Sensing deficiency

Observations in a CAG report on the efficiency of ISRO's seven remote-sensing satellites operational during 2003-08 are not complimentary.

THE celebration of the successful launch on April 20 of the 18th Indian remote-sensing (IRS) satellite Resourcesat-2 aboard the Polar Satellite Launch Vehicle (PSLV-C16) needs to be tempered with the observations in the recent report of the Comptroller and Auditor General (CAG) on the effectiveness and efficiency of utilisation of the Indian Space Research Organisation's (ISRO) seven remote-sensing satellites that were operational during 2003-08, which are less than complimentary.

The CAG conducted a performance audit of the National Remote Sensing Centre (NRSC), the nodal agency for ISRO's remote-sensing activities that involve acquisition and archiving of satellite/aerial remote-sensing data and its dissemination. The activities of the NRSC are governed by the Remote Sensing Data Policy (RSDP) of 2001, which vests it with the sole authority to acquire and disseminate all remote-sensing data in the country. The NRSC undertakes remote-sensing application projects of national importance identified by the National Natural Resources Management System (NNRMS), a body of the Planning Commission. The audit covered a six-year period from 2003-04 to 2008-09.

(Until August 2008, the NRSC was known as the National Remote Sensing Agency, which was an autonomous organisation under the Department of Space (DoS). It was converted into a wing of ISRO apparently because of the constraints faced by the agency in dealing with programmes of national importance on account of it being a society. From September 1, 2008, the NRSC's funding has come from budgetary allocations to the DoS instead of the earlier grants-in-aid.)

At present there are 10 operational remote-sensing satellites, including the recently launched Resourcesat-2, but only three of these Resourcesat-1, Cartosat-1 and Cartosat-2 formed part of the CAG's audit. The remaining four IRS-1C, IRS-P3, IRS-1D and Oceansat-1 have completed their missions since 2008. Remote-sensing satellites that have been launched by ISRO, besides the ones covered by the audit period and are currently operational, include the Technology Experiment Satellite (TES), the Indian Mini Satellite (IMS-1), Cartosat-2A, the Radar Imaging Satellite (RISAT-2), Oceansat-2, Cartosat-2B and Resourcesat-2. The NRSC also acquires remote-sensing data from a few foreign satellites, which for the audit period included Landsat, Ikonos, Quickbird, Radarsat, European Remote-sensing Satellite (ERS), and satellites of the National Oceanographic and Atmospheric Administration (NOAA) and Moderate Resolution Imaging Spectroradiometer (MODIS) aboard NASA's satellites.

While IRS-P3, designated as an experimental remote-sensing satellite, is one of the projects audited, the non-inclusion of the TES, which was launched on October 22, 2001, and had a one-metre resolution imaging capability for the first time, in the audit is interesting. This would seem to confirm the general perception that the satellite is not service-oriented but meant for strategic applications, including military reconnaissance. Even though image acquisition from the TES and processing are done at the NRSC, data utilisation is perhaps entirely administered by the newly established centre under ISRO/DoS called the Advanced Data Processing Research Institute (ADRIN), which is located in Hyderabad. So, TES-related projects are all probably funded directly by the DoS and not by the NRSC. This is probably why the CAG, which was assessing the performance of the NRSC, was not provided with information on the TES' operations and data utilisation for scrutiny. Also, though they were launched in 2008, Cartosat-2A and IMS-1 were not included in the audit for the obvious reason of these having been barely operational at the time of the audit.

The total investment in the seven remote-sensing projects (Table 1) under audit was Rs.2,205.86 crore, of which Rs.1,468.59 crore was towards the cost of the satellites and their launches and Rs.737.27 crore towards expenditure incurred on the earth observation programme during 2003-08. The amount Rs.737.27 crore subsumes in it the total grants-in-aid of Rs.657.78 crore made to the NRSC during the audit period until it became a unit of the ISRO/DoS in August 2008.

It is the number of scenes acquired during its operational life that determines the performance of a given remote-sensing satellite. This, in turn, depends on factors such as the design capacity of the satellite, elevation and the visibility during its orbits over the imaging region, the payloads such as the on-board imaging sensors and the availability of the on-board solid-state recorder. The ground stations of the NRSC acquire the scenes captured by the satellite when the satellite passes over the station.

