year. The persistence showed strong seasonal dependence, with each of the five regions showing memory out to 4 or 5 months from winter and spring. Many aspects of climate in the Australasian region are known to have undergone considerable changes about 1950. We show this to be true for persistence also; its characteristics identified for the entire record were present during the 1951‐1980 period, but virtually disappeared in the previous 30-year period. Much of the seasonal distribution of rainfall persistence on monthly time-scales, particularly in the east, is due to the influence of the SOI. However, most of the persistence identified in winter and spring in the north-west is independent of the ENSO phenomenon. Rainfall anomalies following extreme ‘dry’ and ‘wet’ months, seasons and years (lowest and highest two deciles) persisted more than would be expected by chance. For monthly extreme events this was more marked in the winter semester for the ‘wet events’, except in the south-east region. In general, less persistence was found for the extreme seasons. Although the persistence of dry years was less than would have been expected by chance, the wet years appear to display persistence. 1997 by the Royal Meteorological Society. Int. J. Climatol. 17: 597‐613, 1997.

‘Global and terrestrial precipitation: a comparative assessment of existing climatologies’: a reply

Abstract: Over the past 20 years, several global and terrestrial precipitation climatologies have been developed. These climatologies have been used extensively, for example, to evaluate general circulation model simulations of the present-day climate, as input fields in global hydrological studies, and to validate satellite precipitation algorithms. If these climatologies differ significantly, the results of these applications may be affected adversely. In this paper, three global precipitation climatologies—developed by Schutz and Gates, Jaeger, and Legates and Willmott—are examined comparatively. Additionally, three terrestrial precipitation archives—compiled by Eischeid et al., Leemans and Cramer, and Hulme—are compared with the Jaeger and the Legates and Willmott climatologies to assess precipitation over land areas and the importance of the time period of record on long-term averages. Results indicate that seasonal precipitation estimates from the Schutz and Gates climatology are considerably smaller than either of the other two climatologies and its annual average estimate is much smaller than most estimates of global precipitation made since 1960. Zonally, however, the smaller estimates are predominantly located in lower latitudes although the general spatial pattern is similar to that of the other two global climatologies. Spatial gradients and areas of precipitation maxima are much smaller in the Schutz and Gates climatology, which accounts for its lower estimates. Qualitatively, the Jaeger climatology agrees well with that of Legates and Willmott although Legates and Willmott's estimates are much greater along the Inter Tropical Convergence Zone. Oceanic estimates vary considerably which underscores the uncertainties in estimating oceanic precipitation. Over land, the Legates and Willmott climatology compares favourably with the observation-based estimates from the terrestrial archives, leading to the conclusion that observation-based climatologies are consistent regardless of the time period of record and are preferable to climatologies compiled subjectively.

Predictability of Zimbabwe summer rainfall

Abstract: Predictors of Zimbabwe summer rainfall are investigated with a view to improved long‒range forecasts. Teleconnectivity is assessed in respect of sea‒surface temperatures, the Southern Oscillation index, the Quasi‒biennial Oscillation (QBO), outgoing longwave radiation (OLR) and wind. Spectral analyses of historical rainfall gives an indication of cycles in the range 2·3, 18 and 3·8 years, possibly associated with the QBO, the luni‒solar tide and the El Nĩo–Southern Oscillation (ENSO), respectively. Pair‒wise correlations are found between Zimbabwe summer rainfall and SST in the central Indian Ocean (r<−0·5) in austral spring. Below normal OLR values in September over southern Africa corresponds with good rains in the following summer. Rainfall–upper‒wind correlations are optimum (r<−0·7) over the equatorial Atlantic in spring. Comparatively weak correlation with the QBO may also reflect biennial adjustment of monsoon and global ENSO teleconnections. Additional predictor variables are utilized and multivariate models are formulated for early and late summer rainfall and maize yield in Zimbabwe. The models use three to five predictors, are trained over a 22‒year period and perform well in jack‒knife skill tests. Summer rainfall forecasts with one season lead times are viable and could ameliorate hardship caused by drought. ©1997 Royal Meteorological Society. Int.J.Climatol., Vol.17, 1421-1432 (No. of Figures: 5 No. of Tables: 0 No. of References: 59)

Glacier balance fluctuations and atmospheric circulation patterns over the Southern Alps, New Zealand

Abstract: The links between climate and glacier change are investigated by examining the relationships between atmospheric circulation and glacier balance changes over a 17 year period. Altitudes of the end-of-summer snowlines made on some 48 index glaciers provide surrogates for mass balance, and mean atmospheric pressure maps were computer-generated for the south-west Pacific from long-period meteorological station data. Pressure data are separated into accumulation and ablation seasons. The results show that atmospheric circulation patterns exert a strong control on glacier mass balance, with positive balances associated with anomalous south to south-west flow and negative balances associated with enhanced north to north-east flow.

The association between the BWA index and winter surface temperature variability over eastern Canada and west Greenland

Abstract: Since about 1970, winter surface temperature data from stations on coastal eastern Canada and western Greenland have shown detectable decadal cooling. In this study, we attempt to understand some aspect of this surface cooling trend by relating it to the variability of the Canadian Polar Trough (CPT). In order to facilitate the relationship, we introduce a new 50 kPa index called the Baffin Island–West Atlantic (BWA) index which, although reflecting the variability of the western structure of the North Atlantic Oscillation (NAO), is found to explain temperature variability better in north-eastern North America than the structure characterized by the NAO index. The decadal variability in the winter surface temperature is found to be associated with the BWA index at a statistically significant correlation of 0·85. Two distinctive winter climate regimes are found to exist in the climate record from 1947 to 1995, one before and one after about 1970. Although the magnitude of the variance does not change significantly from one regime to the next, the two regimes are characterized by statistically significantly different means and by two distinct spectral signatures. Variability before 1970 is dominated by interannual fluctuations, whereas afterwards much of the contribution to the variability comes from interdecadal fluctuations. Subtraction of the 1947–1969 winter 50 kPa mean height field from the 1970–1995 mean field shows that the change in the height field over the Northern Hemisphere is reflected in the enhancement of the negative phase of the NAO mode (which corresponds to a strong jet stream over the western Atlantic and a strong Icelandic low) and of the positive phase of the Pacific/North America (PNA) mode. © 1997 The Royal Meteorological Society.