Environ Monit Assess DOI 10.1007/s10661-014-3871-x

Assessment of regional climatic changes in the Eastern Himalayan region: a study using multi-satellite remote sensing data sets Anubha Agrawal & Anu Rani Sharma & Shresth Tayal

Received: 11 October 2013 / Accepted: 6 June 2014 # Springer International Publishing Switzerland 2014

Abstract In this study, an attempt has been made to capture the sensitivity of a mountainous region to elevation-dependent warming and the response of a glacier-laden surface to increasing greenhouse gases (GHGs) and aerosol concentration. Some of the changes Sikkim has undergone due to urban sprawl are as follows: an increase of ~0.7±0.46 °C temperature in the past 40 years at an altitude of 5.5 km; a 2.21 km2/year rate of loss of glacierised area in the past 33 years; an increase in absorbed longwave radiation (6±2.41 W/ m2); an increase in heat fluxes (2±0.97 W/m2); a decrease in albedo during the last 30 years; an increase in the concentrations of carbon dioxide (4.42 %), methane (0.61 %), ozone (0.67 %) and black carbon column optical depth (7.19 %); a decrease in carbon monoxide (2.61 %) and an increase in aerosol optical depth (19.16 %) during the last decade; a decrease in precipitation, water yield, discharge and groundwater; and an increase in evapotranspiration during 1971–2005. Detection of three climate signals (1976, 1997 and 2005) in the entire analysis is the quantification of the fact that the climate of Sikkim is moving away from its inter-annual variability. An increase in temperature (0.23 °C/decade) at higher altitude (~5.5 km), suppression of precipitation, decreasing water availability and rapid loss of glacierised area are the evidences of the fact that air pollution is playing a significant role in bringing about regional climatic changes in Sikkim. In this study, change A. Agrawal (*) : A. R. Sharma : S. Tayal Department of Natural Resources, TERI University, New Delhi 110070, India e-mail: [email protected]

detection method has been used for the first time for the estimation of change in a glacierised area of the region. Keywords Eastern Himalaya . Glaciers . Climate change . Air pollution . Hydrological cycle

Introduction Regional air pollution, glacial retreat and climate change are three large-scale environmental issues which have captured the attention of scientists and policymakers all around the world. Increasing air pollution is impacting the atmosphere, delicate ecosystems and human health negatively (Zlatev and Moseholm 2008; Noyes et al. 2009; Yang et al. 2012). Increasing water demands (502 billion cubic meters (BCM) in 1990 and 813 BCM in 2010), energy emissions from energy consumption (700.4 in 1994 and 1798 million tonnes of carbon dioxide equivalent in 2011) in India (TEDDY 2013), etc. are some of the major stresses which are acting as catalyst to the problem of climate change. The entire world has become vulnerable towards ‘climate change’. This has led to the initiation of an indepth investigation of the problem all across the globe. For example Vuille and Bradley (2000) have analysed the mean annual temperature trends over the past six decades (1939–1998) in the tropical Andes. Their results indicate that temperature over there has been increasing by 0.1–0.11 °C/decade since 1939, and the rate of warming has increased to 0.32–0.34 °C/decade in the last 25 years. According to Liu and Chen (2000), the

Environ Monit Assess

linear rates of temperature have increased over the Tibetan Plateau during the period 1955–1996 by 0.16 °C/decade for the annual mean. Qin et al. (2009) have found that the warming rate is increasing over Tibetan Plateau with respect to elevation. Shrestha et al. (1999) analysed maximum temperature data from 49 stations in Nepal for the period 1971–1994. The analyses reveal warming trends after 1977 ranging from 0.06 to 0.12 °C/year in most of the middle Mountain and Himalayan regions, while the Siwalik and Terai (southern plains) regions show warming trends less than 0.03 °C/year. Rana et al. (2012) have used MannKendall (MK) test and regression analysis to study rainfall trends in Delhi and Mumbai during the period from 1951 to 2004. For Mumbai, significant negative changes for long-term rainfall were detected for different seasons and for the whole year in the 1951–2004 periods. Kiely et al. (1998) have used Pettitt-ManWhitney (PMW) test to analyse 54 years (1940–1993) of precipitation data of Valentia in Ireland and found that there has been a 10 % increase in the mean annual precipitation from pre-1975 to post-1975. Kaufman and Seto (2001) have used Landsat imagery to detect land use land cover changes in the Pearl River Delta of China for the period 1989–1996. Zhang et al. (2002) have analysed Landsat Thematic Mapper (TM) images of years 1984 and 1997 for urban built-up and land change detection and found that the built-up area increased by 183 km2 from 1984 to 1997. Lu et al. (2002) have analysed multi-temporal TM images of years 1985, 1991 and 1996 for vegetation change detection (area and biomass) of the Altamira region of Brazil. Ramachandran et al. 2012 have derived seasonal and annual mean trends in aerosol optical depths (AODs) for 2000–2009 using Moderate Resolution Imaging Spectroradiometer (MODIS) Level 2, 10 km×10 km remote sensing data over different locations in India where AOD is showing an increasing trend over most of the states. Schmidt and Nusser (2012) have reported an area loss of 14 % from 1969 to 2010 of glaciers of Trans-Himalayan Kang Yatze Massif, Ladakh. Shahgedanova et al. (2010) have reported an area loss of 19.7±5.8 % from 1952 to 2004 of 126 glaciers of North and South Chuya ridges, Atlai Mountains. Racoviteanu et al. (2008) have reported an area loss of 22.4 % from 1970 to 2003 of Cordillera Blanca, Peru; for the 18 main Peruvian ranges of the Cordillera of the Andes with glaciers, it is estimated that between the 1960s and the end of the 1990s, there has been a loss