With regard to remote-sensing satellites' utilisation, data acquisition and processing, the CAG found that, in terms of the number of scenes captured or, equivalently, their capacity utilisation, the performance of only four of the seven satellites was above 50 per cent of their maximum capacities. Maximum capacity for the audit purpose was taken as the maximum number of scenes captured by the satellites in any of the years during the audit period. The CAG could not assess the performance with reference to the design capacity, as it should have been, because the NRSC, inexplicably, did not furnish that information to the CAG even though it had been requested for. But even in the comparison with the maximum capacities, the performances of the satellites came out below par.

The NRSC apparently argued that the scenes acquired by the earth station (data product generation capacity) should be taken as the design capacity, which was not acceptable to the CAG. Indeed, the CAG found that there was a significant gap between the number of images captured by the satellite and the data product generation capacity because of the inherent capacity limitations of the ground station at Shad Nagar in Hyderabad. In response to this observation of the CAG, the DoS stated in December 2009 that the data product generation capacity was being augmented by establishing a polar earth station at Svalbard, Norway.

Table 2 gives the percentage of actual average capacity utilisation against maximum capacity utilisation. Satellites IRS-P3, IRS-1C and Oceansat-1 could respectively utilise only 32, 45 and 50 per cent of their maximum capacities. In its response to this observation, ISRO stated that P3 was an experimental satellite which stopped functioning in 2003 itself.

Interestingly, even as only 15 images were captured in 2003-04 and zero images in 2004-05 and 2005-06, ISRO's website says: Mission completed in January 2006 after serving for 9 years and 10 months. Further, in July 2009, ISRO, in its reply, attributed the underperformance of P4 to on-board spacecraft power constraints since 2001-end. Mission P4 was completed in August 2010, according to ISRO. The CAG has commented that the NRSC/DoS did not assess the impact of lower capacity utilisation resulting from technical problems on remote-sensing projects of national importance.

The CAG report has also looked at the Return on Investment (ROI) for the seven satellites. Towards this, the revenue generated from the sale of data products generated from each satellite is compared with the investment, which includes the capital expenditure (satellite and launch costs) and the yearly operational & maintenance (O&M) expenditure on it. The following observation of the CAG is pertinent: No benchmark relating to recovery of expenditure incurred or ROI was fixed by DoS.

The analysis of operational returns alone, where return against operational expenditure is compared instead of return on total investment, revealed that even the operational returns for the seven satellites were negative in all the years, implying that revenue from sale of data products was not sufficient even to meet the operational expenditure. As shown in Table 3, the income from sale was very low and varied from 17.8 per cent to 26.13 per cent of the operational expenditure, and the under-recovery of operational expenditure from the sale of data products was significant and worked out to Rs.578.81 crore. Since even operational returns were in the negative, there was no question of recovery of the capital investment of Rs.1,468.59 crore. If one included that as well, the recovery of Rs.158.46 crore shown in Table 3 is a mere 7.18 per cent of the total investment of Rs.2,205.86 crore.

In its response, the NRSC stated in September 2008 that the price of data products had been arrived at by considering only the operational cost of the satellite product generation at the NRSC. Capital costs such as launch costs and satellite fabrication were not included. Further, the NRSC's contention was that the ROI should include the intangible benefits accrued through utilisation of the data as well.

The reply was not acceptable to the CAG as, for one, the NRSC had not quantified the intangible benefits. Further, since the data product pricing was based on the recovery of operational costs, with a mere 21.49 per cent average of such recovery, even this had not been achieved. The CAG has recommended that the NRSC consider all costs to arrive at its pricing policy and also revisit its operational policy to recover at least its operational costs. For long-term sustainability of the programme, the CAG added, the NRSC should improve the generation of revenue from the sale of data products. The CAG report, in fact, has dwelled in detail on the sale of data products by the NRSC. One of the points that emerge from the CAG analysis is that the government sector was the major consumer of the products during the period 2003-09, accounting for more than 80 per cent, and sales to the private sector showed only a marginal increase during 2008-09 (Table 4). This, the CAG has observed, meant that the NRSC had failed to implement the recommendation made by the departmental Parliamentary Standing Committee nearly a decade earlier. In its report on the DoS for 1996-1997, the Parliamentary Standing Committee had observed that more and more private entrepreneurs should be associated with the remote-sensing data utilisation programme.

