Solar Potential in the Himalayan Landscape
Ramachandra T V 1,2,3,*                Gautham Krishnadas 1                Rishabh Jain 1
1 Energy & Wetlands Research Group, Center for Ecological Sciences [CES], 2 Centre for Sustainable Technologies (astra), 3 Centre for infrastructure, Sustainable Transportation and Urban Planning [CiSTUP], Indian Institute of Science, Bangalore, Karnataka, 560 012, India
*Corresponding author:


Energy plays a pivotal role in the development of a region. However, energy shortages in recent times, the imminent energy crisis and threat of climate change have focused the attention for a viable sustainable alternative through renewable sources of energy. Sun being the vital source of renewable energy manifested in different forms in the solar system, it is necessary to understand the mechanism of energy flow involved. The geometry of the earth-sun movements causes large spatial, diurnal and seasonal variations in the amount of solar radiation received on earth. The 23.5° tilt of the earth’s rotational axis with respect to the plane of orbital revolution causes larger annual variations near the poles and smaller variations near the equator [1]. Due to the variations in the sun-earth distance, intercepted solar radiation fluctuates by ±3.3 % around its mean value. The variations due to sunspots, prominences and solar flares, can be neglected as they constitute small fraction compared to the total energy emitted by the sun. The average solar radiation falling on the earth’s atmosphere called the solar constant is estimated to be 1.36 kW/m2. The presence of clouds, suspended dust, gas molecules and aerosols in the atmosphere, through absorption and scattering, attenuates the incident solar energy radiation (also called insolation). A fraction of 0.3 of the total incident solar energy called the albedo of the earth-atmosphere system is reflected back to the space [2]. The remaining fraction of solar energy in the form of Direct and Diffuse insolation which comprises the Global Horizontal Insolation (GHI) is utilized in many processes like heating and illuminating the earth, photosynthesis, growth of vegetation, evapotranspiration, snow ablation as well as solar energy based applications.

Solar radiation data is essential for designing solar energy based applications as it is for meteorology, climatology, oceanography, agriculture, forestry and other domains. Solar energy applications like photovoltaic based off-grid and grid connected power generation, solar water heaters, thermal concentrators, solar cookers, desalination plants, passive building heating etc. demand reliable information on GHI at different regions of interest. This could be estimated from insolation data procured from on-ground pyranometric network, models based on meteorological parameters or those based on satellite data.

 Solar potential assessment

 Insolation data from pyranometric network: Conventionally, solar potential of a region is assessed using the insolation data obtained from ground based radiation stations. There are 572 radiation stations around the world which provide global, diffuse and direct insolation data on hourly, daily or monthly basis to the World Radiation Data Centre (WRDC) apart from other archived data [3]. An efficient and reliable pyranometric network, incurs heavy investment in terms of capital, maintenance and manpower. Developing countries like India with a vast land area of 3287263 km2, find it hard to have a sufficient spatial coverage for insolation measurements. The Indian radiation network under the Indian Meteorological Department (IMD) has merely 44 radiation stations spread across the country. Data from ground based solar radiation monitoring stations are liable to errors due to calibration drift, manual data collection, soiling of sensors and non-standardization of measuring instruments [4]. The World Climate Research Program in 1989 had estimated a cumulative uncertainty of 6% -12% in measurements from conventional solar monitoring stations [5]. Most of the data available from radiation stations are at non- uniform and interrupted periods as shown by Alberto Ortega et al [6] for Chile which was reported to be 3 months to 21 years.  Measurement uncertainties are below 3% at certain research-class sites meant for atmospheric or precision studies, but these are exceptions [7]. The radiation network in India expanded from 4 stations in 1957 to 44 as of today, measuring one or more parameters like Global, Direct or Diffuse radiation at different locations.  These datasets provide temporal information (in minutes) which is essentially important for solar energy applications and design. Annual insolation varies with change in angle of incidence of the solar radiation. It is reported that a minimum of 7-10 years of solar radiation data are required to get a long term mean within 5 % [8].

Insolation data from meteorological parameters: Insufficient insolation data consequent to sparse radiation networks, lead to different parametric (Iqbal and ASHRAE Model) and decomposition models (Angstrom, Hay, Liu and Jordan, Orgill and Hollands Model). These models employ theoretical or empirical methods for interpolation or extrapolation of measured meteorological values to derive insolation data [1]. These models use parameters like sunshine duration [9, 10], temperature [11], rainfall [12], cloud cover [13], extraterrestrial radiation [14], etc. for estimating solar radiation over a given location. Artificial Neural Networks (ANN) are also being utilized for deriving insolation data based on climatological and geographical parameters [15]. These methods show low Relative Root Mean Square Errors (RRMSE) of 5% on validation with in-situ measurements.