of more than 20 % in surface area and volume (Leavell and Portocarrero 2003). The glacierised area of upper Bhagirathi and Alaknanda basins of Garhwal Himalaya have shown an area loss of 4.6±2.8 % from 1968 to 2006 (Bhambri et al. 2011), glaciers of Khmbu Himal, Nepal have shown an area loss of 5 % from 1962 to 2005 (Bolch et al. 2008), Sikkim glaciers have been found to show maximum retreat (400 m) when mean retreat of snout in six regions from entire Himalayas during 2000/01/02–2010/11 for 248 glaciers has been studied (Bahuguna et al. 2014), etc. Above are some of the studies carried out to detect and assess the climatological changes in various regions all over the world. During the past few years, Himalayan glaciers have been under scrutiny due to the observed changes in them and their vital role in the sustenance of mankind. Himalayan glaciers are divided into three zones, western, central and eastern. Though there are a large number of studies available over Western Himalaya, Eastern Himalaya of India remains sparsely studied because of its difficult terrain. So, satellite data becomes important for the study of regional changes in the Eastern Himalayan glaciers. In this study, an attempt has been made to quantify regional changes in Sikkim (Fig. 1) from 1966 to 2013 mainly with the help of satellite data.

Data sets Ground-based measurements Temperature data Mean air temperature for time period 1966–2000 of Gangtok station has been downloaded from http:// www.imd.gov.in/section/nhac/mean/Gangtok.htm. Water balance data Water balance data under virgin scenario for time period 1971–2005 has been downloaded from Hydrological Information System (HIS) (http://gisserver.civil.iitd.ac. in/natcom/). Precipitation, water yield, evapotranspiration, soil moisture and groundwater data have been procured from HIS. In this data, HIS has obtained precipitation data from the Indian Meteorological Department (IMD) and the rest of the parameters from Soil Water Assessment Tool (SWAT).

Environ Monit Assess

Fig. 1 FCC of Sikkim (Landsat image: 27 August 2011)

Satellite-based measurements Landsat data Landsat has been providing multi-temporal, multispectral and multi-resolution range images of the Earth’s surface since 1972. Till now, eight Landsat satellites have been launched. Landsat 1–4, with MSS

multi-spectral sensor, have been acquiring data in four bands (bands 1–4 with resolution 60 m). Landsat 4–5, with TM multi-spectral and TM thermal sensors, have been acquiring data in seven bands (bands 1–5, 7 with resolution 30 m and thermal band 6 with resolution 120 m). Landsat 7, with Enhanced Thematic Mapper (ETM) + multi-spectral and ETM + thermal sensors, has been acquiring data in seven bands (bands 1–5, 7

Environ Monit Assess

with resolution 30 m and thermal bands 6.1 and 6.2 with resolution 60 m). Landsat 8, launched on 11 February 2013, has two sensors. Sensor Operational Land Imager (OLI) collects data from nine bands, and thermal infrared sensor collects data from two bands. In this study, cloud-free images of 13 December 2001 and 27 August 2011 have been taken for the estimation of change in built-up and barren area (class A) and woodland and grassland area (class B). Sikkim is least covered with cloud and snow after the ablation season. Hence, for the estimation of change in glacierised area over the last four decades, images of 30 November 1976 and 21 November have been acquired from USGS Earth Explorer (http://earthexplorer.usgs.gov/). Atmospheric Infrared Sounder (AIRS) data Aqua, an Earth Observing Satellite (EOS) mission to examine water and other climate variables, is a major EOS mission. It carries six distinct earth-observing instruments: the AIRS, the Advanced Microwave Sounding Unit (AMSU), the Humidity Sounder for Brazil (HSB), the Advanced Microwave Scanning Radiometer for EOS (AMSR-E), the MODIS and Clouds and the Earth’s Radiant Energy System (CERES). AIRS is a 2382-channel high-spectral-resolution sounder. Infrared radiation with wavelength range 3.7– 15.4 μm are measured by 2,378 channels, and the remaining four channels measure visible and near-infrared radiation in the range 0.4–0.94 μm (Parkinson 2003). In this study, AIRS + AMSU version 6 data has been used. Carbon dioxide Carbon dioxide is one of the most important minor gases retrieved from AIRS spectral radiances in the 712–750-cm−1 region (mid-troposphere at a nadir resolution of 90 km×90 km). The AIRS broad swath enables it to map the global distribution of CO2 everyday. The AIRS CO2 level 3 is monthly gridded data with 2.5°×2° grid cell size (Rajab et al. 2009). The data is in mole fraction units (data×106 =ppm in volume). The data used for this study includes CO2 data from AIRS over the study region. Methane Methane (CH4) is the most abundant organic trace gas in the atmosphere (Wuebbles and Hayhoe 2002). According to Goddard Earth Sciences Data and Information Service Center (GES DISC), absorption of radiation at specific wavelengths by methane molecules

