Trends in Rainfall Patterns over North-East India during

Transcription

1 INTERNATIONAL JOURNAL OF EARTH AND ATMOSPHERIC SCIENCE Journal homepage: Trends in Rainfall Patterns over North-East India during Sunit Das a, *C. S. Tomar b, D. Saha b, S. O. Shaw a and C. Singh c ORIGINAL ARTICLE a Regional Meteorological Centre Guwahati, b Meteorological Centre, Agartala, c India Meteorological Department, New Delhi, India. *Corresponding Author: C.S. Tomar cstomar2002@gmail.com Submitted: 22/04/2015 Revised: 22/06/2015 Accepted: 23/06/2015 Abstract This study aims to determine trend pertaining to rainfall at ten selected stations of North-East India. Twenty-seven precipitation indices for each station are tested for trends using 50 years data. For this purpose, the non-parametric Mann-Kendall test at 95% confidence level and the change per unit time is estimated by applying Sen's estimator of slope. Most of the significant increasing trends are found in seasonal and monthly rainfall. The monthly rainfall of June at Imphal has a significant decreasing trend and is decreasing at the rate of 2.19 mm/yr. The frequencies of rainfall receiving 4-6 cm per day at Agartala and Guwahati during monsoon season show decreasing trends. Similar result is also observed in the annual frequency of 4-6 cm rainfall per day at Agartala. There is no significant trend in the frequency of wet spells at any of the stations. Key words: Precipitation indices, Mann-Kendall test, Sen s estimator, Trend analysis, Wet spells. 1. Introduction The rainfall received in an area is one of the determining factors for the socio-economic activities including agriculture, forestry and bio-diversity, water resources management, industry and tourism of the region. The Intergovernmental Panel on Climate Change report (IPCC, 2007) suggests that climate change may lead to changes in runoff and water availability due to alterations in rainfall pattern. The changes in rainfall pattern may cause heavy floods in some areas while other areas may experience frequent droughts (IPCC, 2007). Due to the possible effects of climate change on rainfall pattern, analysis of rainfall characteristics and its long term variability has got special attention worldwide in recent years. Trend analysis of rainfall is the primary tool to understand its temporal variations. However, trend analysis significantly depends on the length of data (Burn and Elnur, 2002; Kahya and Kalayci, 2004; Jain and Kumar, 2012). During monsoon season (Jun-Sept), North-East Regions (NER) of India, receives about 66% of its annual total rainfall (Srinivasan et al., 1972) and remaining contribution comes mainly due to thunderstorm activities during the pre-monsoon months and mid-latitude westerly systems during the winter months (Attri and Tyagi, 2010). According to the Indian Network for Climate Change Assessment (INCCA) Report 2010, the mean annual rainfall for NER is projected to vary from a minimum of 940±149 mm to a maximum of 1330±174.5 mm in the 2030s with respect to 1970s. The increase with respect to 1970 is by 0.3 3%. The report also reflects that in NER, a substantial decrease in rainfall in January and February in 2030 with respect to 1970 with no additional rain projected to be available during the period March to May and October to December. However, the monsoon rainfall during June, July, and August is likely to increase by 5 mm in 2030s with reference to 1970s, a negligible rise indeed (INCCA, 2010). There are several studies in India on the rainfall variability and long term trends (Parthasarathy and Dhar, 1975; Mooley and Parthasarathy, 1984; Sarkar and Thapliyal, 1988; Soman et al., 1988; Thapliyal and Kulshresthra, 1991; Guhathakurta and Rajeevan, 2008; Krishnakumar et al., 2009; Kumar et al., 2010; Bhatla and Tripathi, 2014). Most of these studies investigated the trends in annual and seasonal rainfall series on the country scale or in regional scales. Studies of Mooley and Parthasarathy (1984), Sarkar and Thapliyal (1988), and Thapliyal and Kulshresthra (1991) have concluded that there is no significant trend in average annual rainfall of the country. Kumar et al. (2010) have reported no significant trend for annual, seasonal and monthly rainfall over India. Similarly, there are studies those focused mainly on the trends in intensity of daily rainfall. For example, Rakhecha and Soman (1994),

