Tomato water demand changes from transplant establishment through canopy development, fruit set and late season. It also changes by location and date because atmospheric demand changes. The defensible planning question is therefore not “how many litres does a tomato plant need?” but “what is crop evapotranspiration for this stage and weather, how much effective rain contributes, and how much gross irrigation must this field apply?”
Start with reference ET, not a national average
Reference evapotranspiration, ETo, describes climatic evaporative demand for a standardized reference surface. It incorporates weather variables rather than crop identity. The crop coefficient, Kc, translates that demand for a particular crop and growth stage under defined conditions. FAO’s AQUASTAT documentation and Irrigation and Drainage Paper 56 both use ETc = Kc × ETo.
Obtain ETo from a credible station or calculation appropriate to the farm. Kenya has strong elevation and climate differences; a value copied from a distant county or a monthly global average can mislead. Record the station, period and units. If the source reports weekly or monthly totals, keep the time basis consistent throughout the calculation.
Next choose Kc for the actual growth stage. FAO 56’s typical table gives tomato values around 0.6 initially, 1.15 at midseason and 0.70–0.90 toward the end under its stated standard conditions. FAO notes that values require adjustment for climate, wetting frequency, crop height and management. Staked tomatoes with greater height may require a higher value. These coefficients are not Kenyan prescriptions; they are a documented starting method to refine with local expertise.
Why crop stage changes the pump brief
Early after transplanting, small plants cover little ground. Midseason canopy and fruit development can create the highest demand. Late-season demand changes as the crop matures. A pump sized only from an early-stage observation can fail at peak demand. Conversely, applying peak volume throughout the season can waste water, leach nutrients and create disease pressure.
For investment sizing, use a credible peak design day and also prepare a seasonal schedule. The pump and source must meet the peak duty, while valves, timing and storage control daily application. The system should allow shorter, more frequent applications where root depth, soil infiltration and emitter design justify them.
Worked one-acre screening example
Assume—not assert—that local ETo for a hot design day is 5.5 mm/day, the crop is at FAO’s typical midseason Kc of 1.15, effective rainfall for that day is zero, and measured drip application efficiency is 90%. The crop demand is 6.325 mm/day. Across one acre, that is about 25,596 net litres. Dividing by 0.90 gives about 28,440 gross litres to be pumped or withdrawn from storage.
| Step | Calculation | Result |
|---|---|---|
| Crop ET | 5.5 × 1.15 | 6.325 mm/day |
| Net field volume | 6.325 × 4,046.856 | 25,596 L/day |
| Gross at 90% | 25,596 ÷ 0.90 | 28,440 L/day |
| Six-hour average flow | 28,440 ÷ 6 | 4,740 L/hour |
This flow exceeds the 3,600 L/h headline maximum stated by Futurepump for its SF2 and is far above the SE1’s cited 1,600 L/h maximum. That is useful screening information. It does not establish that the example field always needs 28,440 L/day, nor that a particular alternative pump will meet it. Refine ETo, Kc, effective rainfall, area actually wetted, efficiency, scheduling and TDH; then compare curves.
Drip efficiency must be earned in the field
“Drip” is not automatically 90% efficient. Uniformity falls with incorrect pressure, excessive lateral length, elevation changes, blocked emitters, leaks, poor filtration and manufacturing variation. Measure discharge from representative emitters at the beginning, middle and end of laterals, including high and low parts of the field. Compare actual application with the design and repair causes of variation.
Emitter layout should match plant spacing, soil movement and root zone. Sandy soil tends to spread water differently from clay soil. A high hourly emitter rate can exceed infiltration and create deep percolation or runoff; too low a rate may make irrigation windows impractical. Filter selection must match the source and emitter passage. Include filter pressure loss in total dynamic head, especially in the dirty condition used for maintenance triggers.
Convert field volume into zones and runtime
Count emitters per zone and multiply by measured emitter flow. A zone with 1,000 emitters averaging 1.8 L/h applies 1,800 L/h. Dividing a 28,440 L gross target by that rate gives 15.8 zone-hours across the day—before accounting for how many zones can run simultaneously. This reveals whether the irrigation layout, storage and operating day fit together.
Do not use a runtime calculation to override soil observation. Monitor root-zone moisture, plant condition, drainage, rainfall and system output. Effective rainfall is not identical to measured rainfall: some runs off, evaporates or falls outside useful timing. FAO defines irrigation need as crop water need minus the part of rainfall available to the crop.
For pump screening, enter the refined peak millimetres per day and area in the ShambaPump calculator. Then carry the gross litres, pumping hours and TDH into the one-acre sizing workflow.
Method and limitations
The example demonstrates FAO’s coefficient method and a gross-efficiency adjustment. It is not a Kenya-wide irrigation schedule. It excludes soil-water balance, rooting depth, allowable depletion, salinity leaching, disease management, fertigation, real effective rainfall and measured distribution uniformity. Obtain local ETo and agronomic advice.
Frequently asked questions
How many litres per tomato plant per day?
A fixed figure is unreliable because spacing, canopy stage, weather, soil, emitter layout and rainfall vary. Calculate field ETc and verify distribution; per-plant volume can then be derived for management if plants and wetted area are known.
Should the pump be sized for average or peak demand?
Check the credible peak design duty so the system is not predictably short, then control lower seasonal applications through runtime, zones and storage. Source yield and permitted abstraction remain hard limits.