CONTINUE?

Expanding the text here will generate a large amount of data for your browser to display

Chapter 18. Photovoltaic solar-powered pump irrigation in Pakistan*

* Contributed by the ILO.¹

MOST OF PAKISTAN’S agricultural activity (and population) is concentrated in the Indus basin which includes the Peshawar valleys of North Western Frontier Province (NWFP), most of the province of the Punjab, and part of Sind. Of this area a small part, the barani area, has sufficient rainfall for cultivation without recourse to irrigation. This area is the oldest settled and most heavily populated part of the country. Farm sizes are small and only small amounts of land area are available for cultivation under tenancy arrangements.

Until the twentieth century, the rest of the Indus basin was owned by landlords who had acquired property rights but were unable, in the absence of adequate rainfall, to put the land into good productive use. However, at the beginning of the twentieth century, the development of what is now the largest continuous canal system in the world, was embarked upon. This led to the establishment of “canal colonies” populated mainly by migrants from the north who either acquired the land from landlords or were able to settle for some form of tenancy agreement. Landholding sizes in this area are greater than those in the barani areas, and a large percentage of the total cultivated areas is operated by tenants. The total canal network commands an area of 11 million hectares.

Groundwater has been exploited for cultivation by means of animal-powered Persian wheels, lifting water from open surface wells. However, during the past 20 years or so, deep tube wells driven by diesel or electricity have been introduced. Table 18.1 gives a breakdown of the area irrigated by different methods in 1971/2 and 1978/9.

From this table, it can be seen that the percentage of users of deep tube-wells increased while that of users of the Persian wheel dropped. However, tubewells have mainly benefited large landholders (of holdings above 5 hectares). Two main problems in the past which have prohibited the development of smaller tubewells suitable for farms of less than 10 hectares have been firstly, the absence of suitably-sized irrigation units and secondly, the lack of credit facilities to small farmers. The need for small irrigation units is also evident in areas where canal water distribution is poor and where sufficient canal water is only available in the summer season.

Table 18.1: Area irrigated by different sources 1971-72 and 1978-79 (4 × 10⁵ hectares)

Source: Government of Pakistan, Agricultural statistics of Pakistan. Islamabad, 1979.

Notes:

1. Totals may not add up exactly due to rounding.
2. Figures in brackets indicate percentages

Thus, potential for small photovoltaic (solar) pumping units exists where groundwater level permits, to provide irrigation for farmers with small holdings² as well as to supplement canal supplies in areas where these supplies are seasonal.

I. THE FIELD TRIALS OF SOLAR PUMPS

The trials of solar-powered irrigation pumps in Pakistan started in March 1981 when the Intermediate Technology Industrial Services (ITIS) introduced 20 solar-powered micro-irrigation units. These trials were conducted in collaboration with the Agricultural Development Bank of Pakistan (ADBP) and the Pakistan Agricultural Research Council (PARC).

Objectives of the Field Trials

The field trials which were undertaken in collaboration with the ADBP had the following objectives:³

(i) to demonstrate solar pumps with the aim of creating a demand for them which could be met by purchases with credit provided by the bank;

(ii) to discover what area of different crops could be successfully irrigated with pumps throughout the year;

(iii) to indicate in what circumstances (for what seasons, crops and districts) solar pumps are most suited;

(iv) to indicate possible design changes required to future generations of solar pumps.

Description of the Photovoltaic System

The system consists of three main components:

(i) a set of two solar arrays producing a total of 250 watts under 100 mW/cm^-2 insulation and cell temperature of 28 degrees Centigrade.

(ii) a submersible motor/pump set using a brushless DC motor rotating at 3,000 rpm. at 60 volts, coupled to a single-stage, vertical axis, centrifugal pump. The pump set is made mainly of plastic which makes the unit very light thus facilitating its movement in and out of the wells.

(iii) a maximum power point tracker (MPPT) is used to match the varying power output of the array to the varying pump load to enable maximum utilisation of the available power.

Choice of Field Trial Sites

The criteria used by the ITIS for the choice of the trial sites were: the farm size should not be above five hectares; the farmer should have previous record with the bank; good quality groundwater should be available within 3 m; there should be free vehicular access to the farm; the farmer should have already used surface irrigation on his farm and should accept all responsibility for the security of the pump while it is in his possession. However, in many cases, the choice of sites did not conform to the above criteria partly due to difficulties in locating areas which simultaneously fulfilled all of these. As a result, seven out of the 14 pumps were given to farmers with landholdings well over 30 hectares who already had an adequate supply of irrigation water from their own tubewells.

