The south-west monsoon current, which brings in most of the annual rainfall to Kerala, gets a forced ascent at the Ghats and the windward slopes experience very heavy rainfall. However, rainfall is not uniformly distributed on the windward slopes and there are pockets of very heavy rainfall and relatively less rainfall. The coastal region of the state also receives abundant rainfall. The results suggest that Kerala can be divided into three unique rainfall regions, each region having a similar covariance structure of annual rainfall. Stations north of 10.N (north Kerala) fall into one group and they receive more rainfall than stations south of 10.N (south Kerala). Group I stations (table 2) receive more than 65% of the annual rainfall during the south-west monsoon period, whereas stations falling in Group II (table 3) receive 25-30% of annual rainfall during the pre-monsoon and the north-east monsoon periods. Group III stations (table 4) with IIIrd factor as the primary loading, i.e., IIIrd factor greater than Ist and IInd factor. The meteorology of Kerala is profoundly influenced by its orographical features, however it is difficult to make out a direct relationship between elevation and rainfall. Local features of the state as reflected in the rainfall distribution are also clearly brought out by the study. The classification of Kerala according to the important factors is given in figure 2.
There are 28 rainfall stations in Group I, which can be again classified into 2 according to the contribution of south-west monsoon rainfall (see regions Ia and Ib in figure 2): Stations receiving more than 75% of the annual rainfall during the south-west monsoon, and stations receiving 65-75% of the annual rainfall during the south-west monsoon season. All these stations are situated north of 10.N (north Kerala) and receive more rainfall than stations south of 10.N (south Kerala). This division into north and south Kerala arises from the fact that monsoon rainfall progresses from south to north along the west coast of India and there are differences in the rainfall characteristics between the southern and northern parts of the state. An examination of the monthly rainfall pattern shows that rainfall is highest in June south of 10.N while north Kerala experiences the highest rainfall in July. With each surge in the monsoon current,rainfall is more in north Kerala than in the south and the frequency of surges is higher in July than in June. The active spells of monsoon are also more in July and August. Further, under weak monsoon conditions there is divergence over south Kerala causing little or no rain, while there is convergence resulting in rain over north Kerala on many occasions.
South Kerala stations have the second factor as the primary factor and there are 26 stations in Group II. All these stations receive 25-30% of the annual rainfall during the pre-monsoon and north east monsoon season. (i) Precipitation during pre-monsoon is mainly from thundershowers and there is an increased thunderstorm activity in the southern tip of Kerala state from March onwards increasing progressively with the advance of the season. May makes the maximum contribution to the rainfall of the pre-monsoon period (March - May). The south-west monsoon advances over south Kerala between 11th and 31st May in nearly 50% of the years, and this is a contributory factor for the pronounced rainfall maximum over south Kerala during this season.
The meteorology of Kerala is profoundly influenced by its orography. This is reflected as the third factor in the factor analysis. Of the 77 stations, 23 stations have this factor as the primary loading. Even though these stations have the third factor as their primary loading, all these stations do not have the same rainfall pattern and can be classified into three groups. (i) There are 8 stations that receive very heavy annual rainfall (> than 336 cm) and are located on the windward slope of the Ghats. On the windward side of mountains, moist air is forced up the slope, where it cools and condenses, leading to precipitation. These peaks also provide convection points of instability. This instability will be triggered when the orographic lifting is strong enough to force air parcels to ascend to their level of free convection. Thus, the heavy rainfall over the windward slopes depends not only on the elevation of the station but also on the wind velocity perpendicular to the mountain range and on the moist static stability.
The complex relationship between topography and precipitation in mountainous regions is evident in the pattern of rainfall distribution. A frequency table showing the rainfall and elevation is illustrated in table 5. Stations are classified according to their elevation into 5 groups. The normal rainfall of Kerala for the period of study is 280 cm and stations having rainfall greater than 20% of the state normal (i.e., greater than 336 cm) are classified as excess rainfall stations and those having rainfall less than 20% of the state normal (i.e., less than 224 cm) are grouped as scanty rainfall stations. It can be seen from table 5 that while there are 4 excess rainfall stations with elevation between 500m and 1000m 3 stations with the same altitude experiences only scanty rainfall. Figure 3 illustrates the rainfall versus the elevation of each station. It is evident from the figure that it is difficult to make out a direct relationship between elevation and rainfall. The spatial variability of mean annual precipitation depends upon the topographic factors like exposure of the station to the prevailing wind, elevation, orientation, and slope of the mountain.