is the basis of measurement of methane concentrations in the atmosphere by the satellite instruments. Methane concentrations are measured in the infrared/microwave (IR/MW) spectral range and are sensitive mainly in the mid-troposphere. In the present study, data of methane volume mixing ration ascending has been included from AIRS over the study area with spatial resolution 1°×1°. Carbon monoxide Increasing or decreasing concentrations of carbon monoxide are mainly an indicator of combustion, e.g. use of fossil fuels, biomass burning, forest fires, etc. Since it has a short atmospheric lifetime, it does not have a large influence on climate and is not a significant indicator of climate change. Because more fires occur in drought conditions, increasing carbon monoxide concentrations could indicate meteorologically warmer and/or drier conditions in a specific region. According to the GES DISC, absorption of radiation at specific wavelengths by carbon monoxide molecules is the basis of measurement of carbon monoxide concentrations in the atmosphere by the satellite instruments. Carbon monoxide concentrations are measured in the IR/MW spectral range. With peak sensitivity at an altitude of approximately 5 km, AIRS is sensitive to carbon monoxide in the mid-troposphere at heights between 2 and 10 km. In the present study, total CO ascending data has been extracted from AIRS for the study area with spatial resolution 1°×1°. Ozone Tropospheric ozone affects both climate and air quality. Following water vapor, carbon dioxide and methane, the radiative forcing of tropospheric ozone, 0.35 W/m2, makes it the fourth most important atmospheric greenhouse gas. As for the air quality, tropospheric ozone plays a positive role by acting as a precursor of the idroxyl radical, which is able to remove several pollutants from the middle troposphere through oxidation reactions. It also plays a negative role as it is toxic for both humans and crops when it reaches high concentrations near the Earth’s surface (Noia et al. 2013). In the present study, total column ozone ascending data has been extracted from AIRS for the study area with spatial resolution 1°×1°.

Environ Monit Assess

Modern Era Retrospective-Analysis for Research and Applications (MERRA 2D) data sets MERRA is the reprocessing of atmospheric observations (Badarinath et al. 2010) collected over the satellite era (1979–present). According to GMAO (http://gmao.gsfc. nasa.gov/pubs/brochures/MERRA%20Brochure.pdf), this has been achieved by the use of GOES-5 atmospheric data assimilation system (ADAS) by the Global Modeling and Assimilation Office. MERRA is a climate quality analysis that places NASA’s EOS observations into a climate context. It is being conducted with version 5.2.0 of the GEOS-5-ADAS with a 1/2° latitude×2/3° longitude×72 layer model configuration. The products are distributed through GES-DISC (http://disc.sci.gsfc. nasa.gov/MDISC/dataprods/merra_products.shtml). In the present study, surface albedo, temperature at pressure levels 250, 500 and 850 hPa and absorbed longwave at the surface; latent heat flux over land and sensible heat flux over land have been used to carry out the assessment and quantify changes in the physical parameters of the regional climate. MERRA version 5 data has been used in the study. AOD The MODIS is a remote sensor on board the two EOS Terra and Aqua satellites. MODIS operates from an altitude of 705 km. It measures reflected solar radiance and terrestrial emission in 36 channels in the wavelength range of 0.41–14.4 mm with resolutions varying between 0.25 and 1 km (Ramachandran et al. 2012). MODIS collection version 5.1 monthly 550 nm AOD from Terra is utilised in the study for time period 2001–2012. Black carbon column optical depth (550 nm) The Goddard Chemistry Aerosol Radiation and Transport (GOCART) model simulates major tropospheric aerosol components (Chin et al. 2002). In the present study, black carbon (BC) column optical depth at wavelength 550 nm and spatial resolution 2°×2.5° has been extracted for the study area.

Aqua, CloudSat, Aura, OCO-2, GCM-W1 and PARASOL satellites as well. CALIPSO has a 98° inclination orbit and flies at an altitude of 705 km, providing daily global maps of the vertical distribution of aerosols and clouds. The CALIPSO payload consists of three instruments: the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), an imaging infrared radiometer and a moderate spatial resolution wide field-of-view camera. CALIOP provides profiles of backscatter at 532 and 1,064 nm, as well as two orthogonal (parallel and perpendicular) polarization components at 532 nm (Sharma et al. 2009; Yerramestti et al. 2013; Chen et al. 2010). A detailed discussion of CALIOP data products has been described by Powell et al. (2009). In the present study, the version 3.30 image has been acquired.

Methodology Landsat data Estimation of change in glacierised area Change in glacierised area of East Rathong watershed East Rathong watershed consists of 36 glaciers (Agrawal and Tayal 2013). The glacierised area of the 36 glaciers of East Rathong basin for the years 1976 and 2013 was estimated individually using unsupervised and supervised classification. Difference in the sum of the glacierised areas computed for all the glaciers together for 1976 and 2013 gave the change in the glacierised area (area I) of the basin from 1976 and 2013. The change in glacierised area of the East Rathong basin (area II) was again estimated using change detection method (image differencing approach). Hence, error involved in the change in the glacierised area estimated using image differencing approach is as follows: ErrorðE Þ ¼

area II area I

The actual change in the glacierised area (dC) becomes:

Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO)

dC ¼

The CALIPSO was launched on 28 April 2006. It is a part of the A-train constellation which includes the

where dCi = change estimated using image differencing approach.

dC i E

Environ Monit Assess

Change in glacierised area of Sikkim Change in the glacierised area estimated using change detection method was divided with E to estimate the actual change in the glacierised area.

data uniquely for the state. The data was also statistically analysed to detect the presence of trends and shifts at the sub-catchment level. MK, Theil and Sen’s median slope estimator, PMW and Buishand’s U tests were used to detect trends and shifts in the data.