Insolation data from satellites: Remote sensing data from polar satellites at ~850 km with higher spatial resolution and geostationary satellites at ~36000 km with higher temporal resolution are being utilized for estimation of insolation based on different subjective, empirical and theoretical methods [16, 17]. Subjective methods involve subjective interpretation of cloud cover from the satellite images and its statistical relationship with atmospheric transmittance. Empirical methods depend on functional relations based on satellite derived data and available solar radiation data which are customized for any place and time. Empirical methods are classified into two models: statistical and physical. Another method which simulates radiant energy exchanges taking place within the earth atmospheric system, hypothetically not demanding empirical calibration of model parameters is called theoretical method. The broadband and spectral models are attached to this method [18].

The solar potential of Kampuchea was estimated based on a statistical model [19] with the visible and infrared images obtained from Japanese Stationary Meteorological Satellite GMS-3 along with ground based regression parameters and concluded that seasonal average of daily insolation depended more on the topography of the region than on seasonal variations. The RRMSE of the monthly average insolation was shown to vary from 5.7% to 11.6% with the mean of ground values as the cloudiness of the region increased. The solar potential of Pakistan was estimated by employing a physical model [17] based on the images collected from the Geostationary Operational Environmental Satellite (GOES) INSAT which scanned the region four times per day. The results indicated that the desert and plateau regions received favourable GHI while the monsoon had its influence on reduced insolation in the Eastern and Southeastern Pakistan, for the months of August, October and May.

The daily Direct and Global insolation in 6 locations of India (from 2000 to 2007) were estimated based on a statistical model using Meteosat images of 5 X 5 km spatial resolution addressing the aerosol component in the atmosphere. These were validated against the surface global insolation for 5 locations with a RRMSE of 12 % [20]. Stretched Visible Infrared Spin Scan Radiometer (S-VISSR) images of Geostationary Meteorological Satellite (GMS) of 15 minute interval high spatial resolution (0.01⁰X0.01⁰) for a period of 3 months in 1996 [21] were analysed to develop a bispectral threshold technique with respect to earth-atmosphere albedo and infrared temperature. The RRMSE with respect to ground data from 67 weather stations was calculated to be 25 % for hourly and 12% for daily insolation. Similar endeavor generated hourly and monthly basis Global as well as Direct insolation maps using SOLARMET physical model applied over 7 years high resolution Meteosat satellite images [22]. A physical model based on visible range satellite images considering the climatological aspects of hourly GHI useful for designing solar energy systems was designed [23] and monthly average hourly insolation datasets with RMSE ~10% when validated with 25 pyranometric stations were obtained. The extent of high solar resource availability zones in India was quantified on the basis of NASA SSE GHI data and power generation potential of concentrated and photovoltaic applications was demonstrated in the wake of encouraging solar policies in the country [24].

Satellite-data based models do not show much difference in performance as the primary source of ambiguity for all these models is the influence of cloud pattern. It has been proved that interpolations and extrapolations of available ground data for predicting solar radiation over distances beyond 34 km show higher RMSE compared to satellite-data based measurements [25]. RMSE for different models have been found to be within 20% for hourly values and 10% for daily values [26]. Today satellite data for about 20 years are available, which reduces the possibility of error due to annual variations in insolation. Nevertheless, ground data is indispensable at this juncture for validation and model improvement of satellite based data.

Citation : Ramachandra. T.V., Gautham Krishnadas and Rishab Jain, 2012. Solar Potential in the Himalayan Landscape., International Scholarly Research Network, ISRN Renwable Energy, Volume 2012, 13 pages.
* Corresponding Author :
  Dr. T.V. Ramachandra
Energy & Wetlands Research Group, Centre for Ecological Sciences, Indian Institute of Science, Bangalore – 560 012, India.
Tel : 91-80-23600985 / 22932506 / 22933099,      Fax : 91-80-23601428 / 23600085 / 23600683 [CES-TVR]
E-mail :,,     Web :
E-mail   |   Sahyadri   |   ENVIS   |   GRASS   |   Energy   |   CES   |   CST   |   CiSTUP   |   IISc   |   E-mail