In response to the CAG's observation, ISRO stated in July 2009 that the private sector procured high-resolution data mainly for infrastructure and utility mapping applications and this was made available by the launches of Cartosat-1 and Cartosat-2 satellites. But the CAG has noted that despite the launch of Cartosat-1 in May 2005 and Cartosat-2 in January 2007, sales to the private sector did not increase significantly and has also commented that the efforts of the NRSC were not adequate in customising the data according to the needs of the private sector and also in exploring the possibility of widening the customer base.

It may be pointed out here that, while Cartosat-1 with its 2.5-m resolution data and stereo-imaging capability, useful particularly for Geographic Information System (GIS) applications involving digital elevation models and the like, has performed well, it is widely known that useful data from Cartosat-2, with its 1-m resolution panchromatic data capability, have not been made available to customers at all, a fact that is not acknowledged by ISRO. The percentage capacity utilisation of Cartosat-2 shown in Table 2 is also perhaps a reflection of that.

The CAG has also compared the international rate of data products with ISRO's pricing policy to international customers. The NRSC is the nodal agency that sells IRS data products directly to Indian customers, whereas for international customers data are routed through Antrix Corporation Ltd. (ACL), ISRO's marketing arm, at 2.5 times the international price, which includes a 100 per cent commission to ACL.

According to the report, the rate at which the NRSC sold high resolution data to ACL ranged from $0.08/km {+2} to $2.5/km {+2}. ACL, in turn, sold these products to international customers whose rate varied from $0.16/km {+2} to $8.6/km {+2}. As compared to ACL's average price of $4.38/km {+2}, international market prices for similar high resolution data varied from $17/ km {+2} to $35/km {+2}, which is perhaps the rate at which the NRSC itself procured remote-sensing data from foreign satellites such as Radarsat, Ikonos and Quickbird on demand. Thus the average international price was about six times that of the price charged by ACL from international customers.

According to ISRO's response to the CAG in July 2009, a strategic pricing policy had to be adopted to help penetrate the global market, especially after the launch of Cartosat-2. Accepting the CAG's recommendation to streamline the pricing of IRS data products, especially those sold internationally, the NRSC stated in February 2010 that it would periodically review the pricing policy keeping in view the costing principles and appropriate marketing strategy. It also stated that it was entering into a memorandum of understanding (MoU) with ACL for international distribution of data, which is expected to be ready shortly.

As per ISRO's policy, a proposal to launch a remote-sensing satellite is approved taking into account user requirements and data needs in various remote-sensing applications such as resource survey, mapping applications and oceanographic applications. Having found that the NRSC has been unable to recover even the operational costs of the satellites from the sale of data products, the CAG also tried to establish how effectively data needs were assessed by the DoS before approving satellite proposals.

Scrutinising the proposal documents of Oceansat-1, Cartosat-1 and Resourcesat-1, the CAG has observed that proper assessment of data needs in various thematic areas was not carried out. Further, according to the report, ISRO also did not make available the proposal files pertaining to IRS-1C, ID and Cartosat-2, which too had been requested for. ISRO similarly expressed its inability to produce the information that the CAG had sought in August 2008 on demand and supply of theme-wise data to assess the data gap in the areas of urban planning, drought monitoring, land-use and land cover mapping, underground water resource mapping, mineral prospecting, environmental impact analysis, and so on. In October 2008 the NRSC responded stating that it had plans to collect theme-wise data only in the future. Accepting the CAG's observation, ISRO stated in July 2009 that efforts would be made to collect information on the intended use of data products disseminated by the NRSC.

One of the chief aims of initiating remote-sensing activities in the country, which began with acquiring data from foreign satellites such as Landsat and subsequently establishing the IRS satellite system with the launch of the first indigenous operational remote-sensing satellite IRS-1A in 1988, was to provide efficient operational resource survey services and implement remote-sensing application projects in various sectors of the economy. To carry out earth observation activities effectively with the above objectives, the NNRMS was set up in 1983 under the Planning Commission, which was mandated to address specific issues relating to applications of remote sensing in different thematic areas.