2 Sen Roy and Balling (2004), Joshi and Rajeevan (2006), Goswami et al. (2006) and Guhathakurta et al. (2010) have studied the trend in extreme rainfall over India. On the other hand, there are only a few studies (Deka et al., 2013; Jain et al., 2013; Das et al., 2011; Das and Goswami, 2003) available on the rainfall variability and trends over northeast India. They have concluded that there is no significant trend in annual rainfall for northeast region whereas Rupa Kumar et al. (1992) have reported a decreasing trend in monsoon rainfall over this region. Guhathakurta and Rajeevan (2006) have studied the trend in rainfall series in meteorological sub-divisional scale and observed a decrease in winter rainfall in Nagalnd-Manipur- Mizoram-Tripura region and a significant increase in annual rainfall in Assam-Meghalaya subdivision. In NER, agriculture practices are primarily dependent on rainfall. Therefore any change in rainfall pattern will potentially impact the agricultural activities of the region. A study by Ravindranath et al. (2011) indicates that the most parts of the region are subjected to climate induced vulnerability currently and in the near future. In view of above facts, it is of special interest to know the present trends in rainfall amount and its intensity and the wet spellsover NER in monthly, seasonal and annual timescales. Most of the studies on rainfall trend and variability over NER are primarily confined to seasonal and annual scale varied data length. However, there is no study available in the region on trends of wet spells and rainfall intensities of different categories. We, therefore, aim to review the recent trends in rainfall amount using a 50 years long data set on monthly, seasonal and annual scale over 10 selected stations of NER. The other objective is to investigate trends in frequency of rainfall intensity of 2.5 mm/day, 1-3 cm/day, 4-6 cm/day, 7cm/day and frequency of wet spells of length for 2 days, 3 days and 4 days respectively during monsoon season and also study the same in annual scale at each station. 2. Rainfall Characteristics of the Study Area The latitudinal and longitudinal boundary of the study area is roughly from 22 o N to 30 o N and 90 o E to 97 o E. Within this boundary, the entire Bangladesh and a part of Myanmar also lies towards south and south east of the region respectively (Fig 1). The region is marked by eastern Himalayas in the north in Arunachal Pradesh, Garo-Khasi-Jaintia hills extending from west to east in the centre and Naga-Patkai hill range from south to north over the eastern parts of the region. The average height of the Garo, Khasi and Jaintia hills is around 1 to 1.5 km and Naga-Patkai hills have an average height of 3 kms whereas the height of the mountains in Arunachal Pradesh ranges between 3 and 5 km. Another important aspect of the physiography is the Brahmaputra valley extending from east Arunachal Pradesh in the east to west Assam in the west. Around 68% of the study area is mountainous. Although distinguished in the altitude differences, both the mountain and plain areas are closely inter-related in terms of precipitation regime. Based on homogeneity of rainfall, NER is divided into three Meteorological sub-divisions as Arunachal Pradesh, Assam and Meghalaya region and Nagaland-Manipur-Mizoram- Tripura (NMMT) by India Meteorological Department (IMD). The four seasons considered here are as per IMD's categorization i.e. winter (January-February), premonsoon (March-April-May), monsoon (June-July- August-September) and post-monsoon (October- November-December). The seasonal and annual rainfall over the different meteorological sub-divisions of this region is furnished in Table 1. The Table 1 is based on the available data on IMD's website ( In winter season, Arunachal Pradesh receives maximum rainfall (148.1mm) whereas NMMT receives the least amount of rainfall (44.0 mm). This may be attributed to the fact that the precipitation during this season mainly occurs while western disturbances pass over the northern parts of NER. The rainfall during premonsoon season is essentially due to thunderstorms and the seasonal rainfall decreases gradually from Arunachal Pradesh to NNMT. This variation of precipitation is well supported by the fact that thunderstorm activities are less over NMMT as compared to Arunachal Pradesh and Assam and Meghalaya region. During monsoon season, Assam and Meghalaya region receives the highest rainfall of mm whereas Arunachal Pradesh and NMMT receive mm and mm respectively. For occurrence of rainfall over NER during this season, trough of low, axis of monsoon trough passing through NER are the predominant contributors. The synoptic systems like low pressure or depression formed over Bay of Bengal during this season generally move in north-westerly direction and seldom affect NER except areas proximity of these systems. The above facts justify well the spatial distribution of rainfall over the region during the monsoon season. During post-monsoon period, Arunachal Pradesh receives the highest rainfall (267.2 mm) and NMMT records more rainfall than Assam and Meghalaya region. The spatial distribution of rainfall during the season is supported by the facts that the - 38

