* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Spatial and temporal trends of mean and extreme
Survey
Document related concepts
Surveys of scientists' views on climate change wikipedia , lookup
General circulation model wikipedia , lookup
Effects of global warming on human health wikipedia , lookup
Climate change and poverty wikipedia , lookup
Early 2014 North American cold wave wikipedia , lookup
Climate sensitivity wikipedia , lookup
Attribution of recent climate change wikipedia , lookup
Climatic Research Unit documents wikipedia , lookup
Climate change, industry and society wikipedia , lookup
Global warming hiatus wikipedia , lookup
Effects of global warming on humans wikipedia , lookup
Urban heat island wikipedia , lookup
IPCC Fourth Assessment Report wikipedia , lookup
North Report wikipedia , lookup
Transcript
Atmospheric Research 138 (2014) 73–90 Contents lists available at ScienceDirect Atmospheric Research journal homepage: www.elsevier.com/locate/atmos Spatial and temporal trends of mean and extreme rainfall and temperature for the 33 urban centers of the arid and semi-arid state of Rajasthan, India Santosh M. Pingale a,⁎, Deepak Khare a, Mahesh K. Jat b, Jan Adamowski c a b c Department of Water Resource Development and Management, Indian Institute of Technology Roorkee, Roorkee 247 667 UK, India Department of Civil Engineering, Malaviya National Institute of Technology, Jaipur, Rajasthan, India Department of Bioresource Engineering, McGill University, Quebec, Canada a r t i c l e i n f o Article history: Received 7 May 2013 Received in revised form 21 September 2013 Accepted 29 October 2013 Keywords: Extreme events Interpolation Mann–Kendall test Temperature Rainfall India a b s t r a c t Trend analysis of the mean (monsoon season, non-monsoon season and annual) and extreme annual daily rainfall and temperature at the spatial and temporal scales was carried out for all the 33 urban centers of the arid and semi-arid state of Rajasthan, India. Statistical trend analysis techniques, namely the Mann–Kendall test and Sen's slope estimator, were used to examine trends (1971–2005) at the 10% level of significance. Both positive and negative trends were observed in mean and extreme events of rainfall and temperature in the urban centers of Rajasthan State. The magnitude of the significant trend of monsoon rainfall varied from (−) 6.00 mm/hydrologic year at Nagaur to (−) 8.56 mm/hydrologic year at Tonk. However, the magnitude of the significant negative trends of non-monsoon rainfall varied from (−) 0.66 mm/hydrologic year at Dungarpur to (−) 1.27 mm/hydrologic year at Chittorgarh. The magnitude of positive trends of non-monsoon rainfall varied from 0.93 mm/hydrologic year at Churu to 1.70 mm/hydrologic year at Hanumangarh. The magnitude of the significant negative trends of annual rainfall varied from (−) 6.47 mm/year at Nagaur to (−) 10.0 mm/year at Tonk. The minimum, average and maximum temperature showed significant increasing warming trends on an annual and seasonal scale in most of the urban centers in Rajasthan State. The magnitude of statistically significant annual extreme daily rainfall varied from 2.00 mm at Jhalawar to (−) 1.64 mm at Tonk, while the magnitude of statistically significant extreme annual daily minimum and maximum temperature varied from 0.03 °C at Ganganagar to 0.05 °C at Jhalawar, respectively. The spatial variations of the trends in mean (monsoon season, non-monsoon season and annual) and extreme annual daily rainfall and temperature were also determined using the inverse-distance-weighted (IDW) interpolation technique. IDW results are helpful to identify trends and variability in mean and extreme rainfall and temperature in space and time for the study locations where the data is not available and the quality of data is not good. These spatial maps of temperature and rainfall can help local stakeholders and water managers to understand the risks and vulnerabilities related to climate change in terms of mean and extreme events in the region. © 2013 Elsevier B.V. All rights reserved. 1. Introduction Increasing urbanization, populations and economies are causing adverse impacts on the natural environment. With ⁎ Corresponding author currrently at: Department of Water Resources and Irrigation Engineering, Arba Minch University, Ethiopia. E-mail address: [email protected] (S.M. Pingale). 0169-8095/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.atmosres.2013.10.024 increases in urbanization, problems related to climate change are becoming worse, since additional pervious areas are becoming impervious, leading to decreases in vegetation cover and increases in temperature. Rapid economic development and population growth in various parts of the world such as India have caused concerns regarding the quantity and quality of natural resources, and in particular water resources. As such, information on the magnitude, duration and frequency of 74 S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 extreme events (such as floods and droughts), along with trends in various hydro-meteorological variables, is required for the planning of optimum adaptation strategies for the sustainable use of natural resources such as water (e.g., Kampata et al., 2008; Some'e et al., 2012; Huang et al., 2013). Mall et al. (2006) have studied the potential for sustainable development of surface water and groundwater resources within the constraints imposed by climate change. The authors suggested that little work has been done on hydrological impacts of climate change for Indian regions/basins. Therefore, assessment of climate change is required in India which considers the causative factors of climate change (i.e., anthropogenic factors, global warming in terms of variations in meteorological parameters) at spatial and temporal scales. Further, it is essential in hydrological modeling to assess the adverse impacts of climate change on water resources. Such studies are even more significant for urbanized watersheds/ areas to better understand the relationship between climate change and hydrological processes, which will facilitate the sustainable utilization of water resources. There are different methods that can be used in the assessment of climate change. Various parametric and non-parametric statistical tests have been used in the recent past by many researchers to assess trends in hydrometeorological time series in India (e.g., Gadgil and Dhorde, 2005; Singh et al., 2008a, 2008b; Basistha et al., 2009; Kumar et al., 2010; Pal and Al-Tabbaa, 2010; Deka et al., 2012; Duhan et al., 2013; Jeganathan and Andimuthu, 2013; Jain et al., 2013), and in other regions of the world (e.g., Zhang et al., 2000; Zhang et al., 2001; Roy and Balling, 2004; Cheung et al., 2008; Motiee and McBean, 2009; Sahoo and Smith, 2009; Zin et al., 2010; Tabari et al., 2011; Nalley et al., 2012; Nemec et al., 2012; Saboohi et al., 2012; Jiang et al., 2013). Detection of past trends, changes, and variability in the time series of hydro-climatic variables is very important in understanding the potential impact of future changes in the region (Sahoo and Smith, 2009). Statistical analysis could be extended to analyze climatic parameters and their relationships with water resources, land use/cover changes, urbanization etc. It has been seen that the Earth's average temperature has increased by 0.6 °C in the later part of the 20th century; there is also a dramatic shift in temperature change from a minimum of 1.4 °C to a maximum of 5.4 °C as reported by the projections made by various climate prediction models (IPCC, 2001). The IPCC has demonstrated that climate change assessments at sub-national (i.e., state or province) and local scales are needed. The IPCC has also found several gaps in the knowledge that exists in terms of observations and research needs related to climate change and water (IPCC, 2007). These include GCM scale resolutions and downscaling techniques to locale scales to assess the climate change and to quantify the contribution of anthropogenic activity to climate change etc. It has been found that seasonal and annual air temperatures have been increasing at a rate of 0.57 °C per 100 years (1881 to 1997) in India (Pant and Kumar, 1997). The trends of certain hydro-climatic variables have been studied in Indian urban centers such as Hyderabad, Patna, Ahmedabad, Surat, Bangalore, Mumbai, Nagpur, Pune, Jaipur, Chennai, New Delhi, Kanpur, Lucknow and Kolkata, whose populations are more than 1 million (De and Rao, 2004). Significant increasing trends were found in annual and monsoon rainfall over Chennai, New Delhi, Kolkata and Mumbai. Other climatic parameters were not considered to assess climate change impacts on timing, distribution, pattern, frequency, variability of precipitation and their inter-relationships in urban cities. Arora et al. (2005) explored trends of annual average and seasonal temperatures using the MK test at the country and regional scales in India. It was observed that annual mean temperature, mean maximum temperature and mean minimum temperature have increased at the rates of 0.42, 0.92 and 0.09 °C (100 year)−1, respectively. On a regional basis, stations in Southern and Western India show rising trends of 1.06 and 0.36 °C (100 year)−1, respectively, while stations in the North Indian plains show a falling trend of 0.38 °C (100 year)−1. The seasonal mean temperature has increased by 0.94 °C (100 year)−1 for the post-monsoon season and by 1.1 °C (100 year)−1 for the winter season. This study deals only with trend detection of temperature; in addition, trends in annual extreme daily events of rainfall and temperature have not been studied. Basistha et al. (2007) assessed the spatial trends of rainfall over Indian subdivisions from 1872 to 2005. Their results show decreasing trends of rainfall over North India excluding Punjab, Haryana, West Rajasthan, and Saurashtra, and increased trends in South India excluding Kerala and Madhya Maharashtra. The study also concluded that the arid portion (i.e., western part) of India has not been investigated in detail. Further, they highlighted that more research is required to assess the spatial patterns of trends of other climatic variables (average, minimum and maximum temperatures, relative humidity, wind speed, evapotranspiration (ET), number of rainy days etc.) and their inter-relationships including trends of annual and seasonal climatic parameters to assess the impacts of climate change. Gowda et al. (2008) studied the region of Davangere district over a period of 32 years using statistical analysis. Climatic parameters (i.e., rainfall, relative humidity, maximum temperature, minimum temperature, sunshine hour and wind speed) were analyzed to assess climate change. The statistical analysis showed that such a small data set may not represent the correct picture of climate change and requires long term data. A significant change in climatic variables was found in and around the Davangere region. Ghosh et al. (2009) observed the varied trend in Indian summer monsoon rainfall, which has not only been affected by global warming, but may also be affected by local changes due to rapid urbanization, industrialization and deforestation. Patra et al. (2012) detected rainfall trends in the twentieth century in India using parametric and nonparametric statistical trend analysis tests. The temporal variation in monthly, seasonal and annual rainfall was studied for the Orissa State using data from 1871 to 2006. The analysis revealed a long term, insignificant, declining trend in annual as well as monsoon rainfall, and an increasing trend in the post-monsoon season. Rainfall during winter and summer seasons showed an increasing trend. The study only assessed rainfall for Orissa State; trends in mean and extreme annual daily events were not explored. Duhan and Pandey (2013) investigated the spatial and temporal variabilities of precipitation in 45 districts of Madhya Pradesh (MP), India on an annual and seasonal basis. The MK test and Sen's slope estimator test were used for trend analysis, and an increase and decrease in the precipitation trend were found on annual and seasonal bases, respectively, in the districts of MP. In this study, little attention was given to urban trends of S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 mean and extreme rainfall and temperature. A literature review shows that in India climate related studies are restricted to statistical analysis of some of the meteorological parameters only. The effects of climate change on water resources were not considered and seasonal changes are not clearly indicated with mean and extreme events, especially in the urban centers. As described above, the analysis of extreme trend and variability in annual daily rainfall and temperature has been carried out in recent years by a number of researchers in various parts of the world, including India. However, there is a still a lack of information on trends and variability of extreme annual daily events, especially for the semi-arid and arid region of Rajasthan, India. For the region of Rajasthan, there is a lack of studies of climate change assessments that consider various region specific aspects, and their inter-relation with the climatic parameters at the spatial and temporal scales. Climate change in terms of inter-seasonal and inter-annual variations in mean and extreme annual daily rainfall and temperature, for urban centers of arid and semi-arid regions in India, has not been reported in much detail in the literature. In light of this, the present study was aimed at studying climate change in terms of mean (monsoon, non-monsoon and annual) and extreme annual daily events in the 33 main urban centers of Rajasthan in India via trend analysis of two important climatic parameters (i.e., rainfall and temperature (minimum, average and maximum)) at the spatial and temporal scales. Statistical trend analysis techniques, namely the Mann–Kendall (MK) test (Mann, 1945; Kendall, 1975) and Sen's slope estimator (Sen, 1968), were used in the present study. These methods have several advantages over parametric methods (Duhan and Pandey, 2013). The paper is organized in five sections. Section 2 describes the study area and data used. Section three describes the methodology adopted in the study. Sections 4 and 5 describe the findings of the study on climate change assessment in terms of trends in mean (monsoon season, non-monsoon season and annual) and extreme annual daily rainfall and temperature in the 33 urban centers of Rajasthan State in India. Further, the spatial variations in mean (seasonal and annual) and extreme annual daily rainfall and temperature with their trends are presented using the inverse-distance-weighted (IDW) interpolation method. 2. Study area and data used Rajasthan is the largest state in India, with an area of 342,000 km2 of which about 60% falls under the arid category. Administratively, the state is divided into 33 districts (Fig. 1) and its climate varies from arid to semi-arid. Geographically, Rajasthan State extends from 23° 3.50′ to 30° 14′N latitude and 69° 27′ to 78° 19′E longitude. The study region has unique climatological characteristics (low and erratic rainfall, extremes of diurnal and annual temperatures, low humidity and high wind velocity). It also has unique physiographic characteristics. For example, the Arawali Hill Range is unique because it lies along the direction of the southwest monsoon, which brings rains in the region. There is also the Great Indian Desert (called the Thar Desert) in the northwestern part of Rajasthan, which is India's largest desert, covering almost 70% of the state; in this region there is less vegetation and the region has a significant influence on rainfall and temperature distribution across the 75 state. The climate in the state varies from arid to semi-arid. This makes Rajasthan State different from other regions in neighboring states. The state is bordered by Gujarat, Madhya Pradesh, Uttar Pradesh, Haryana and Punjab, and is also bordered with Pakistan. The climate in Rajasthan State is characterized by low and erratic rainfall, extremes of diurnal and annual temperatures, low humidity and high wind velocity. The annual average rainfall is highly variable, and it is most erratic in the western region with frequent