The mean rainfall of the recent past three decades was well below the centennial average rainfall, clearly indicating decreased monsoon activity. This could be related to increase in tropical deforestation and land use change occurring in the areas surrounding this semi-arid ecosystem, like the Montane cloud forest high waives (Mehamalai) and the cardamom hills during the report period. No alternating pattern of increasing and decreasing rainfall in the decades is seen and no epochal trends in ISM as observed for the rest of the country. The 20, 30, 40 and 50 year running means showed declining trends (2.7 to -3.4 mm per year, with R2 values ranging from 0.42 to 0.85, significant at 95%). This means the future rainfall in the rain shadow slopes of the cardamom hills is likely to decrease. Though historical records say that the eastern slopes have enjoyed good rainfall, in the last century the slopes had lost their thick evergreen forest (90% of primary forest). Whether this has triggered or will trigger changes in rainfall pattern, questionable and sensitivity analysis performed elsewhere showed some teleconnected effects of deforestation in the precipitation change of south East Asia. El Nino and La Nina events have profound influence on the rainfall of this region. The most intense El Nino year in the last century was 1997, when the rainfall was well above the average. Nearly 58% of the El Nino years had below average rainfall. On the other hand 42% of the El Nino years accounted above normal rainfall. Amazingly the lowest rainfall of the last century was in 1983(328.5mm), which happened to be a typical La Nina year, and the effect was spectacular. 70% of the 17 La Nina years brought about below average rainfall and only 30% of the La Nina years experienced more than average rainfall. This shows a changing relationship between semi-arid rainfall climatology and ENSO in this region. Up to many decades ago, the entire area was under cardamom cultivation. For the past 8-10 years, the actual area under cardamom cultivation has shrunk to about 40,000 hectares. Relatively drought resistant crops like coffee and silk cotton are cultivated in the remaining areas. This is indicate the insufficiency of rainfall on the eastern and northern borders of CHR, and the severity of aridity is advancing towards up in the cardamom hills' reserves into the state of Kerala.
There was a strong association of El-Nino phenomenon with below average rainfall (67% of the El-Nino years), indicating a good ENSO relationship. In the case of La-Nina years, fifty per cent of the years enjoyed excess rainfall and rest of the years were low rainfall years. However, the impact of El-Nino phenomenon was not severe during the first pentad of this century in this mountain range. This shows the relationship seems to have weakened in the recent past. Considering these, the scientists concerned should take precautions in forecasting local climate. The coefficient variation of rainfall (CV) for the entire series ranged from 9 to 27%. The decadal averages for the first three decades (1896-1925) and the last three decades (1976-2005) studied, reported values less than the annual mean (5000 mm) for the whole period (1896-2005). Only during the middle period (1926-1975), the decadal average rainfall did exceed the series annual mean. Interestingly the strongest El-Nino years (1997-1998) had received (5120 mm) fairly well above the annual mean. Since 1896, of the total El Nino years, only ten years each in El-Nino and La-Nina years had envisaged less than average. Otherwise all other El-Nino and La-Nina years found to be more than the annual average (5000 mm). These results are found to be for and contradictory to the results reported for many regions of varied environment in India Therefore, extreme rainfall anomaly can occur even in the absence of strong ENSO events, thus uncertainty in ENSO effects is quite large from one environment to another even that are close to each other. The production trend of major spices and plantation crops shows a negative trend for the first one and half a pentads (2000-2007) of this century. Cardamom reported a maximum percentage change of -12.5.1% followed by black pepper (-10.0%) in CHR. In particular, tea production in the cardamom hills has gone down significantly reporting a percentage change of -20% during the recent past seven years. Therefore, this unhealthy trend of production of specific crops is certainly due to the change in the climatic condition which affects both quantity and quality of crops.