3 Fig 1: Study area and locations of the rain gauge stations on topographic map. rainfall activities over the region during this season are due to retreat of monsoon and passage of western disturbances over the northern latitudes of the region, sometimes due to the recurvature and north-eastwards movement of cyclonic storms formed over Bay of Bengal as well. For this study, one station from Arunachal Pradesh, six stations from Assam and Meghalaya region and three stations from NMMT are considered. The geographical details of the stations and the basic statistics i.e. mean (M), Standard deviation (SD), median (Me), co-efficient of kurtosis (Ck) and coefficient of skewness (Ck) of monsoon and annual rainfall at each station is presented in Table 2a & 2b. The skewness and kurtosis were computed to test whether the annual (monsoon) rainfall data follow a normal distribution. Skewness is a measure of symmetry or, more precisely, the lack of symmetry. The data set is said to be symmetric if it looks the same to the left and right from the center point. The skewness for a normal distribution is zero, and any symmetric data should have skewness near zero. Negative values for the skewness indicate that data are skewed to the left and positive values for the skewness indicate that data are skewed to the right. Kurtosis is a measure of data peakedness or flatness relative to a normal distribution. That is, data sets with a high kurtosis tend to have a distinct peak near the mean, decline rather rapidly, and have heavy tails. Data sets with low kurtosis tend to have a flat top near the mean rather than a sharp peak. The standard normal distribution has a kurtosis of zero. Positive kurtosis indicates a peaked distribution and negative kurtosis indicates a flat distribution. Hence the annual (monsoon) rainfall distribution under consideration did not follow normal distribution. The correlations of rainfall pattern from site to site in different seasons are also presented in Fig Data and Methodology 3.1 Data Used Daily rainfall records for the ten selected sites of NER are considered in this study. The sites are selected based on the availability of continuous data. A minimum 45 years of continuous data for each station is utilized for the desired purpose. More specifically, 45 years ( ) data for Cherrapunjee and in remaining nine stations 50 years ( ) data is used. The daily rainfall data of each station is used to compute the monthly, seasonal and annual precipitation totals of respective station. On similar fashion, the frequencies of daily rainfall intensity of 2.5 mm/day, 1-3 cm/day, 4-6 cm/day, 7 cm/day for annual and monsoon season are considered for each station. IMD's criterion of rainy day i. e. rainfall 2.5mm per day (Forecaster's Guide, IMD) is applied to identify a wet spell (the number of consecutive rainy days). Following this definition, the 39