dry spells, punctuated occasionally by heavy downpours in some years associated with passing low pressure systems over the region. The western and southern districts of Rajasthan State frequently experience severe droughts. Coping with low rainfall, with a high coefficient of variation across time and space, is the major challenge (Rathor, 2005). Gridded data sets of rainfall and temperature have been used in many hydrological and climatological studies worldwide, including Australia, for hydro-climatic forecasting, climate attribution studies and climate model performance assessments (e.g., Tozer et al., 2012; Dash et al., 2013). Therefore, in the present study 0.5° × 0.5° and 1° × 1° gridded data sets of daily rainfall and temperature from the period 1971 to 2005 were used for the spatial and temporal variability and trend analyses in the present study. These data were obtained from the Indian Meteorological Department (IMD), Pune. From all of India, 395 stations were used for developing the high resolution daily gridded data sets (1° × 1°) of daily temperature (minimum, maximum and mean) for the Indian region by the Indian Meteorological Department (IMD) — Pune. The modified Shepard's angular distance weighting algorithm was used for interpolating the stations of temperature data into 1° × 1° grids. These data sets were evaluated using the cross validation technique to estimate the interpolation error (root mean square error), which was found to be less than 0.5 °C. These data sets were also compared with another (Asian Precipitation — Highly Resolved Observational Data Integration Towards the Evaluation of Water Resources (APHRODITE)) high resolution monthly data set, and a correlation of more than 0.8 was found in most parts of the country. Similarly, high resolution daily gridded data sets of rainfall (0.5° × 0.5°) were also developed (using Shepard's method), and validated using the network of 6076 rain gage stations for the Indian region by IMD, Pune. Additional detail on the methodology is given in IMD reports (Rajeevan and Bhate, 2008; Srivastava et al., 2008). These data sets have been found to be useful for various regional applications (such as studies of extreme temperatures, validation of numerical and climate model simulations and many environmental applications including trend analysis of mean and extreme events of rainfall and temperature) (Rajeevan and Bhate, 2008 and Srivastava et al., 2008), and they can be used for water resource planning purposes. The data was subjected to thorough quality checks (location of stations, typing error, missing data etc.) before the data analysis was performed. Hamlet et al. (2005) have reported that existing gridded data sets are not appropriate for the long term trend estimation of precipitation and temperature fields for the continental United States. The gridding procedures introduce artificial trends due to the incorporation of stations with different record lengths and locations. 76 S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 Fig. 1. Location of urban centers considered in the state of Rajasthan, India. However, the gridded set developed by IMD, Pune addressed these issues before and after developing the gridded data sets from the observed station data before releasing the data to users. In addition, these data sets were validated with other gridded data sets developed in the region, and their application was demonstrated successfully through case studies and by other researchers (e.g., Dash et al., 2013). Basic data quality checks (such as rejecting values, greater than exceeding known extreme values, minimum temperature, greater than maximum temperature, same temperature values for many consecutive days, unusual high values, homogeneity, location of the station etc.) were performed for each station, and then stations were selected based on data quality or the development of the gridded data sets of temperature and rainfall. Further, homogeneity, persistence and periodicity in the data were also analyzed by autocorrelation. Normalizing the time series before the analysis was performed at each urban center. The district headquarters of Rajasthan State, which has the largest population, greatest developmental activities, and administrative centers of the districts, are considered as the ‘urban centers’. The geographical extent of the urban centers is defined by their municipal S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 boundaries defined by the development authorities. The sizes and populations of the urban centers are included in Table 1. The climate of individual urban centers has been extracted from the gridded surface of temperature and rainfall corresponding to the municipal limits of the urban centers. Therefore, it preserves the meteorological characteristics of individual urban centers. The IDW assumes that the values of the variables at a point to be predicted are similar to the values of nearby observation points. The majority of the stations are located in the urban areas. Therefore, this has a major influence of urban climate. The grid covering the urban center of the district headquarters of Rajasthan State is considered to represent the climate of the urban center. First, the grids that represent the urban center from the developed data sets are identified, and then daily data (1971–2005) was extracted. The daily rainfall and temperature (minimum, average and maximum) data were extracted for the 33 urban centers of district headquarters in Rajasthan State (Table 1). These urban centers were selected taking into account the length and availability of records, so that most of the urban centers in the arid and semi-arid region of Rajasthan State were covered by the corresponding data to assess trends in mean (seasonal and annual) and extreme annual daily rainfall and temperature. The grids covering the urban centers were chosen 77 for extracting the daily rainfall and temperature data, and it was assumed that the grids covering the urban center represent the rainfall and temperature of that urban center in Rajasthan State over a period of time (Fig. 1 and Table 1). Further, the rainfall and temperature (minimum, average and maximum) data were arranged according to monsoon season, non-monsoon season and annual scale from 1971 to 2005 in each urban center. The monsoon season represents the months of June to September, and the non-monsoon season represents the months of January to May and October to December. 3. Methodology 3.1. Normalization and autocorrelation analysis of time series Normalized rainfall and temperature time series were used to test the outliers present in the time series data using following relationship (Rai et al., 2010): Xt ¼ ðxt ‐xÞ=σ ð1Þ where Xt is the normalized anomaly of the series, xt is the observed time series, and x and σ are the long-term mean and standard deviation of annual/seasonal time series, respectively. Table 1 List of urban centers, size and population. ID Urban center Longitude Latitude Elevation (m.a.s.l.), m Size (km2)a approx. Population (2011 census) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Ajmer Alwar Banaswara Baran Barmerb Bharatpur Bhilwara Bikaner Bundi Chittorgarh Churu Dausab Dholpur Dungarpur Ganganagar Hanumangarh Jaipur Jaisalmer Jalore Jhalawarb Jodhpur Jhunjhunu Karauli Kota Nagaur Pali Rajsamandb Sawai Madhopur Sikar Sirohib Tonk Udaipur Pratapgarh 74° 76° 74° 76° 71° 77° 74° 73° 75° 74° 75° 76° 77° 73° 73° 74° 75° 70° 72° 76° 73° 75° 77° 75° 73° 73° 73° 76° 75° 72° 75° 73° 74° 26° 27° 23° 25° 25° 27° 25° 28° 25° 24° 28° 26° 26° 23° 29° 29° 26° 26° 25° 24° 26° 28° 26° 25° 27° 25° 25° 26° 27° 24° 26° 24° 24° 486 271 302 262 227 183 421 242 268 394 292 333 177 225 164 177 431 229 178 312 231 323 275 271 302 214 547 266 427 321 289 598 491 31 – – – – – – 65 – – – – – – – – 168 – – – 83 – – 62 – – – – – – – – – 542,580 315,310 100,128 118,157 83,517 252,109 360,009 647,804 102,823 116,409 119,846 61,589 126,142 43,000 224,773 151,104 3,073,350 78,000 54,185 48,054 1,033,918 118,966 105,690 1,001,365 100,618 229,956 55,671 120,998 237,579 35,544 165,363 451,735 48,900 42′ E 38′ E 24′ E 33′ E 25′ E 30′ E 40′ E 22′ E 41′ E 42′ E 01′ E 20′ E 52′ E 43′ E 50′ E 21′ E 52′ E 57′ E 58′ E 10′ E 04′ E 20′ E 04′ E 52′ E 40′ E 25′ E 53′ E 21′ E 15′ E 51′ E 47′ E 42′ E 47′ E 27′ 34′ 30′ 05′ 45′ 15′ 21′ 01′ 27′ 54′ 19′ 53′ 41′ 50′ 49′ 35′ 55′ 55′ 22′ 35′ 18′ 06′ 30′ 10′ 00′ 46′ 04′ 01′ 36′ 53′ 10′ 34′ 02′ N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N a For size of urban center of district headquarter Rajasthan see e.g., Jhalawar, https://maps.google.co.in/maps?q=Jhalawar&ie=UTF8&hq=&hnear= 0x39653ede53cfaf07:0x31263af08bba421e,Jhalawar,+Rajasthan&gl=in&ei=TekFUuK8G4bLrQfk74F4&ved=0CLABELYD. More details are also available in http://www.citypopulation.de/php/india-rajasthan.php. b Population census 2001. 