The majority of the pumps were located in three regions: Muzaffargarh, Sukkur and Swabi. This was because these regions are characterised by: availability of shallow (3 m deep) low salinity (less than 2,000 EC) groundwater; a non-perennial canal supply; a comparatively small number of public and private tubewells; an abundance of open surface wells at which to install pumps without farmers incurring additional costs for construction, and a large number of small-holder farmers.

Results from Field Trials

Reported use of solar pumps in the test sites is given in Table 18.2. However, the information collected could not provide the material required for rigorous technical and/or socio-economic evaluation of the system. This was due to several reasons. First, the data were scanty due to failure of the farmers to maintain adequate records. Second, the fact that some of the pumps were installed in large landholdings which already possessed other sources of irrigation resulted in underutilisation of these solar pumps. Third, due to lack of experience in the use of the pumps the data collected during the first year could not be treated as an accurate indication of the pattern of use. Fourth, some of the pumps had developed mechanical faults during the first year.

II. EFFECTS OF SOLAR PUMPS ON FARM PRACTICE

In two of the farms in the Peshawar region the solar pumps replaced the Persian wheel and the farmers had already sold their bullocks and dismantled the wheel. No change in cropping pattern was observed as the performance (flow rate) of solar pumps was comparable to that of the Persian wheel which they replaced.

At Muzaffargarh where water was purchased from private tubewell owners to supplement canal supply, the area of fallow land was found to decline thereby increasing cropping-intensity. All the farmers planted vegetables which were to be irrigated by the solar pumps.

Reliability of Solar Systems

The importance of reliability of irrigation systems in farming cannot be overemphasised, especially in cases where the degree of sophistication of the technology renders immediate maintenance and repair difficult.

During the field trials, 12 systems experienced some kind of fault (70 per cent failure rate). Failure of the motor/pumpset which occurred in seven systems was possibly caused by:

- a fault in the MPPT thus exposing the system to overvoltage (140 V) and causing the motor to overload;

- fault in the motor circuitry;

- motor overheating caused by running the pump while it is not immersed.

Although the first two faults could have originated from manufacturing defects,⁴ the lack of adequate supervision and experience could also be contributory factors. The third fault could be eliminated through training and experience.

Inability of the plastic connectors to withstand the heat and misuse to which they were subjected was another problem encountered by the system. The non-availability of spare or substitute connectors left the systems idle for some time.

No faults were detected on the solar panels themselves, which were handled by the farmers with the utmost care. All of them had built mud enclosures or huts in which they stored the panels overnight and animal-proof fences to protect them during the day.

Table 18.2. Summary of data obtained from ADBP farm sites

N.R. no records kept

* Farmer used alternative irrigation source in conjunction with the solar pump.

Sources: M. Howes, The potential for small-scale solar-powered irrigation in Pakistan, Institute of Development Studies, Brighton, 1982 and R.G. Pallett, Solar-powered irrigation pumps in Pakistan, Hydraulics Research Station Ltd., Wallingford, 1982.

Viability of Solar-pumping

The viability of solar pumps is influenced by technical and socio-economic factors which are discussed below.

Technical factors

1. Meteorological. Solar radiation is required to power the pumps and the amount available at different times of the year is important for their operation. Variables such as latitude, season, cloud cover and atmospheric pollution can affect the amount of solar radiation and thus pump operation. In Pakistan, dust in the atmosphere during the period immediately preceding the monsoon season reduces peak radiation values from 85 mW cm^-2 to as low as 70 mW cm^-2 as compared to the related value of 100 mW cm^-2 for the SEI cells.

Field trials revealed that the cell temperatures far exceeded the rated value of 28 degrees Centigrade (cell temperatures as high as 75 degrees Centigrade were experienced) resulting in reduced efficiency of the system.⁵

As a result of the above two factors the peak array output was recorded as 55 to 60 per cent of the rated power output and the measured head/discharge curve for the system was considerably lower than expected.

2. Open well hydrology. The solar pumps were installed in open wells which were originally operated by the Persian wheel, and powered by bullocks or camels. Peak flows were estimated to be between 1.5 and 3 litres per second, independent of water table depth which could be 6m or more. Since Persian wheels are operated intermittently (usually four hours each in the morning and evening) the wells are allowed to recharge in between pumpings.

In the case of solar photovoltaic pumps operating all day the water level would drop depending on the pumping head, the insolation rate and the specific yield of the well. A drop in the water level results in a low discharge rate until an equilibrium is struck.

Increasing the well-depth would not affect the recharge rate since it appeared that the major recharge of wells in the areas covered in Pakistan was through the well base and not through the sides. Instead, it is recommended that boreholes be drilled through the well base thus converting the wells into dug-cum-bore-wells.

3. Irrigation water management. The large diversity between discharge rates of the solar pumps (two litres per second) and that obtained from the canal system (28 litres per second) created a number of problems. Existing farm layout is with level borders or basins which are particularly adapted for large water supply rates.⁶ With the low pumping rates, there was difficulty in estimating the depth of application; there were large losses of water through seepage and deep percolation caused by inappropriate channels and plot sizes; more time was spent by the farmer on irrigating with the low-flow solar pump which was unable to supply large pre-irrigation application in a short time period to enable sowing to be carried out soon after harvesting the previous crop. Since it is unlikely that farmers would want to change their familiar layout systems to furrow or trickle irrigation in order to accommodate the low-flow rates, it becomes imperative that solar pumps be redesigned for use in existing large farms.

Water use efficiencies (water pumped - water stored in the root zone) for the solar pump were quite low. No accurate measurement of water flow or of soil moisture was obtained but it was estimated that water use efficiency in the case of solar pumps was 60 per cent in plots where they were used to replace the Persian wheel and 30 to 40 per cent with larger plots and channel sizes.

Consideration has to be given to matching crop water requirement to pump output where the sole use of solar pumps is for irrigating farms. Since variations in the former are more marked than those in the latter, it might be expedient to choose a pump rating below peak-crop water requirements and storing surplus energy (or water) which can be used when required. This can be done via storage batteries, reservoirs or by using the water retention capacity at the root zone as a reservoir. The last option appears to be the most attractive since it involves no additional cost to the farmers.

B. Economic factors. Farmers can utilise groundwater for irrigation purposes through:

- using an animal-operated Persian wheel and an open well;
- purchasing water from either public or private tubewells;
- purchasing a diesel or electric pumpset.

The solar system cannot be compared as an alternative to canal irrigation or to electrically-operated tubewells as these are much cheaper sources given the present level of subsidisation. However, there is a good chance that solar pumps might replace the Persian wheel and, at lower price levels, they may compete with water purchased from diesel-operated tubewell owners.

Howes’ economic analysis which is based on theoretical calculations compares four solar pump systems including the type tested in Pakistan (system A) with the Persian wheel and the purchase of water from tubewells. Specifications for the four solar systems labelled A, B, C and D are given in Table 18.3.

Table 18.3: Characteristics of solar pumping systems

Source: M. Howes, The potential for small-scale solar-powered irrigation in Pakistan, Institute of Development Studies, Brighton. 1982.

Data were collected on the amount of water typically lifted in a year with the Persian wheel and from a 16 hp diesel-operated tubewell. These data were compared with theoretical figures obtained for solar pumps, assuming that all available radiation is used for pumping (see Table 18.3).

Using these figures and with some assumptions, calculations were made to determine the price level to which solar pumping would have to fall to replace the Persian wheel or water purchased from diesel-operated tubewells. To do this, farm sizes equal to the areas which can be irrigated by systems A to D in the peak farming season (rabi) at 4 m head have been assumed (see Table 18.4).

Costs of irrigation using all four solar systems were calculated and compared with reported costs of irrigation when the Persian wheel and diesel-operated tubewells were used. For these calculations, the following assumptions were made:

- a 15-year life of the solar system;

- once initial fixed capital investment has been made, the only cost for the solar systems is that of maintenance and repair for which a fixed sum was allowed annually;

- for the Persian wheel and diesel tubewells allowance was made for initial expenditure, scrap value, repair and maintenance, operators’ wages, opportunity cost of grazing land (in the case of the animal-drawn Persian wheel) and fuel (in the case of the diesel tubewell);

- a 12 per cent discount rate.

Solar Systems and the Persian Wheel

Price reductions required in the solar systems in order to displace the Persian wheel are shown in Table 18.5. The upper part of the table refers to the case where the Persian wheel and oxen are due for replacement immediately (are at the end of their economic lives). The lower part concerns cases where the farmers have just replaced their wheel and oxen (no resale value is taken into account). The table thus shows the level at which all Persian wheel users would be persuaded to immediately change to solar pumps. Both calculations are revised to take into account uncertainties associated with the use of a new technology. It is assumed that a further 20 per cent reduction in price levels would be required to account for such uncertainties.

From the table the following points emerge:

- none of the solar systems would appear cheap enough to compete with the Persian wheel even under the most favourable conditions where Persian wheel and oxen are to be replaced immediately and where no allowance is made for uncertainty;

- System A which was field-tested in this project appears to be unattractive as an immediate replacement of the Persian wheels unless large subsidies are provided;

- the most attractive solar system seems to be system B which could immediately be adopted at 12-30 per cent subsidy in cases where the Persian wheel is ready for immediate replacement.