4 Table 1: Spatial distribution of rainfall (mm) over the study area based on long period averages ( ) Met-subdivisions Total Rainfall in mm Winter Pre-monsoon Monsoon Post-Monsoon Annual ARUNACHAL PRADESH ASSAM & MEGHALAYA NMMT Winter Pre-monsoon Monsoon Post-monsoon Fig 2: Correlation matrix of seasonal rainfall of the stations. frequency of wet spells of length at least of two days (2-d), 3 days (3-d) and 4 days (4-d) is computed for the yearly basis and monsoon season at each station. Monthly rainfall during the month of January, February, November and December for all the stations are found to be very small as compared to the other months of the year. Hence rainfall series of these months are not considered for trend analysis. The frequency of occurrence of rainfall 7 cm per day during the monsoon months at Imphal is very less and in some years it is nil. Therefore, this time series is also ignored in trend analysis. Similarly, the annual (monsoon) frequency of wet spells of length at least 5 days are very less and in many years it is nil for each of the stations. Therefore time series with frequency of wet spells of length of at least 5 days (5-d) is not considered for trend analysis. Altogether, 27 precipitation indices are considered in this study. The station wise details of the precipitation indices tested for trends are presented in Table 3. 40

5 Table 2a: The geographical locations of the stations and corresponding Mean (M), standard deviation (SD), median (Me), Kurtosis (Ck) and Skewness (Cs) of annual total rainfall of the stations during ( for Cherrapunjee) Table 2b: The geographical locations of the stations and corresponding Mean (M), standard deviation (SD), median (Me), Kurtosis (Ck) and Skewness (Cs) of monsoon rainfall of the stations during ( for Cherrapunjee) Met- Subdivision Station Latitude ( N) Longitude ( E) Elevation (m) Annual M (mm) SD (mm) Me (mm) Ck Cs Arunachal Pradesh Passighat Cherrapunjee Guwahati Dibrugarh Assam & North Meghalaya Lakhimpur Shillong Tezpur Agartala NMMT Kailashashar Imphal Met- Subdivision Latitude ( N) Longitude ( E) Elevation (m) Monsoon M (mm) SD (mm) Me (mm) Ck Station Cs Arunachal Pradesh Passighat Assam & Meghalaya NMMT Cherrapunjee Guwahati Dibrugarh North Lakhimpur Shillong Tezpur Agartala Kailashashar Imphal Adopted Methodology A comprehensive review of statistical approaches used for trend analysis in water resources data is provided by Helsel and Hirsch (1992). Studies like Lettenmaier et al. (1976), Hirsch et al. (1982), Berryman et al. (1988), and Hirsch et al. (1991) have recommended to use non-parametric methods as a general approach for the detection of trends in hydrology and water resources since in most of the cases it is not known a priori whether the data is normally distributed or not. Many studies indicate that non-parametric Mann-Kendall (M-K) test as an effective tool for identifying trends in hydrometeorological data (Hirsch et al., 1982; Lettenmaier et al., 1994; Xu et al., 2003; Ludwig et al., 2004) because it is simple, robust and allows missing values besides not being affected by gross data errors and outliers. In this study, K-test is used to detect the plausible positive or negative trends in each precipitation variable. 41

6 Precipitation Indices Totals Frequency Wet Spell Monthly Table 3: Station wise details of the precipitation indices tested for trends. Station A C G I K D NL P S T JANUARY FEBRUARY MARCH APRIL MAY JUNE JULY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER WINTER PRE- Seasonal POST- Annual 2.5 mm /day 1-3 cm /day 4-6 cm/ day 7 cm /day 2-d 3-d 4-d : Considered :Neglected A: Agartala; C: Cherrapunjee; G: Guwahati; I: Imphal; K: Kailashahar; D: Dibrugarh; NL: North Lakhimpur; P: Passighat; S: Shillong; T: Tezpur The null hypothesis of no trend versus the alternate hypothesis of the existence of trend is tested at a significant level 0.05 α = 0.05 significant level. Sen's non-parametric test (Sen, 1968; Gilbert, 1987) is applied for quantification of trend. 4. Results and Discussion The changing pattern and trends in rainfall both significantly depend on the data period (Kahya and Kalayci, 2004; Jain and Kumar, 2012). Burn and Elnur (2002) suggested that data record of 25 years is long enough for finding trends, if any exists in climate change research. In this study, M-K test is performed for trend with minimum 45 years of data. Details of trend analysis are presented in the following subsections. 42