78 S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 The autocorrelation test was performed to check the randomness and periodicity in the time series (Modarres and da Silva, 2007). If lag-1 serial coefficients are not statistically significant then the MK test can be applied to the original time series (Luo et al., 2008; Karpouzos et al., 2010). The serial correlation coefficients of the normalized climatic series were computed for lags L = 0 to k, where k is the maximum lag (i.e., k = n/3), and n is the length of the series. The autocorrelation coefficient rk of a discrete time series for lag-k was estimated as: The Kendall test statistic S can be computed as: S¼ n −1 X sgn x j −xk ð5Þ k¼1 where sgn(xj − xk) is the signum function. The test statistic S is assumed to be asymptotically normal, with E(S) = 0 for the sample size n ≥ 8 and variance as follows: ½nðn−1Þð2n þ 5Þ− n‐k X Xt −Xt Xtþk −Xtþk ð2Þ k¼1 where rk is the lag-k serial correlation coefficient. The hypothesis of serial independence was then tested by lag-1 autocorrelation coefficient as H0 : r1 = 0 against H1 : |r1| ≥ 0 using the test of significance of serial correlation (Yevjevich, 1971; Rai et al., 2010): tðt−1Þð2t þ 5Þ t VðSÞ ¼ rk ¼ " k¼1 #0:5 n‐k X 2 2 Xt −Xt Xtþk −Xtþk X ð6Þ 18 where ti denotes number of ties up to sample i. The standardized MK test statistic (Zmk) can be estimated as: Zmk 8 S−1 > > > pffiffiffiffiffiffiffiffiffiffi if S N 0 > < VðSÞ ¼ 0 if S ¼ 0 > > Sþ1 > > p ffiffiffiffiffiffiffiffiffi ffi if S b 0 : VðSÞ : ð7Þ 0:5 ðrk Þtg ¼ −1 tg ðn‐k‐Þ n‐k ð3Þ where ðrk Þtg is the normally distributed value of rk, and tg is the normally distributed statistic at ‘g’ level of significance. The values of tg are 1.645, 1.965 and 2.326 at the significance level of 0.10, 0.05 and 0.01, respectively. The null hypothesis of serial independence was rejected at the significance level α (here 0.05), if jrk j ≥ ðrk Þtg . For the non-normal time series data, the MK test is an appropriate choice for trend analysis (Yue and Pilon, 2004). Therefore, the MK test was used in the present study, where autocorrelation was found to be non-significant at the 5% level of significance. 3.2. Mann–Kendall test p ¼ 0:5−φðjZmk jÞ The Mann–Kendall (MK) test is a non-parametric test that can be used for detecting trends in a time series (Mann, 1945) where autocorrelation is non-significant. The non-linear trend, as well as the turning point can be derived from Kendall test statistics (Kendall, 1975). This method searches for a trend in a time series without specifying whether the trend is linear or non-linear. It has been found to be an excellent tool for trend detection and many researchers have used this test to assess the significance of trends in hydroclimatic time series data such as water quality, stream flow, temperature and precipitation (e.g., Ludwig et al., 2004; Zhang et al., 2004; McBean and Motiee, 2008; Basistha et al., 2009; Rai et al., 2010; Patra et al., 2012). The MK test can be applied to a time series xi ranked from i = 1, 2 … n − 1 and xj ranked from j = i + 1, 2 … n such that: 8 > 1 if x j −xi N0 > > < 0 if x j −xi ¼ 0 sgn x j −xi ¼ > > > : −1 if x −x b0 j i The standardized MK test statistic (Zmk) follows the standard normal distribution with a mean of zero and variance of one. If ±Zmk ≤ Zα/2 (here α = 0.1), then the null hypothesis for no trend is accepted in a two sided test for trend, and the null hypothesis for the no trend is rejected if ± Zmk ≥ Zα/2. Failing to reject H0 (i.e., null hypothesis) does not mean that there is no trend. Rather, it is a statement that the evidence available is not sufficient to conclude if there is a trend (Helsel and Hirsch, 2002). A positive value of Zmk indicates an ‘upward trend’ and a negative value indicates a ‘downward trend’. The significance levels (p-values) for each trend test can be obtained from the relationship given as (Coulibaly and Shi, 2005): : ð4Þ ð8Þ where φ() denotes the cumulative distribution function (CDF) of a standard normal variate. At a significance level of 0.1, if p ≤ 0.1 then the existing trend is considered to be statistically significant. 3.3. Sen's estimator of slope and percentage change over a period If a linear trend is present in a time series, then the true slope of the trend is estimated using a simple non-parametric procedure (Theil, 1950; Sen, 1968), which is given by: Q i ¼ median x j −xk ∀k ≤ j j−k ð9Þ where xj and xk are data values at times j and k (j N k), respectively. The median of N values of Qi is Sen's estimator of slope. If N is odd, then Sen's estimator is computed by Qmed = Q(N + 1)/2 and if N is even, then Sen's estimator is computed by Qmed = [QN/2 + Q(N + 2)/2]/2. Finally, Qmed is S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 tested by a two-sided test at 100(1 ‐ α)% confidence interval, and the true slope is obtained by a non-parametric test. The percentage change over a period of time can be estimated from Theil and Sen's median slope and mean by assuming the linear trend (Yue and Hashino, 2003; Basistha et al., 2009) in the time series: % change ¼ median slope length of period : mean ð10Þ 79 in rainfall and temperature were also determined. The corresponding spatial and temporal distributions of the mean (monsoon season, non-monsoon season and annual) and extreme events of annual daily rainfall and temperature using the MK test statistics (Zmk) were determined by the IDW interpolation technique. The increasing, decreasing, and no trend at 10% level of significance are represented with the symbols ↑, ↓ and О, respectively. 4.1. Trend analysis of mean events 3.4. Interpolation technique In the present study, the spatial variation and trends of the mean (monsoon, non-monsoon and annual) and extreme annual daily rainfall and temperature were determined using the inverse-distance-weighted (IDW) interpolation technique. The IDW method is a simple and effective interpolation technique that is based on the assumption that the values of the variables at a point to be predicted are similar to the values of nearby observation points. This method implies that each station has a local influence, which decreases with distance by means of the use of a power parameter (Isaaks and Srivastava, 1989; Robinson and Metternicht, 2006). The general formula of IDW (Yavuz and Erdogan, 2012) is as follows: ^ ðS Þ ¼ m Z 0 N X λi ZðSi Þ ð11Þ i¼1 where ^ ðS0 Þ Z N λi Z(Si) is the value of prediction for location S0 is the number of measured sample points surrounding the prediction location is the weight assigned to each measured point is the observed value at the location Si. The weighting is a function of inverse distance. The formula determining the weighting is (Yavuz and Erdogan, 2012): λi ¼ d−p io N X −p dio i¼1 : ð12Þ The IDW method requires the choice of a power parameter and a search radius. The power parameter (p) controls the significance of measured values on the interpolated value based upon their distance from the output point (d) (Erdoğan, 2009). The choice of a relatively high power parameter ensures a high degree of local influence and gives the output surface in more detail (Yavuz and Erdogan, 2012). 4. Results and discussion Detailed trend analysis of the mean (monsoon season, non-monsoon season and annual) and extreme annual daily rainfall and temperature was carried out for the period of 1971–2005 using the Mann–Kendall test for all 33 urban centers of Rajasthan State. Similarly, Sen's slope and percentage change 4.1.1. Rainfall The spatial distributions of trends in seasonal (monsoon and non-monsoon) and annual average rainfall for the 33 urban centers of Rajasthan State are shown in Fig. 2. The magnitude of the Sen's slope and percentage change in average rainfall at seasonal (monsoon and non-monsoon) and annual temporal scales for the 33 different urban centers of Rajasthan State is also given in Tables S1 to S3 (Supporting material), respectively. Both positive and negative trends were observed by the MK test and Sen's slope estimator in seasonal and annual average rainfall in the urban centers. This is in good agreement with observations in different parts of India (Pal and Al-Tabbaa, 2011). All of the significant trends in monsoon and annual scale were found to be decreasing at the 10% level of significance. However, both positive and negative trends in non-monsoon average rainfall were observed. As a result, the expected decrease in rainfall will possibly cause a decrease in water availability in the future (Gosain et al., 2006; Tabari and Talaee, 2011b). The monsoon season rainfall pattern and trends were very similar to the annual ones, which demonstrate the major contribution of monsoon rainfall in annual rainfall in the urban centers of Rajasthan State. Similar results indicating the major contribution of seasonal precipitation in annual precipitation have also been observed in Iran (Tabari and Talaee, 2011b). Significant decreasing trends were observed in monsoon average rainfall for Jodhpur (at −6.82 mm/hydrologic year and −88.16%), Karauli (at −7.59 mm/hydrologic year and −39.03%), Nagaur (at −6.00 mm/hydrologic year and −66.92%) and Tonk (at −8.56 mm/hydrologic year and −50.63%). The significant decreasing trends of average monsoon rainfall had the highest value of decreasing slope (i.e., −8.56 mm/hydrologic year) for Tonk, and percentage change of −88.16% in Jodhpur (Table S1). Significant trends in non-monsoon average rainfall were found for Dungarpur (at −0.66 mm/hydrologic year and −58.10%), Chittorgarh (at −1.27 mm/hydrologic year and −80.29%), Jaipur (at 1.22 mm/hydrologic year and 61.53%), Churu (at 0.93 mm/hydrologic year and 49.61%), Ganganagar (at 1.39 mm/hydrologic year and 79.46%) and Hanumangarh (at 1.70 mm/hydrologic year and 102.85%). A significant increasing trend was observed in non-monsoon rainfall at Hanumangarh (1.70 mm/hydrologic year), and a decreasing trend was found at Chittorgarh (−1.27 mm/hydrologic year) (Table S2). The percentage change of significant trend was found to be at 102.85% and −80.29% in Hanumangarh and Chittorgarh, respectively (Table S2). Significant decreasing trends were observed in annual average rainfall for Jodhpur (at −7.51 mm/year and −85.05%), Karauli (at −7.41 mm/year and −34.65%), Nagaur (at −6.47 mm/year and −61.93%) and Tonk (at −10 mm/year and −54.54%). In 80 S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 Fig. 2. Location of urban centers with increasing, decreasing and no trends at 10% significance level for a) monsoon, b) non-monsoon and c) annual average rainfall. S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 the case of annual average rainfall, maximum values of decreasing slopes (−10 mm/year) and percentage change (−85.05%) were observed in Tonk and Jodhpur, respectively (Table S3). The IPCC also reported increases in precipitation at a global level, however, in the present study, precipitation was found to increase in some urban centers and decrease in other centers. This may be due to changing land use/land cover, and several other anthropogenic activities at a local scale (Gowda et al., 2008). 4.1.2. Minimum temperature Significant increasing trends were observed in minimum temperature at seasonal and annual temporal scales for the majority of the urban centers in the northeastern part of Rajasthan State. However, significant trends were not found in other urban centers at a 10% level of significance by the MK test (Fig. 3). The magnitude of the Sen's slope of significant trends in seasonal (monsoon, non-monsoon and annual) minimum temperature for the urban centers of Rajasthan State is shown in Fig. 3. The maximum value of Sen's slope (0.04 °C/hydrologic year) and percentage change (5.07%) for a significant increasing trend of average minimum temperature in the monsoon season were observed in Churu and Sikar, respectively. In the case of non-monsoon average minimum temperature, maximum values of increasing (0.05 °C/hydrologic year) slope were observed in Bikaner, Jaipur and Sikar. Similarly, percentage change in non-monsoon minimum temperature varied from 3.91% at Bharatpur to 13.45% at Sikar. Also, a significant trend in annual average minimum temperature had maximum values of increasing slopes (0.05 °C/year) at Sikar. The percentage changes in annual average minimum temperature were found to be at a minimum (2.61%) and maximum (9.52%) at Banaswara and Sikar, respectively. The positive trends of seasonal and annual minimum temperatures were also observed by other researchers in India (e.g., Jain et al., 2013). 81 4.1.4. Maximum temperature Significant increasing trends in monsoon average maximum temperature were found in Barmer, Bikaner, Dausa, Jaisalmer, Jalore and Pali (Fig. 5a). The significant increasing trends in monsoon season average maximum temperature were observed with a maximum value (0.03 °C/hydrologic year) of Sen's slope in Jaisalmer and Bikaner. Percentage change in average maximum temperature during the monsoon period was at a minimum (2.09%) and maximum (2.90%) in Jalore and Bikaner, respectively. In the case of non-monsoon average maximum temperature, the southwestern urban centers in Rajasthan State had significant increasing trends. No significant trend was found in the remaining urban centers during the non-monsoon period at a 10% level of significance (Fig. 5b). Significant increasing trends were observed in annual average maximum temperature for urban centers that are located in the southeast and southwestern parts of Rajasthan State (Fig. 5c). The significant increasing trends in non-monsoon season average maximum temperature were observed with a maximum value of Sen's slopes (0.03 °C/hydrologic year) in Barmer, Jaisalmer, Jalore, Pali, Sirohi and Udaipur. Percentage changes in non-monsoon average maximum temperature were found to be at a minimum (2.08%) and maximum (3.31%) in Dungarpur and Pali, respectively. No significant trends were found in the average maximum temperature at an annual temporal scale for the urban centers located in the northern region of the state (Fig. 5c). Similarly, significant trends in annual average maximum temperature were observed with maximum values of slopes (0.03 °C/year) in Barmer, Jaisalmer, Jalore, Pali and Udaipur. The percentage change was found to be at a minimum (1.47%) and at a maximum (3.24%) in Pali and Baran, respectively. The detail magnitude of the Sen's slope for average maximum temperature for the urban centers of Rajasthan State is shown in Fig. 5. These results are in good agreement with previous studies in India (e.g., Singh et al., 2008a, b). 4.2. Trends in extreme events 4.1.3. Average temperature Significant increasing trends in average temperature at seasonal and annual temporal scales were observed for the majority of the urban centers in Rajasthan State, accounting for an average of 80% of the urban centers (Fig. 4). The significant increasing trends in monsoon average temperature were observed in Bikaner, Bundi, Churu, Dausa, Jaipur, Jaisalmer, Kota and Sikar, which is indicated by the maximum value of slope (i.e., 0.03 °C/hydrologic year). Also, percentage change in monsoon average temperature was found to be at a maximum for Jaisalmer (3.47%) and at a minimum for Alwar (2.43%). Similarly, significant trends in non-monsoon and annual average temperature showed the maximum value of 0.03 °C in the urban centers. The percentage changes in significant trends in non-monsoon average temperature were observed to be at a minimum (2.43%) and at a maximum (4.84%) in Banaswara and Sikar, respectively. The percentage change in annual average temperature was found to be at a minimum (1.53%) and maximum (4.32%) in Dungarpur and Sikar, respectively. The detail magnitude of the Sen's slope of the significant trends in average temperature is shown in Fig. 4. This warming trend in average temperature is also supported by other studies in India (e.g., Singh et al., 2008a, b). 4.2.1. Extreme annual daily rainfall The spatial distribution of extreme annual maximum daily rainfall trends and variation is shown in Fig. 6a, while the MK test statistics (Zmk) with increasing, decreasing and no trends are shown in Fig. 6b. Both positive (southeastern) and negative (northeastern) trends were observed in extreme annual daily rainfall in the urban centers of Rajasthan State. The results revealed that significant decreasing trends and variability were observed in annual extreme daily rainfall for Hanumangarh (at −0.65 mm/year and −49.16%), Karauli (at −1.56 mm/year and −60.31%), Nagaur (at −1.56 mm/year and −80.33%), Sikar (at −1.33 mm/year and −63.86%) and Tonk (at −1.64 mm/year and −65.14%). However, a significant increasing trend was found only in Jhalawar (at 1.38 mm/year and 45.97%). The maximum values of decreasing (−1.64 mm/year) slopes and percentage change (−80.33%) in significant trends in annual maximum daily rainfall were observed in Tonk and Nagaur, respectively (Table S4). Significant trends were not found for extreme events of annual daily rainfall in other urban centers (Fig. 6). The detail magnitudes of slope and percentage change in extreme annual maximum daily rainfall are presented in Table S4 and Fig. 6. Similarly, only significant trends in extreme annual maximum daily rainfall are shown in Fig. 7. Investigating 82 S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 Fig. 3. Location of urban centers with increasing, decreasing and no trends for a) monsoon, b) non-monsoon and c) annual average minimum temperature. S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 83 Fig. 4. Location of urban centers with increasing, decreasing and no trends at 10% significance level for a) monsoon, b) non-monsoon and c) annual average temperature. 84 S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 Fig. 5. Location of urban centers with increasing, decreasing and no trends for a) monsoon, b) non-monsoon and c) annual average maximum temperature. S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 85 Fig. 6. Location of urban centers with increasing, decreasing and no trends at 10% significance level for extreme annual daily rainfall. trends in extreme events is necessary to assess the impact of global climate change over regional climatic variability on present and future water resources (Wang et al., 2013). The assessment of trends of extreme annual daily rainfall events is crucial for adaptation planning in arid and semi-arid regions of India such as Rajasthan. 4.2.2. Extreme annual daily minimum temperature The spatial distribution of extreme annual daily minimum temperature is shown in Fig. 8a, while the MK test statistic (Zmk) indicating trends is shown in Fig. 8b. Significant increasing trends were observed in extreme annual daily minimum temperature (at 0.03 to 0.07 °C/year) for the urban centers located in the northeastern and western regions of Rajasthan State (Fig. 8). However, significant trends were not found in other urban centers of Rajasthan State (Fig. S1) (Supporting material). A significant increasing trend in annual minimum daily temperature was observed with a maximum value of slope (0.073 °C/year) and percentage change (102.76%) in Sikar, while significant increasing annual daily minimum temperature was observed with a minimum value of slope (0.040 °C/year) and percentage change (31.01%) in Banaswara and Hanumangarh, respectively. The details of slope magnitude and variability of extreme annual daily minimum temperature in the urban centers of Rajasthan State are presented in Table S5 and Fig. 8. Only the trends in significant extreme events of annual daily minimum temperature are shown in Fig. S1. From the results, it is clear that the magnitude of the minimum temperature trends and variability is greater than the maximum temperature extremes. This was also observed in other parts of India (Duhan et al., 2013). S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 1998 2005 Rainfall Linear (Rainfall) 350 300 250 y = 1.7368x -3347.5 R² = 0.1206 200 150 100 50 0 1970 1977 Year y = -1.2439x + 2540.6 R² = 0.0493 200 150 100 50 0 1970 1977 1984 1991 1998 2005 Annual max. daily rainfall, mm Annual max. daily rainfall, mm Rainfall Linear (Rainfall) 300 250 1991 1998 2005 400 350 Rainfall Linear (Rainfall) 160 250 200 150 100 50 0 1970 1977 1984 1991 1998 2005 Sikar 120 100 80 60 40 y = -1.298x + 2653.4 R² = 0.1839 0 1970 1977 1984 1991 1998 2005 Year y = -2.3587x + 4779.6 R² = 0.1561 Year 140 20 Rainfall Linear (Rainfall) 300 Year Nagaur 350 1984 Karauli Year Tonk Annual max. daily rainfall, mm 100 90 80 70 60 50 40 30 20 y = -0.5473x + 1134.5 10 R² = 0.0907 0 1970 1977 1984 1991 Jhalawar Annual max. daily rainfall, mm Annual max. daily rainfall, mm Hanumangarh Rainfall Linear (Rainfall) Annual max. daily rainfall, mm 86 250 Rainfall Linear (Rainfall) 200 150 y = -2.2845x + 4629.9 R² = 0.3097 100 50 0 1970 1977 1984 1991 1998 2005 Year Fig. 7. Trends in statistically significant extreme annual daily rainfall in the urban centers of districts of Rajasthan State. 4.2.3. Extreme annual daily maximum temperature The spatial distribution of extreme annual daily maximum temperature (obtained using MK test statistic (Zmk)) was determined using the IDW interpolation technique for the 33 urban centers of Rajasthan State (Fig. 9). A significant increasing trend in extreme annual daily maximum temperature was observed in the south (i.e., Banaswara, Baran, Bundi, Chittorgarh, Jhalawar, Pali and Sirohi), east (i.e., Churu and Jaipur) and west (i.e., Jaisalmer) urban centers of Rajasthan State (Fig. 9). Significant increasing trend in the annual maximum daily temperature was observed with a maximum value of Sen's slope (0.049 °C/year) and percentage change (3.86%) in Jhalawar, while the minimum value of slope (0.029 °C/year) and percentage change (2.35%) was observed in Sirohi. The minimum and maximum annual daily maximum temperatures were found to be at 44.48 °C and 47.62 °C in Sirohi and Dholpur, respectively. The detail magnitude of the Sen's slope and percentage change of annual daily maximum temperature in urban centers of Rajasthan State is presented in Table S6. The results of significant extreme events of annual daily maximum temperature are shown in Fig. S2. The results clearly indicate the warming trends in the extreme annual maximum daily temperature in the urban centers, which shows attribution of anthropogenic influence on climate (Fig. 9a and b). These may include drastic changes in land use/land cover, industrialization and urbanization. In summary, the mean and extreme rainfall and temperature trends and variability at a spatial and temporal scale were evaluated for the 33 urban centers of Rajasthan State. The results of mean and extreme annual daily rainfall showed both positive and negative trends over the urban centers. An increasing trend was found in annual and seasonal mean and extreme minimum, and average and maximum temperatures over the urban centers. This may be due to several natural and anthropogenic activities such as land use/land cover, urbanization and industrialization, which contribute to the changes in rainfall and temperature in terms of their distribution, intensity and occurrence in space and time. The detailed evaluation of the magnitude and variability in rainfall, and minimum and maximum temperatures in terms of mean and extremes, is very important for impact assessment and adaptation planning for floods, drought and extreme events. It is suggested that in the future, the assessment of mean and extreme events considering climate change scenarios can be carried out on seasonal and annual temporal scales at each urban center of Rajasthan State, India. This would be very useful to facilitate the proper planning and management of water resources in a changing climate in Rajasthan State, India. 5. Conclusions In the present study, the trends in mean (monsoon season, non-monsoon season and annual) and extreme annual daily rainfall and temperature were determined for the 33 urban centers of Rajasthan State. This was carried out using the non-parametric Mann–Kendall (MK) test. The Sen's slope and percentage changes in rainfall and temperature were also estimated over the study period (1971–2005). Further, the spatial variations in mean (seasonal and annual) and extreme annual daily rainfall and temperature with their trends were determined using the inverse-distance-weighted (IDW) interpolation method. S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 87 Fig. 8. Location of urban centers with increasing, decreasing and no trends at 10% significance level for extreme annual daily minimum temperature. Both positive and negative trends were observed by the MK test and Sen's Slope estimator in seasonal and annual average rainfall in the urban centers of Rajasthan State. All of the significant trends in monsoon and annual rainfall were found to be decreasing at the 10% level of significance. However, significant trends in non-monsoon average rainfall were observed (both positive and negative trends). The significant decreasing trends of average monsoon rainfall had the highest value of decreasing slope (i.e., −8.56 mm/hydrologic year) for Tonk and percentage change of −88.16% in Jodhpur. A significant increasing trend was observed in non-monsoon rainfall at Hanumangarh (1.70 mm/hydrologic year), and a decreasing trend was found at Chittorgarh (−1.27 mm/hydrologic year). In the case of annual average rainfall, maximum values of decreasing slopes (−10 mm/year) and percentage change (−85.05%) were observed in Tonk and Jodhpur, respectively. Significant increasing trends were observed in minimum temperature at seasonal (0.04 and 0.05 °C/hydrologic year) and annual (0.05 °C/year) temporal scales for the majority of the urban centers in the northeastern part of Rajasthan State. Significant increasing trends (at 0.01 to 0.05 °C/year) were observed in average temperature for most of the urban centers (about 80% of the urban centers), in particular, all the urban centers showed significant warming trends in annual average temperature. Significant increasing trends in monsoon average maximum temperature were found in the some of the urban centers. However, the southwestern urban centers had significant increasing trends in non-monsoon average maximum temperature. Significant increasing trends were observed in annual average maximum temperature for urban centers that are located in the southeast and southwestern parts of Rajasthan State. The trends in maximum and minimum temperature may 88 S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 Fig. 9. Location of urban centers with increasing, decreasing and no trends at 10% significance level for extreme annual daily maximum temperature. be influenced in different ways due to local physical geographic (e.g., topography, urbanization, air pollution etc.) and atmospheric circulation features (Turkes and Sumer, 2004; Dhorde et al., 2009; Tabari and Talaee, 2011a). In the case of extreme events for annual daily rainfall both positive (southeastern) and negative (northeastern) trends were observed in some of the urban centers. Other urban centers did not show any significant trends in extreme annual daily rainfall. Significant increasing trends were observed in extreme annual daily minimum temperature (at 0.03 to 0.07 °C/year) for the urban centers located in the northeastern and western regions of Rajasthan State. However, significant trends were not found in the other urban centers of Rajasthan State. A significant increasing trend in extreme annual daily maximum temperature (at 0.03 to 0.05 °C/year) was observed in the south (i.e., Banaswara, Baran, Bundi, Chittorgarh, Jhalawar, Pali and Sirohi), east (i.e., Churu and Jaipur) and west (i.e., Jaisalmer) urban centers of Rajasthan State. The results of mean (monsoon, non-monsoon and annual) and extreme annual daily rainfall and temperature clearly indicate prominent changes in the urban centers, whether due to natural variability or anthropogenic activity over the period. The trends and variability of rainfall and temperature found in this study could be one part of a cycle with a frequency shorter than 35 years. A similar observation was also made for precipitation trends in Iran (Tabari and Talaee, 2011a). There is no doubt that changes in natural factors like sea surface temperature, the Earth's rotation and solar cycles can cause changes in the climate. However, we do not have any control of these factors. Anthropogenic activities like urbanization and land use change can also affect local climates significantly. The effects of these factors could be minimized by taking appropriate adaptation and mitigation measures. The assessment of mean S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 and extreme events of rainfall and temperature is necessary to help and prepare suitable adaptation strategies in the face of uncertain changing climate and extreme weather events. The outcome of this study should prove to be useful to policy makers, hydrologists, and water resource planners dealing with climate change in all the urban centers in Rajasthan State for the sustainable development and planning of water resources. Acknowledgments We acknowledge the Indian Meteorological Department (IMD) in Pune for providing useful meteorological gridded datasets for this research work. The financial assistance provided by the Ministry of Human Resource Development, Government of India in the form of a scholarship is duly acknowledged. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.atmosres.2013.10.024. References Arora, M., Goel, N.K., Singh, P., 2005. Evaluation of temperature trends over India. Hydrol. Sci. J. 50 (1), 81–93. Basistha, A., Goel, N.K., Arya, D.S., Gangwar, S.K., 2007. Spatial pattern of trends in Indian sub-divisional rainfall. Jalvigyan Sameeksha 22, 47–57. Basistha, A., Arya, D.S., Goel, N.K., 2009. Analysis of historical changes in rainfall in the Indian Himalayas. Int. J. Climatol. 29, 555–572. Cheung, W.H., Senay, G.B., Singh, A., 2008. Trends and spatial distribution of annual and seasonal rainfall in Ethiopia. Int. J. Climatol. 28, 1723–1734. Coulibaly, P., Shi, X., 2005. Identification of the Effect of Climate Change on Future Design Standards of Drainage Infrastructure in Ontario, Final Report. McMaster University, Department of Civil Engineering, Ontario. Dash, S.K., Saraswat, Vaishali, Panda, S.K., Sharma, N., 2013. A study of changes in rainfall and temperature patterns at four cities and corresponding meteorological subdivisions over coastal regions of India. Glob. Planet. Chang. http://dx.doi.org/10.1016/j.gloplacha.2013.06.004. De, U.S., Rao, G.S.P., 2004. Urban climate trends — the Indian scenario. J. Indian Geophys. Union 8 (3), 199–203. Deka, R.L., Mahanta, C., Pathak, H., Nath, K.K., Das, S., 2012. Trends and fluctuations of rainfall regime in the Brahmaputra and Barak basins of Assam, India. Theor. Appl. Climatol. http://dx.doi.org/10.1007/s00704-0120820-x. Dhorde, A., Dhorde, A., Gadgil, A.S., 2009. Long-term temperature trends at four largest cities of India during the twentieth century. J. Indian Geophys. Union 13 (2), 85–97. Duhan, D., Pandey, A., 2013. Statistical analysis of long term spatial and temporal trends of precipitation during 1901–2002 at Madhya Pradesh, India. Atmos. Res. 122, 136–149. Duhan, D., Pandey, A., Gahalaut, K.P.S., Pandey, R.P., 2013. Spatial and temporal variability in maximum, minimum and mean air temperatures at Madhya Pradesh in central India. C. R. Geosci. 345, 3–21. Erdoğan, S., 2009. A comparison of interpolation methods for producing digital elevation models at the field scale. Earth Surf. Proc. Land. 34, 366–376. Gadgil, A., Dhorde, A., 2005. Temperature trends in twentieth century at Pune. India Atmos. Environ. 