Solar Systems and the Purchase of Diesel Tubewell Water

Table 18.6 shows the price reductions required in various solar systems to displace the purchase of diesel deep tubewell water (DTW). While the upper part assumes fixed diesel costs in real terms the lower part of the table takes into account an annual rise in real terms of 10 per cent in diesel cost.

Table 18.4. Comparison of performance of solar and other irrigation systems

Notes:

1. Theoretical figure based on the assumption that all available effective radiation will be used for pumping.

2. Figure based on recordings and interviews which suggested that Persian wheels would on average lift 2.5 litres per second × 4 metres and be used for 8 hours a day and 180 days a year.

3. Derived from M.A. Chaudhary, “Determination of cost of tubewell water and estimates of economic rent in canal irrigation”, in The Pakistan Development Review, Islamabad, Vol. 18, No. 2, 1978.

Table 18.5: Price reductions required in various solar systems in order to replace the Persian wheel (US$ and 12 per cent discount)

Notes:

1. Price to which solar pumping systems would have to fall to replace the Persian wheel.

2. Level of subsidy which would be required to justify immediate switch to solar pumping.

3. Figures in brackets indicate percentages.

Source: M. Howes: The potential for small-scale solar-powered irrigation in Pakistan, op. cit.

Table 18.6: Price reductions required in various solar systems to replace the purchase of diesel DTW water (US$ and 12 per cent discount)

Notes:

1. Price to which solar pumping systems would have to fall to replace the Persian wheel.

2. Level of subsidy which would be required to justify immediate switch to solar pumping.

3. Figures in brackets indicate percentages.

Source: M. Howes: The potential for small-scale solar-powered irrigation in Pakistan, op. cit.

The calculations did not take into account the following:

- the low discharge rate of solar pumps might result in higher labour costs;

- lower discharge rate for solar systems as compared to DTW irrigation might result in higher transmission losses in the case of the former;

- tubewell water purchasers who wish to switch to solar pumping would need to build an open well, the cost of which is not included in the calculations.

From Table 18.6 the following can be concluded:

- solar pumping could be attractive if diesel prices increase in real terms by 10 per cent for system B;

- with small subsidies (1 to 29 per cent), systems B, C, and D could immediately replace purchased water from DTW owners;

- if diesel prices do not increase in real terms, the cost of solar systems would have to be reduced even further below the level where they can compete with the Persian wheel and the tubewells.

V. CONCLUSIONS

At this stage, it would be difficult to come to a firm conclusion about the viability of solar pumps in Pakistan. It seems evident however that price reductions would be required in order that they can compete with existing irrigation systems. In the case where no other form of lift irrigation is available and solar pumps are to replace the Persian wheel, relatively modest reduction in prices would enable the more efficient solar photovoltaic pumps to be adopted. Where diesel deep tubewells exist and diesel prices remain constant in real terms, a considerably larger price decrease in solar systems would be required before adoption is expected to take place.

Several technical problems in solar water pumping would have to be addressed in the design of solar pumps. In particular, pumping units should be designed to operate under climatic conditions expected in the field. System efficiency and reliability would have to be improved.

Solar pumping technology has the potential to benefit small farmers judging from the low discharge rate of such pumps.

Utilisation of the existing solar photovoltaic pumps in Pakistan might involve a change in the irrigation practices (particularly as related to system layout) to enable furrow and/or trickle irrigation.

This project has demonstrated the feasibility of irrigation by means of pumps powered by electricity which is generated by the photovoltaic cells. Such a method of water pumping in general and irrigation in particular, would prove invaluable in areas where no alternative exists for irrigation and where diesel is unavailable, scarce or expensive.

NOTES AND REFERENCES

1. Material for this chapter is drawn mainly from two publications: (i) M. Howes: The potential for small-scale solar-powered irrigation in Pakistan, Institute of Development Studies, Brighton, 1982 and (ii) R.G. Pallett: Solar-powered irrigation pumps in Pakistan, Hydraulics Research Station Ltd, Wallingford, 1982.

2. Fifty-eight per cent of farms in Pakistan have an effective land holding acreage of 0 to 2 hectares (See Howes, ibid.).

3. R.G. Pallett: Solar-powered irrigation pumps in Pakistan, op. cit.

4. An improved variety of the pumpset has since been produced. This version is of higher efficiency and incorporates improved electric circuitry and a thermal overload switch.

5. System efficiency drops 4.44 per cent per degree Centigrade rise in temperature above rated value.

6. This is true also for farms irrigated by Persian wheels.