7 Fig 3: Stationwise Sen s slope estimator for monthly, seasonal and annual total rainfall during ( for Cherrapunjee). ( ) indicates statistical significant increasing (decreasing) trend at 95% confidence level as per Mann Kendall test. 4.1Trends in Annual, Seasonal and Monthly Rainfall The results of trend analysis of annual, seasonal and monthly rainfall are presented in Table 4. Station 43

8 Fig 4: Stationwise time series plot of annual rainfall total with Sen's slope estimator. 44

9 Table 4: Sen s slope estimator for monthly, seasonal and annual total rainfall during ( for Cherrapunjee). Station MAR APR MAY JUN JUL AUG SEP OCT ANN WIN PREMON MON POSTMON Agartala Cherrapunjee * Guwahati * * Imphal * -2.19* * * Kailashashar * * Dibrugarh N Lakhimpur Passighat Shillong Tezpur * * *Indicates statistical significance at 95% confidence level as per Mann Kendall test (+ for increasing and for decreasing). wise Sen's slope estimator of these precipitation indices are depicted in Fig 3. The results reveal that except Agartala and Passighat, the annual rainfall at all other stations are increasing whereas at Agartala and - Passighat, it is decreasing. But the trends are not statistically significant. The time series plot of annual rainfall at each station with Sen's slope is presented in Fig 4. Likewise, no significant trend in monsoon rainfall has been observed at any station. In case seasonal rainfall, significant increasing trend is found in winter rainfall of Cherrapunjee at a positive rate 1.30 mm/yr while in all other stations, there is no significant trend in winter rainfall. For pre-monsoon season, rainfall at Imphal and Tezpur is increasing significantly at 2.73mm/yr and 2.26mm/yr respectively whereas other stations do not show any significant trends. A statistically significant increasing trend at 1.68 mm/yr is found at Guwahati in the post-monsoon rainfall and the rest of stations have no significant trend in rainfall during this season. On monthly scale, it is found that there is significant increase in rainfall at the rate 1.59 mm/yr during April at Tezpur and during May, rainfall is increasing significantly at 1.49 mm/yr and 3.36 mm/yr at Imphal and Kailashahar respectively. Similarly, positive rate of 1.18 mm/yr, 1.71 mm/yr and 1.83 mm/yr are observed in rainfall respectively at Imphal, Kailashahar and Guwahati in the month of October. Contrary to these, a significant decrease in rainfall at the rate mm/yr is found at Imphal during the month of June. No significant trends in monthly rainfall have been observed at the other stations. 4.2 Trends in Frequency of Rainfall Intensity Table 5 depicts the results of trend test for annual and monsoon seasonal frequency of rainfall 2.5 mm /day, 1-3 cm /day, 4-6 cm/day and 7 cm /day. The annual and monsoon seasonal frequency of 2.5 mm /day has no significant trend at any of the stations. Similar results hold good for frequency of 1-3 cm/day and 7 cm/ day at each station. On the other hand, the annual of frequency of 4-6 cm/day at Agartalais decreasing significantly at the rate 0.10/yr and no other station has any decisive trend. Similarly, significantly decreasing trends are observed at Agartala and Guwahati during the monsoon season in the frequency of 4-6 cm/day and are decreasing at the rate /yr and -0.05/yr respectively while no other station has any significant trend. 4.3 Trends in Wet Spells The trend test results subjected to the frequency of wet spells of length of 2-d, 3-d and 4-d for annual and monsoon season is presented in Table 6. The results show that there exists no significant trend in wet spell indices for the monsoon season. Similar results are also found for the three annual wet spell indices. 5. Conclusions In this study, M-K test is used for detecting trends in 27 precipitation indices for 10 major rain gauge stations of north-east region of India. 50 years of data for nine stations and 45 years of data for one station are analyzed for temporal trends in monthly, 45