Estimation of built-up and vegetated area Following Zha et al. (2003), two classes were made, barren land and built-up area as one class and vegetated area (woodland and farmland) as one class. The area was estimated for both the classes for years 2001 and 2011. Climatological parameters Satellite data Time series monthly average data was downloaded in netcdf file format. Average for every year was computed and plotted. For the detection of presence of trends in carbon monoxide, ozone, carbon dioxide, methane and black carbon data, regression analysis was used. And for detection of trends in albedo, temperature, sensible heat flux, latent heat flux and absorbed longwave at the surface, the MK test was used. The details of MK test are given elsewhere (Mann 1945; Kendall 1975; Chen and Grasby 2008). For the assessment of practical significance of the trend, slope of the trend and percentage change over mean for the period concerned were estimated using Theil and Sen’s median slope estimator test (Narayanan et al. 2013). For shift detection and to find the years in which the shifts were induced, the data sets were further subjected to Buishand’s U test and PMW test, respectively. The details of Buishand’s U test are given in Buishand (1984), and PMW test is given in Kiely et al. (1998). Water balance data Sikkim is a small state with area of 7,454.95 km2 and population of 54,0851 (HIS). For water balance analysis of the state precipitation, evapotranspiration (ET), groundwater, water yield and soil water, under virgin scenario, data for time period 1971–2005 were downloaded for sub-catchments Goma 3023502, Bharamaputra 3023501 and Daina 3024101. The downloaded sub-catchment level data and the area of the sub-catchments lying within the premises of Sikkim (calculated in Tables 1 and 2) were used to calculate the

Results Changes in the landscape of Sikkim Landsat images of years 1976 and 2013 of the months of October and November were analysed to estimate the change in the glacierised area of Sikkim from 1976 to 2013. The average rate of loss of the glacierised area from 1976 to 2013 has been estimated to be ~2.21 km2/year. Landsat images of years 2001 and 2011 were analysed to assess changes in classes A and B. It has been found that from 2001 to 2011 class A (built-up and barren area) has decreased by 36 % and class B (woodland and grassland area) has increased by ~6.5 %. Changes in class A and in class B are almost the same in terms of area. Changes in the climatological variables Albedo Surface albedo of Sikkim for the time period 1979–2011 was analysed which showed a statistically insignificant negative trend. The presence of statistically significant shift was detected in the year 1997. Change in the mean of the albedo before and after the shift is −0.002. In Sikkim, snowfall and glacier melt take place simultaneously. Most of the time of the year, the area is covered with wet snow. Since wet snow does not have very high albedo, the area does not experience significant changes in the albedo during the entire hydrological year. Hence, any significant trend in the albedo could not be detected. Temperature Variations in annual average air temperature data of Sikkim region at 250 (T250), 500 (T500) and 850 hPa (T850) for time period 1979–2011 were analysed. As per the analysis, no trends were detected in the T250 and T850 data sets, but air temperature of Sikkim at 500 hPa (Fig. 2) has shown statistically significant increase of

Environ Monit Assess Table 1 Details of the data analysed in the study Landsat data Date

Sensor

Mission

Resolution (m)

Path/row

30 November 1976

MSS

Landsat 2

60

149/41

13 December 2001

ETM+

Landsat 7

30

139/41

27 August 2011

TM

Landsat 5

30

139/41

21 November 2013

OLI

Landsat 8

30

139/41

IMD data Time period

Type of data

Data archived from station

Type of data

1971–2005

Water balance

Sub-catchment level

1966–2000

Temperature

IMD grid data (downloaded from Hydrological Information System in Virgin condition) Gangtok

VFM nighttime image

Northeast India

Monthly average

CALIPSO data 1 August 2013 From Giovanni portal Time period

Type of data

Longname of the data downloaded

1979–2011

Albedo

Surface albedo

Resolution 2=3  1=2 2=3  1=2

1979–2011

Temperature

1979–2011

Shortwave flux

Temperature at 250 hPa, temperature at 850 hPa, temperature at 500 hPa Net downward shortwave flux over land

1979–2011

Longwave flux

Net downward longwave flux over land

1979–2011

Latent heat

Latent heat flux from land

1979–2011

Sensible heat flux

Sensible heat flux (positive upward)

2001–2011

AOD

2003–2011 2003–2011

Methane Vmr CO

Aerosol optical thickness at 0.55 μm for both ocean (best) and land (corrected): mean of daily mean CH4_VMR_eff_A (volume mixing ratio ascending)

2003–2011

Ozone

Total column ozone ascending

2003–2011

CO2 mole fraction

Mole fraction of carbon dioxide in free troposphere

Total column carbon monoxide ascending

0.7±0.46 °C at 0.05 significance level. The slope of the trend is 0.02. Percentage change over mean for 1979– 2011 is 0.24 %. By applying Buishand’s U test, a statistically significant shift at 0.05 significance level was detected in all the three data sets. By applying the PMW test, it was found that the shifts were induced in

2=3  1=2 2=3  1=2 2=3  1=2 2=3  1=2 0:5  0:5 1  1 1  1 1  1 2:5  2

the years 1997 and 2005 in T500 and only in 1997 in the cases of T850 and T250. Change in the mean of the T500 before and after the shift in 1997 is 0.413; change in the mean of the T250 before and after the shift is 1.03; and change in the mean of the T850 before and after the shift is 0.14.

Table 2 Details of the basins, catchments and sub-catchments lying in Sikkim Area (km2)

%Area within Sikkim

km2

5,601.447

Basin

Catchment

Sub-catchment

Bharamaputra 302

Bharamaputra 30235

Goma 3023502

5,850.4

95.75

Bharamaputra 3023501

2,227

78.38

1,745.487

Daina 3024101

3,676.5

2.35

863.98

Bharamaputra 30241 Total area

8,210.914

Environ Monit Assess 266.0

265.5

265.0

264.5

264.0

Fig. 2 Graph showing increasing temperature at 500 hPa from 1979 to 2011

The IMD data of Gangtok station was also statistically analysed. Statistically insignificant negative trend was detected in the data. The presence of shift in the year 1976 was detected. Change in the mean of the temperature before and after the shift is −0.83. T500 data analysis is quiet close to the global temperature analysis published by IPCC AR4 WG I. According to IPCC AR4, the total temperature increase from 1850–1899 to 2001–2005 is 0.76±0.19 °C, and 1998 and 2005 have been reported as the warmest 2 years in the instrumental global surface air temperature record since 1850. According to Mitchell (1971), depending on the nature of aerosols and the underlying surface, the net effect of aerosols near Earth’s surface may be one of either cooling or warming. In this study, the temperature near Earth’s surface is decreasing.