39, 6550–6556. Ghosh, S., Luniya, V., Gupta, A., 2009. Trend analysis of Indian summer monsoon rainfall at different spatial scales. Atmos. Sci. Lett. 10, 285–290. Gosain, A.K., Rao, S., Basuray, D., 2006. Climate change impact assessment on hydrology of Indian river basins. Curr. Sci. 90, 346–353. Gowda, K.K., Majuantha, K., Manjunath, B.M., Putty, Y.R., 2008. Study of climate changes at Davangere region by using climatological data. Water Energy Int. 65 (3), 66–77. Hamlet, A.F., Mote, P.W., Clark, M.P., Lettenmaier, D.P., 2005. Effects of temperature and precipitation variability on snowpack trends in the Western United States. J. Clim. 18, 4545–4561. Helsel, D.R., Hirsch, R.M., 2002. Statistical methods in water resources techniques of water resources investigations. Book 4, Chapter A3. U.S. Geological Survey. (522 pages). 89 Huang, J., Sun, S., Zhang, J., 2013. Detection of trends in precipitation during 1960–2008 in Jiangxi Province, Southeast China. Theor. Appl. Climatol. http://dx.doi.org/10.1007/s00704-013-0831-2. IPCC, 2001. Third assessment report. The Scientific Basis, vol. 1. Cambridge Univ. Press. IPCC, 2007. AR4 working group I report. The physical science basis. Isaaks, E.H., Srivastava, R.M., 1989. An Introduction to Applied Geostatistics. Oxford University Press, New York. Jain, S.K., Kumar, V., Saharia, M., 2013. Analysis of rainfall and temperature trends in Northeast India. Int. J. Climatol. 33, 968–978. Jeganathan, A., Andimuthu, R., 2013. Temperature trends of Chennai City, India. Theor. Appl. Climatol. 111, 417–425. Jiang, F., Hu, R., Wang, S., Zhang, Y., Tong, L., 2013. Trends of precipitation extremes during 1960–2008 in Xinjiang, the Northwest China. Theor. Appl. Climatol. 111, 133–148. Kampata, J.M., Parida, B.P., Moalafhi, D.B., 2008. Trend analysis of rainfall in the headstreams of the Zambezi River Basin in Zambia. Phys. Chem. Earth 33, 621–625. Karpouzos, D.K., Kavalieratou, S., Babajimopoulos, C., 2010. Trend analysis of precipitation data in Pieria region (Greece). Eur. Water 30, 31–40. Kendall, M.G., 1975. Rank Correlation Methods, 4th ed. Charles Griffin, London. Kumar, V., Jain, S.K., Singh, Y., 2010. Analysis of long-term rainfall trends in India. Hydrol. Sci. J. 55 (4), 484–496. Ludwig, W., Serrat, P., Cesmat, L., Esteves, J.G., 2004. Evaluating the impact of the recent temperature increase on the hydrology of the Tˆet River (Southern France). J. Hydrol. 289, 204–221. Luo, Y., Lio, S., Sheng, L., Fu, S., Liu, J., Wang, G., Zhou, G., 2008. Trends of precipitation in Beijing River Basin, Guangdong Province, China. Hydrol. Process. 22, 2377–2386. Mall, R.K., Gupta, A., Singh, R., Singh, R.S., Rathore, L.S., 2006. Water resources and climate change: an Indian perspective. Curr. Sci. 90 (12), 1610–1626. Mann, H.B., 1945. Non-parametric test against trend. Econometrica 13, 245–259. McBean, E., Motiee, H., 2008. Assessment of impacts of climate change on water resources: a long term analysis of the Great Lakes of North America. Hydrol. Earth Syst. Sci. 12, 239–255. Modarres, R., da Silva, R.V.P., 2007. Rainfall trends in arid and semi-arid regions of Iran. J. Arid Environ. 70, 344–355. Motiee, H., McBean, E., 2009. An assessment of long-term trends in hydrologic components and implications of water levels in Lake Superior. Hydrol. Res. J. 40 (6), 564–579. Nalley, D., Adamowski, J., Khalil, B., 2012. Using discrete wavelet transforms to analyze trends in streamflow and precipitation in Quebec and Ontario (1954–2008). J. Hydrol. 475, 204–228. Nemec, J., Gruber, C., Chimani, B., Auer, I., 2012. Trends in extreme temperature indices in Austria based on a new homogenized dataset. Int. J. Climatol. 33, 1538–1550. Pal, I., Al-Tabbaa, A., 2010. Long-term changes and variability of monthly extreme temperatures in India. Theor. Appl. Climatol. 100, 45–56. Pal, I., Al-Tabbaa, A., 2011. Assessing seasonal precipitation trends in India using parametric and non-parametric statistical techniques. Theor. Appl. Climatol. 103, 1–11. Pant, G.B., Kumar, K.R., 1997. Climates of South Asia. John Wiley, UK 320. Patra, J.P., Mishra, A., Singh, R., Raghuwanshi, N.S., 2012. Detecting rainfall trends in twentieth century (1871–2006) over Orissa State, India. Clim. Chang. 111 (3), 801–817. Rai, R.K., Upadhyay, A., Ojha, C.S.P., 2010. Temporal variability of climatic parameters of Yamuna River Basin: spatial analysis of persistence, trend and periodicity. Open Hydrol. J. 4, 184–210. Rajeevan, M., Bhate, J., 2008. A high resolution daily gridded rainfall data set (1971–2005) for mesoscale meteorological studies. National Climate Centre, India Meteorological Department, Pune, Research Report No: 8/2008. Rathor, M.S., 2005. State level analysis of drought policies and impacts in Rajasthan, India. Colombo, Sri Lanka: IWMI. 40p. Working paper 93: Drought Series Paper 6. Robinson, T.P., Metternicht, G., 2006. Testing the performance of spatial interpolation techniques for mapping soil properties. Comput. Electron. Agric. 50, 97–108. Roy, S.S., Balling Jr., R.C., 2004. Trend in extreme daily precipitation indices in India. Int. J. Climatol. 24, 457–466. Saboohi, R., Soltani, S., Khodagholi, M., 2012. Trend analysis of temperature parameters in Iran. Theor. Appl. Climatol. 109, 529–547. Sahoo, D., Smith, P.K., 2009. Hydroclimatic trend detection in a rapidly urbanizing semi-arid and coastal river basin. J. Hydrol. 367, 217–227. Sen, P.K., 1968. Estimates of the regression coefficient based on Kendall's tau. J. Am. Stat. Assoc. 39, 1379–1389. Singh, P., Kumar, V., Thomas, T., Arora, M., 2008a. Basin-wide assessment of temperature trends in Northwest and Central India. Hydrol. Sci. J. 53 (2), 421–433. 90 S.M. Pingale et al. / Atmospheric Research 138 (2014) 73–90 Singh, P., Kumar, V., Thomas, T., Arora, M., 2008b. Changes in rainfall and relative humidity in river basins in Northwest and Central India. Hydrol. Process. 22, 2982–2992. Some'e, B.S., Ezani, A., Tabari, H., 2012. Spatiotemporal trends and change point of precipitation in Iran. Atmos. Res. 113, 1–12. Srivastava, A.K., Rajeevan, M., Kshirsagar, S.R., 2008. Development of a high resolution daily gridded temperature data set (1969–2005) for the Indian region. National Climate Centre, India Meteorological Department, Pune, Research Report No: 8/2008. Tabari, H., Talaee, P.H., 2011a. Analysis of trends in temperature data in arid and semi-arid regions of Iran. Glob. Planet. Chang. 79, 1–10. Tabari, H., Talaee, P.H., 2011b. Temporal variability of precipitation over Iran: 1966–2005. J. Hydrol. 396, 313–320. Tabari, H., Somee, B.S., Zadeh, R.M., 2011. Testing for long-term trends in climatic variables in Iran. Atmos. Res. 100, 132–140. Theil, H., 1950. A rank-invariant method of linear and polynomial regression analysis. I, II, III. Nederl. Akad. Wetensch. Proc. 53, 386–392. Tozer, C.R., Kiem, A.S., Verdon-Kidd, D.C., 2012. On the uncertainties associated with using gridded rainfall data as a proxy for observed. Hydrol. Earth Syst. Sci. 16, 1481–1499. Turkes, M., Sumer, U.M., 2004. Spatial and temporal patterns of trends and variability in diurnal temperature ranges of Turkey. Theor. Appl. Climatol. 77, 195–227. Wang, H., Chen, Y., Xun, S., Lai, D., Fan, Y., Li, Z., 2013. Changes in daily climate extremes in the arid area of Northwestern China. Theor. Appl. Climatol. 112, 15–28. Yavuz, H., Erdogan, S., 2012. Spatial analysis of monthly and annual precipitation trends in Turkey. Water Resour. Manag. 26 (3), 609–621. Yevjevich, V., 1971. Stochastic Processes in Hydrology. Water Resources Publications, Fort Collins: CO. Yue, S., Hashino, M., 2003. Long term trends of annual and monthly precipitation in Japan. J. Am. Water Resour. Assoc. 39 (3), 587–596. Yue, S., Pilon, P., 2004. A comparison of the power of the t test, Mann– Kendall and bootstrap tests for trend detection. Hydrol. Sci. J. 49 (1), 21–37. Zhang, X., Vincent, L.A., Hogg, W.D., Niitsoo, A., 2000. Temperature and precipitation trends in Canada during the 20th century. Atmos. Ocean 38 (3), 395–429. Zhang, X., Hogg, W.D., Mekis, F., 2001. Spatial and temporal characteristics of heavy precipitation events over Canada. J. Clim. 14, 1923–1936. Zhang, X.B., Zwiers, F.W., Li, G.L., 2004. Monte Carlo experiments on the direction of trends in extreme values. J. Clim. 17, 1945–1952. Zin, W.Z.W., Jamaludin, S., Deni, S.M., Jemain, A.A., 2010. Recent changes in extreme rainfall events in Peninsular Malaysia: 1971–2005. Theor. Appl. Climatol. 99, 303–314.