10 Table 5: Sen s slope estimator for frequency of 2.5 mm/day, 1-3 cm/day, 4-6 cm/day and 7 cm/day for annual and monsoon season respectivelyduring ( for Cherrapunjee). Station 2.5 mm/day 1-3 cm/day 4-6 cm/day 7 cm /day ANN MON ANN MON ANN MON ANN MON Agartala * -0.06* Cherrapunjee Guwahati * Imphal Kailashashar Dibrugarh N-Lakhimpur Passighat Shillong Tezpur *indicates statistical significance at 95% confidence level as per Mann Kendall test (+ for increasing - for decreasing) Table 6: Sen s slope estimator for frequency of wet spells 2-d, 3-d and 4-d for annual and monsoon season respectivelyduring ( for Cherrapunjee). Station 2-d 3-d 4-d ANN MON ANN MON ANN MON Agartala Cherrapunjee Guwahati Imphal Kailashashar Dibrugarh N-Lakhimpur Passighat Shillong Tezpur seasonal, annual variation of total rainfall. Annual and monsoon rainfall have no significant trends at any of the selected stations, which is in agreement with previous findings (Deka et al., 2013; Jain et al., 2013; Das et al., 2011; Das and Goswami, 2003). However, on seasonal scale, rainfall at Imphal and Tezpur shows 46

11 significantly increasing trends during pre-monsoon season; rainfall at Guwahati has increased significantly during the post-monsoon and the winter rainfall at Cherrapunjee has also increased significantly during the study period. On monthly basis, increasing trend in rainfall is observed at Tezpur during April, at Impahal and Kailashahar during May and September and at Guwahat during October while the rainfall of June at Imphal has a decreasing tendency during the period The frequency of 4-6 cm/day have decreasing tendency at Agartala on annual scale while during the monsoon season, it has a decreasing tendency both at Agartala and Guwahati. Some recent studies on the rainfall variability and trends over northeast India (Deka et al., 2013; Jain et al., 2013; Das et al., 2011; Das and Goswami, 2003) have concluded that there is no significant trend in annual rainfall which is in agreement with the present study. Guhathakurta and Rajeevan (2006) have observed a decrease in winter rainfall in Nagalnd- Manipur-Mizoram-Tripura on meteorological subdivision scale. However, in the present study, no station of Nagalnd-Manipur-Mizoram-Tripura has References Attri SD and Tyagi A (2010). Climate profile of India. Met Monograph No. Environment Meteorology-01/2010. India Meteorological Department, p Bhatla R and Tripathi A (2014). The Study of Rainfall and Temperature Variability over Varanasi. International Journal of Earth and Atmospheric Science, 1(2): Burn DH and Elnur MAH (2002). Detection of hydrologic trends and variability. Journal of Hydrology, 255(1): Das PJ and Goswami DC (2003). Long-term variability ofrainfall over northeast India. Indian Journal of Landscape Systems and Ecological Studies, 26(1): Das S, Bhattacharjee K, Shaw SO, Pathak HG and Patowary B (2011). Characteristic pattern and recent trend in rainfall over Guwahat. Proceedings of "Water for Cities: responding to the Urban Challenges", Guwahati, May 30, Deka RL, Mahanta C, Pathak H, Nath KK and Das S (2013). Trends and fluctuations of rainfall regime in the Brahmaputra and Barak basins of Assam, India. Theoretical and Applied Climatology, 114(1-2): Forecaster's Guide; India Meteorological Department, p Gilbert RO (1987). Statistical methods for environmental pollution monitoring, Van Nostand Reinhold, New York. Goswami BN, Venugopal V, Sengupta D, Madhusoodanam MS and Xavier PK (2006). Increasing trends of extreme rain events over India in a warming environment. Current Science, 314: depicted a significant increase or decrease in winter rainfall. The absence of any significant trend in rainfall amount on monthly, seasonal and annual scale and in frequency of rainfall intensities of 2.5 mm/day, 1-3 cm/day, 4-6 cm/day, 7cm/day and wet spells of length of 2 days, 3 days and 4 days at most of the stations confirms that there is no clear cut indication of any major changes in rainfall pattern over NER in terms of rainfall intensity and persistence. However, further research is needed for better understanding of the temporal pattern of rainfall in the region. Acknowledgements The authors are thankful to the Director General of Meteorology, India Meteorological Department, for his encouragement to carry out the work and to Deputy Director General of Meteorology, Regional Meteorological Centre, Guwahati for providing the data to carry out the study. The authors are also thankful to the anonymous reviewers for their expert comments and valuable suggestions for improving the manuscript. Guhathakurta P, Menon P, Mazumdar AB and Sreejith OP (2010). Changes in extreme rainfall events and flood risk in India during the last century. Research Report No: 3/2010, National Climate Centre, India Meteorological Department, Pune. Guhathakurta P and Rajeevan M (2006). Trends in the rainfall pattern over India. Research Report No: 2/2006, National Climate Centre, India Meteorological Department, Pune, Guhathakurta P and Rajeevan M (2008). Trends in the rainfall pattern over India. International Journal of Climatology, 28: IPCC (2007). Summary for policymakers. In Climate Change 2007: The Physical Science Basis, Solomon, S., Qin, D., Manning, M. Chen, Z., Marquis, M., Averyt, K. B., Tignor, M., Miller, H.L. (eds).intergovernmental Panel on Climate Change, Cambridge University Press: UK. INCCA (2010). Climate change and India: A 4X4 assessment - A sectoral and regional analysis for 2030s. Ministry of Environment and Forests, Government of India. Jain SK and Kumar V (2012). Trend analysis of rainfall and temperature data for India. Current Science, 102: Jain SK, Kumar V and Saharia M (2013). Analysis of rainfall and temperature trends in northeast India. International Journal of Climatology, 33(4): Joshi UR and Rajeevan M (2006). Trends in precipitation extremes, India.Research Report No: 3/2006, National Climate Centre, India Meteorological Department, Pune. Kahya E and Kalayci S (2004). Trend analysis of streamflow in 47