Greenhouse gases (GHGs) and aerosols Annual average values of AOD, carbon monoxide, ozone, carbon dioxide, methane and BC column optical depth over Sikkim region were analysed. It has been observed that AOD has increased from 0.241167 in 2001 to 0.281125 in 2012 (Fig. 3). It is showing a statistically significant positive trend at 0.05 significance level. The slope of the trend is 0.005. Percentage change over mean for 2001–2012 is 14.42 %.

Carbon dioxide mole fraction has increased from 0.000374 in 2003 to 0.000394 in 2011 (Fig. 4). It is showing a statistically significant positive trend at 0.05 significance level. The slope of the trend is 2.09. Percentage change over mean and annual growth rate for 2003–2011 are 4.92 % and 1.89 ppm/year, respectively. According to IPCC 2007, the annual growth rate of CO2 at global scale for 1995–2005 is 1.9 ppm/year. Methane volume mixing ratio has increased from 1.789243×10−6 in 2003 to 1.809254×10− 6 in 2011 (Fig. 5). It is showing a statistically significant positive trend at 0.05 significance level with slope 0.001. Percentage change over mean for 2003–2011 is 0.7 %. Total column ozone has increased from 263.8 to 267 DU from 2003 to 2011 (Fig. 6). Total column carbon monoxide has decreased from 163 to 157.5 from 2003 to 2011 (Fig. 7). (A decrease of CO is an indicator of an increase in tropospheric ozone as the main precursors of tropospheric ozone formation are methane, NOx and CO. (The fact that the concentration of stratospheric ozone tends to remain constant over time, and the observations like a decrease in total column density of CO and an increase in total column density of ozone indicate that the concentration of tropospheric ozone is increasing in the atmosphere.) Ozone is showing a statistically significant positive trend at 0.05 significance level with slope 0.23. Percentage change over mean for 2003–2011 is 0.8 %.

Environ Monit Assess 0.38

0.36

0.34

0.32

0.30

Fig. 3 Graph showing increasing aerosol optical depth (AOD) from 2001 to 2012

Carbon monoxide is showing a statistically insignificant negative trend with slope 0.006. Percentage change over mean for 2003–2011 is −3.11 %. Black carbon column optical depth has increased from 0.0138 to 0.0151 from 2000 to 2007 (Fig. 8). It is showing a statistically significant positive trend at 0.05 significance level with slope 1.8 × 10 − 5 . Percentage change over mean for 2003–2011 is 0.5 %.

Heat fluxes Latent heat flux and sensible heat flux data for the time period 1979 to 2011 have been analysed for the state of Sikkim. It has been found that average sensible heat flux is 54.669 W/m2 and average latent heat flux is 66.701 W/m2. Sensible heat flux data set shows slight negative trend, and latent

0.0003925 0.0003900 0.0003875 0.0003850 0.0003825 0.0003800 0.0003775 0.0003750

Fig. 4 Graph showing increasing concentration of mole fraction of carbon dioxide from 2003 to 2011

Environ Monit Assess 1.810

1.805

1.800

1.795

Fig. 5 Graph showing increasing volume mixing ratio (ascending) of methane from 2003 to 2011

heat flux data set shows slight positive trends. The trends are insignificant, but the sum of sensible and latent heat fluxes shows a statistically significant positive trend at significance level 0.05 with slope 0.05. It has been found that the sum during the study period, i.e. from 1979–2011, has increased from 120 to 122 W/m2 with percentage change over mean being 1.41 %.

Absorbed longwave at the surface On the analysis of absorbed longwave at surface data for the time period 1979−2012 (Fig. 9), statistically significant positive trend at 0.05 significance level has been detected. The slope of the trend is 0.21. An increase of 6 W/m2 with percentage change over mean being 2.33 % has been

271 270 269 268 267 266 265 264 263 262

Fig. 6 Graph showing increasing concentration total column ozone (ascending) from 2003 to 2011

Environ Monit Assess 1.64 1.63 1.62 1.61 1.60 1.59 1.58 1.57 1.56

Fig. 7 Graph showing decreasing concentration of total column carbon monoxide (ascending) from 2003 to 2011

observed from 1979−2012. No shift has been detected in the data. Changes in the hydrological cycle On the analysis of decadal average water balance data of Sikkim, it has been found that from 1971 to 2005, water yield, precipitation and groundwater have been decreasing, and evapotranspiration and soil water have been increasing (Fig. 10).

Statistical analysis of the water balance data of individual sub-catchment was carried out. Precipitation and water yield are showing negative and ET is showing positive statistically significant trends at 0.05 significance level for the sub-catchment Bharamaputra 3023501. Precipitation, water yield and groundwater are showing negative and ET is showing positive statistically significant trends at 0.05 significance level for the sub-catchment Goma 3023502. Glacierised area of Sikkim is mostly present in the Goma basin. No trends

0.0155

0.0150

0.0145

0.0140

0.0135

Fig. 8 Graph showing increasing optical depth of black carbon column from 2000 to 2007

Environ Monit Assess 316 314 312 310 308 306 304 302

Fig. 9 Graph showing increasing absorbed longwave at surface from 1979 to 2011

were detected in the water balance data of subcatchment Daina 3024101. Statistical analyses of the hydrological parameters are further summarised in Table 4.

Discussion The mountain stations in the tropics have higher warming rates relative to their lower counterparts (Rangwala and Miller 2012). This may be due to the presence of aerosol layer, which extends from the

surface to high elevation (Fig. 11). It heats the midtroposphere by absorbing solar radiation. The heating produces an atmospheric dynamical feedback, also known as elevated-heat-pump (EHP) effect. EHP increases moisture, cloudiness and deep convection over northern India and also enhances the rate of snow melt in the Himalayas and Tibetan Plateau (Lau et al. 2010). Melting of Himalayan glaciers has been related to enhance heating from BC aerosols and GHGs of 0.25 K/ decade from 1950 to present (Menon et al. 2010). In this study, detection of the presence of increasing concentration of aerosols, heat fluxes and a temperature change of

30

25

20

Water Yield PrecipitaƟon

15

Soil Water 10

5

0

Fig. 10 Details of water balance (IMD) data of Sikkim from 1971 to 2005

Ground Water

Environ Monit Assess

Fig. 11 Vertical feature mask image for nighttime on 1 August 2013 over northeast India obtained from CALIPSO observations

+0.7 or +0.23 °C/decade at an altitude of 5.5 km indicates that increasing GHGs and BC aerosols are playing a significant role in the loss of glacierised area of Sikkim Himalayan glaciers. The glacierised area of Sikkim has been observed to be retreating at a rate of 2.77 km2/year from 1976 to 2009. The population of the Sikkim has increased two times from 1976 to 2001 (MoEF 2011), and area under production of crops (fruits and vegetables) has increased by 4.5 times from 2000 to 2006 (Basic Statistics of NER 2006) (Table 3). Increasing population, their needs and urbanization, has led to an increase in the total built-up area and area under agriculture, whereas fallow land has decreased. Also because of the decrease in glacierised area, the thermal state of the barren lands at higher elevations where vegetation could not grow earlier has possibly undergone some change. Due to this as well, the barren area has decreased and vegetated area has increased.

Over time, Sikkim’s radiation budget has also changed. Absorbed longwave at the surface is showing a statistically significant increase of ~6 W/m2 from 1979 to 2011. Increasing aerosol concentration in the atmosphere could be the main reason of the increase. Consequences of increasing pollution and temperature on the hydrological cycle are suppression of rain and increase of ET (due to the increase in saturation vapour pressure), respectively (Ramanathan et al. 2001). Similar results were obtained in this study as well. Most of the hydrological parameters analysed using data downloaded from HIS website are decreasing except for ET and soil water content (Table 4). An increase in ET can be attributed to the increase in air temperature of the region, and an increase in the area under crop and forest cover of Sikkim is the most probable explanation of the increase in soil water content of Sikkim. This is further acting as a positive feedback to the decreasing groundwater of the

Table 3 Statistics of crop area of northeastern region from 1996 to 2006 and (Basic Statistics of NER 2006) population of Sikkim from 1975 to 2001 (MoEF 2011) Crop area of NER (ha)

Population of Sikkim

Total pulses

Total food grains

1996–1997

157.4

3,761.9

1997–1998

157.4

3,756.3

3,913.7

1975–1976

262,000

2000–2001

152

3,893

670,200

674,245

1980–1981

316,000

2002–2003

177

3,898

597,600

601,675

1990–1991

406,000

300,6800

300,6800

2000–2001

540,000

2005–2006

Fruits and vegetables

Total 3,919.3

Environ Monit Assess Table 4 Summary of statistical analyses of hydrological parameters Parameter

Time period (years)

Statistically significant trend

Slope of trend

Water yield

35

Negative

−28.48

−46.77

period100 %change over mean ¼ slopestudy mean

Sub-catchment Goma Precipitation

35

Negative

−26.81

−34.63

Evapotranspiration

35

Positive

2.7

17.82

Soil water

35

Absent

2.07

9.5

Groundwater

35

Negative

−5.33

−27.64

Sub-catchment Bharamaputra 3023501 Water yield

35

Negative

−28.56

−46.78

Precipitation

35

Negative

−26.82

−34.63

Evapotranspiration

35

Positive

2.51

16.74

Soil water

35

Absent

2.4

5.8

Groundwater

35

Absent

−5.1

−25.31

state along with decreasing precipitation and increasing ET. Increase in the concentration of anthropogenic aerosols also leads to decrease in surface solar radiation. To compensate for this decrease, one of the components from radiation, latent heat flux and sensible heat flux, which tend to remain in balance with each other at the surface decreases (Ramanathan et al. 2001). In this study as well, one of these components is decreasing (sensible heat flux), whereas the other two (ET and latent heat flux) are increasing. According to Bonan (2008), over land sensible and latent heat fluxes are important determinants of microclimates. Hence, the significant positive trend shown by the sum of these fluxes is the evidence of the fact that Sikkim’s microclimate is undergoing a significant change. Positive significant trend shown by absorbed longwave radiation on the surface also indicates the same.

&

Conclusions

Acknowledgments The authors are thankful to IMD, NASA and Hydrological Information System (created by IIT Delhi) for making the data easily available. Also, analyses and visualizations used in this paper were produced with the Giovanni online data system, developed and maintained by the NASA GES DISC.

With the help of satellite data, this study has captured regional changes in the landscape, hydrological cycle, atmospheric chemistry and radiation budget of Sikkim during the past 46 years. From the analyses, the following conclusions have been drawn:

&

& &

Urban sprawl, increasing energy demands and increasing air pollution are causing rapid environmental degradation of Sikkim. An increase in temperature at an altitude of 5.5 km and the suppression of precipitation caused by increasing aerosol concentration have led to rapid retreat of Sikkim Himalayan glaciers. Eastern Himalayan glaciers are found at the altitude at which generally the aerosol layers are found. Hence, increasing AOD and concentration of aerosols have made Eastern Himalayan glaciers highly vulnerable to the impacts of climate change. This study is another proof of the fact that mountainous regions are vulnerable to elevationdependent warming. The technique of change detection can be effectively used to estimate change in a glacierised area.

Author Contributions Anubha Agrawal designed the study and conducted all analyses. All authors have contributed to the writing of the paper.

Environ Monit Assess

References Agrawal, A., & Tayal, S. (2013). Assessment of volume change in East Rathong glacier, Eastern Himalaya. International Journal of Geoinformatics, 9, 73–82. Badarinath, K. V. S., Sharma, A. R., Kaskaoutis, D. G., Kharol, S. K., & Kambezidis, H. D. (2010). Solar dimming over the tropical urban region of Hyderabad, India: effect of increased cloudiness and increased anthropogenic Aerosols. Journal of Geophysical Research, 115, 1–18. Bahuguna, I. M., Rathore, B. P., Brahmbhatt, R., Sharma, M., Dhar, S., Randhawa, S. S., et al. (2014). Are the Himalayan glaciers retreating? Current Science, 106(7), 1008–1013. Basic Statistics of NER 2006. viewed on 9 May 2013 http://www. asthabharati.org/Dia_Jan%2009/the%20N.E.%20ind.htm. Bhambri, R., Bolch, T., Chaujar, R. K., & Kulshreshtha, S. C. (2011). Glacier changes in the Garhwal Himalaya, India, from 1968 to 2006 based on remote sensing. Journal of Glaciology, 57, 543–556. Bolch, T., Buchroithner, M., Pieczonka, T., & Kunert, A. (2008). Planimetric and volumetric glacier changes in the Khumbu Himal, Nepal, since 1962 using Corona, Landsat TM and ASTER data. Journal of Glaciology, 54, 592–600. Bonan, G. B. (2008). Chapter 7: Surface energy fluxes, ecological climatology: concepts and applications. Cambridge University Press, 2nd edition. Buishand, T. A. (1984). Testa for detecting shifts in the mean of hydrological time series. Journal of Hydrology, 73, 51–69. Chen, Z., & Grasby, S. E. (2008). Impact of decadal and century scale oscillations on hydroclimate trend analyses. Journal of Hydrology, 365, 122–133. Chen, B., Huang, J., Minnis, P., Hu, Y., Liu, Z., Zhang, D., et al. (2010). Detection of dust aerosols by combining CALIPSO active lidar and passive IIR measurements. Atmospheric Chemistry and Physics, 10, 4241–4251. Chin, M., Ginoux, P., Kinne, S., Torres, O., Holben, B. N., Duncan, B. N., et al. (2002). Tropospheric aerosol optical thickness from the GOCART model and comparisons with satellite and sun photometer measurements. Journal of the Atmospheric Sciences, 59, 461–483. Goddard Earth Sciences data and Information Services Center (GES DISC), viewed on 17 June 2013 (http://disc.sci.gsfc.nasa.gov). Global Modeling and Assimilation Office (GMAO), viewed on 18 June 2013 (http://gmao.gsfc.nasa.gov/pubs/brochures/ MERRA%20Brochure.pdf). Hydrological Information System, IIT Delhi, viewed 15 February 2012, http://gisserver.civil.iitd.ac.in/natcom/. IPCC. (2007). Climate change 2007: the physical science basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press. Kaufman, R. K., & Seto, K. C. (2001). Change detection, accuracy, and bias in a sequential analysis of Landsat imagery in the Pearl River Delta, China: econometric techniques. Agriculture Ecosystems & Environment, 85, 95–105. Kendall, M. G. (1975). Rank correlation methods. London: Griffin. Kiely, G., Albertson, J. D., & Parlange, M. B. (1998). Recent trends in diurnal variation of precipitation at Valentia on the west coast of Ireland. Journal of Hydrology, 207, 270–279.

Lau, W. K. M., Kim, M. K., Kim, K. M., & Lee, W. S. (2010). Enhanced surface warming and accelerated snowmelt in the Himalayas and Tibetan Plateau induces by absorbing aerosols. Environmental Research Letters, 5, 1–10. Leavell, D. L., Portocarrero, C. (2003). Sustainability of Peruvian water resources in light of climate change. Paper presented at the XI World Water Congress, Madrid, Spain, 5–9 October 2003. Liu, X., & Chen, B. (2000). Climatic warming in the Tibetan plateau during recent decades. International Journal of Climatology, 20, 1729–1742. Lu, D., Mausel, P., Brondizio, E., Moran, E. (2002). Change detection of successional and mature forests based on forest stand characteristics using multi-temporal TM data in Altamira, Brazil, ACT publication no. 2-5. Mann, H. B. (1945). Nonparametric tests against trend. Econometrica, 13, 245–259. Menon, S., Koch, D., Beig, G., Sahu, S., Fasullo, J., & Orlikowski, D. (2010). Black carbon aerosols and the third pole ice cap. Atmospheric Chemistry and Physics, 10, 4559–4571. Ministry of Environmental and Forest (MoEF). (2011). Sikkim: state action plan on climate change (2012-2030). New Delhi: Government of Sikkim, Government of India. Mitchell, J. M., Jr. (1971). The effect of atmospheric aerosols on climate with special reference to temperature near Earth’s surface. Journal of Applied Meteorology, 10, 703–714. Narayanan, P., Basistha, A., Sarkar, S., & Sachdeva, K. (2013). Trend analysis and ARIMA modelling of pre-monsoon rainfall data for western India. Comptes Rendus Geosciences, 345, 22–27. Noia, A. D., Sellitto, P., Frate, D. F., Cervino, M., Iarlori, M., & Rizi, V. (2013). Tropospheric ozone column retrieval from OMI data by means of neural networks: a validation exercise with ozone soundings over Europe. EURASIP Journal on Advances in Signal Processing, 1–12. Noyes, P. D., McElwee, M. K., Miller, H. D., Clark, B. W., Tiem, L. A. V., Walcott, K. C., Erwin, K. N., & Levin, E. D. (2009). The toxicology of climate change:Environmental contaminants in a warming world. Environment International, 35, 971–986. Parkinson, C. L. (2003). Aqua: an Earth-Observing Satellite mission to examine water and other climate variables. IEEE Trans Geoscience and Remote Sensing, 41, 173–183. Powell, K. A., Hostetler, C. A., Vaughan, M. A., Lee, K., Trepte, C. R., Rogers, R., et al. (2009). CALIPSO Lidar calibration algorithms. Part I—nighttime 532-nm parallel channel and 532-nm perpendicular channel. Journal of Atmospheric and Oceanic Technology, 26, 2015–2033. Qin, J., Yang, K., Liang, S., & Guo, X. (2009). The altitudinal dependence of recent rapid warming over the Tibetan Plateau. Climate Change, 97, 321–327. Racoviteanu, A. E., Arnaud, Y., Williams, M. W., & Ordonez, J. (2008). Decadal changes in glacier parameters in the Cordillera Blanca, Peru, derived from remote sensing. Journal of Glaciology, 54, 499–510. Rajab, J. M., MatJafri, M. Z., Lim, H. S., & Abdullah, K. (2009). Satellite mapping of CO2 emission from forest fires in Indonesia from AIRS measurements. Modern Applied Science, 3, 68–75. Ramachandran, S., Kedia, S., & Srivastava, R. (2012). Aerosol optical depth trends over different regions of India. Atmospheric Environment, 39, 338–347.

Environ Monit Assess Ramanathan, V., Crutzen, P. J., Kiehl, J. T., & Rosenfeld, D. (2001). Aerosols, climate, and the hydrological cycle. Science, 294(5549), 2119–2124. doi:10.1126/science.1064034. Rana, A., Bengtsson, C. B. U. L., & Sarthi, P. P. (2012). Trend analysis for rainfall in Delhi and Mumbai, India. Climate Dynamics, 38, 45–56. Rangwala, I., & Miller, R. (2012). Climate change in mountains: a review of elevation-dependent warming and its possible causes. Climate Change, 114, 527–547. Schmidt, S., & Nusser, M. (2012). Changes of high altitude glaciers from 1969-2010 in the Trans-Himalayan Kang Yatze massif, Ladakh, Northwest India. Arctic Antarctic and Alpine Research, 44, 107–121. Shahgedanova, Nosenko, G., Khromova, T., Muraveyev, A. (2010). Glacier shrinkage and climatic change in the Russian Altai from the mid‐20th century: an assessment using remote sensing and PRECIS regional climate model, Journal of Geophysical Research, 115, D16107. Sharma, A. R., Kharol, S. K., & Badrinath, K. V. S. (2009). Satellite observations of unusual dust event over North-East India and its relation with meteorological conditions. Journal of Atmospheric and Solar-Terrestrial Physics, 71, 2032–2039. Shrestha, A., Wake, C. P., Mayewski, P. A., & Dibb, J. E. (1999). Maximum temperature trends in the Himalaya and its vicinity: an analysis based on temperature records from Nepal for the period 1971–94. American Meteorological Society, 12, 2775–2786. TERI (The Energy and Resources Institute). (2013). TERI energy data directory and yearbook 2012/13. New Delhi: TERI.

United States Geological Survey: Earth Explorer, viewed on 7 January 2011 http://earthexplorer.usgs.gov/. Vuille, M., & Bradley, R. S. (2000). Mean annual temperature trends and their vertical structure in the tropical Andes. Geophysical Research Letters, 27, 3885–3888. Wuebbles, D. J., & Hayhoe, K. (2002). Atmospheric methane and global change. Earth Science Reviews, 57, 177–210. Yang, C., Yang, H., Guo, S., Wang, Z., Xu, X., Duan, X., & Kan, H. (2012). Alternative ozone metrics and daily mortality in Suzhou: The China Air Pollution and Health Effects Study (CAPES). Science of the Total Environment, 426, 83–89. Yerramestti, V. S., Sharma, A. R., Navlur, N. G., Rapolu, V., Dhulipala, N. S. K. C., & Sinha, P. R. (2013). The impact assessment of Diwali fireworks emissions on the air quality of a tropical urban site, Hyderabad, India, during three consecutive years. Environmental Monitoring and Assessment. doi:10.1007/s10661-013-3102-x. Zha, A., Gao, J., & Ni, S. (2003). Use of normalized difference built-up index in automatically mapping urban areas from TM imagery. International Journal of Remote Sensing, 24, 583–594. Zhang, Q., Wang, J., Peng, X., Gong, P., & Shi, P. (2002). Urban built-up land change detection with road density and spectral information from multi-temporal Landsat TM data. International Journal of Remote Sensing, 23, 3057–3078. Zlatev, Z., & Moseholm, L. (2008). Impact of climate changes on pollution levels in Denmark. Ecological Modelling, 217, 305–319.