The DoS was made the nodal department responsible for the implementation of the mandate of the NNRMS through the NRSC. The NRSC thus was responsible for ensuring that the deliverables supplied by them for implementing projects of national importance, such as Command Area Development (CAD), National Wasteland Mapping (NWM) and the Rajiv Gandhi National Drinking Water Mission (RGNDWM), were effectively utilised. However, the CAG has observed that the NRSC did not play the necessary proactive role in ensuring that the objectives of the various thematic projects were fully achieved. Further, the various thematic Standing Committees of the NNRMS did not meet periodically to coordinate and monitor implementation.

In December 1997, the Ministry of Water Resources (MoWR) had entrusted the NRSC with a project costing Rs.2.5 crore for completion by December 1999. The project, which was aimed at providing information on irrigated area, major crops, water logging, soil salinity and so on for three crop seasons in 14 selected CAD projects, was coordinated by the Water Resources Standing Committee of the NNRMS. The NRSC was also to conduct end-of-study workshops in the five participating States. According to the CAG report, the NRSC completed the study only in September 2001. Further, the workshops were held only after a lapse of three years, between November 2004 and March 2005. The workshops concluded that the data of 1985-86 to 1997-98 that the NRSC had studied and interpreted would no longer be useful in planning irrigation schemes after 2005.

While in September 2008 the NRSC laid the blame for the delay on the participating States, in July 2009 it stated that the primary responsibility for implementing application projects was that of the Ministry concerned. Rejecting their responses, the CAG has noted that the NRSC, as the member secretary of the concerned Standing Committee, was responsible for coordination with the authorities of the CAD projects. No meeting of the committee was apparently convened between April 2003 and March 2005. This indicates, notes the CAG, that the NRSC did not coordinate effectively leading to delays and consequent wastage of resources as the data collected and archived could not be used in planning irrigation schemes beyond 2005.

In 1986, the Ministry of Rural Development (MoRD) entrusted the NRSC the task of mapping the wasteland inventory. While the scheduled date of completion was apparently not to be found in any of the NRSC's records, the wasteland mapping mission was actually completed only in 2000, after 14 years. The CAG has observed that, as a result, only 13.44 per cent of the target set for rehabilitation of wasteland could be achieved. In 2003, the Ministry entrusted another project to update details of wasteland at an estimated cost of Rs.4.98 crore. The project was to be coordinated by the relevant Standing Committee of the NNRMS. One of the objectives of the project was to assess the impact of the rehabilitation activities. Though the updating was completed in March 2005, the NRSC did not carry out the impact assessment study. Further, the Standing Committee also did not meet between April 2003 and March 2005. The ISRO responded in July 2009 by stating that the Ministry did not furnish the necessary input data.

The drinking water mission RGNDWM provides an interesting example. In 1998, the MoRD entrusted the NRSC with the work of preparation of geo-morphological maps for identification of sources of drinking water for all the non-covered' and partially covered' habitats by 2000. The NRSC, notes the report, could complete the work only in 10 States by 2005. The NRSC planned to complete the groundwater survey in another 10 States by June 2010 and work was yet to be taken up in the balance seven States. Given the fact that remote sensing had immense success in identifying water resources, the success rate being as high as 90 per cent, there was need to complete the work expeditiously in the remaining 17 States, the CAG has observed.

The report also notes that the concerned Standing Committee meant to coordinate the activities of this project met only once between April 2003 and November 2008, in February 2006. As in the case of the above two projects, lack of coordination resulted in non-achievement of the objective of identification of sources of drinking water in all States, a pressing need in these days of extreme water stress all over the country. The CAG report has critically examined several other application projects as well. In all, the CAG had identified a total of 105 projects (45 government and 60 user projects) out of a total of 348 (132 government and 216 user projects) for audit scrutiny.

The CAG has also audited the NRSC's other activities such as aerial remote sensing using two beach aircraft as against satellite remote sensing, training in remote sensing and financial management of the organisation. According to the report, the NRSC had accepted most of the recommendations made by the CAG and also submitted details of actions already taken and actions proposed to be taken on the recommendations made. It remains to be seen perhaps in another CAG report or by the Parliamentary Standing Committee whether these actions have made any difference in the effective utilisation of the data from remote-sensing satellites launched after the audit period, in particular the most recently launched Resourcesat-2.