12 Turkey. Journal of Hydrology, 289(2): Krishnakumar KN, Rao GSLHVP and Gopakumar CS (2009). Rainfall trends in twentieth century over Kerala, India. Atmospheric Environment, 43: Kumar V, Jain SK and Singh Y (2010). Analysis of longterm rainfall trends in India. Hydrology Science Journal, 20: Mooley DA and Parthasarthy B (1984). Fluctuations of all India summer monsoon rainfall during Climatic Change, 6: Parthasarathy B and Dhar ON (1975). Trend analysis of annual Indian rainfall. Hydrology Science Bulletin, 20: Rakhecha PR and Soman MK (1994). Trends in the annual extreme rainfall events of 1 to 3 days duration over India. Theoretical and Applied Climatology, 48: Ravindranath NH, Rao S, Sharma N, Nair M, Gopalakrishnan R, Rao AS, Malaviya S, Tiwari R, Sagadevan A, Munsi M, Krishna N and Bala G (2011). Climate change vulnerability profiles for North East India. Current Science, 101: Rupa Kumar K, Pant GB, Parthasarathy B and Sontakke NA (1992). Spatial and sub-seasonal patterns of the long term trends of Indian summer monsoon rainfall. Internaional Journal of Climatology, 12: Sarkar RP and Thapliyal V (1988). Climate change and variability. Mausam, 39: Sen Roy S and Balling RC (2004). Trends in extreme daily precipitation indices in India. International Journal of Climatology, 24: Sen PK (1968). Estimates of the regression coefficient based on Kendall's tau. Journal of American Statistical Association, 63: Soman MK, Krishna Kumar K and Singh N (1988). Decreasing trend in the rainfall of Kerala. Current Science, 57: Srinivasan V, Raman S and Mukherjee S (1972). South- West Monsoon Typical situation over West Bengal and Assam and adjoining states. IMD FMU Report III 3.6, Thapliyal V and Kulshreshtha SM (1991). Climate changes and trends over India. Mausam, 42: