How this calculation works
Apply the entered rainfall loss and collection fraction, add harvest to storage, spill above capacity and only then serve that row’s demand. Compare whole-schedule totals and retain the first tank’s daily ledger.
Collected litres = max(0, rainfall mm − event flush mm) × roof m² × runoff fraction; storage then spills and supplies demandA worked example
The illustrated roof collects 720 L in the first event. The 500 L tank spills 220 L, then meets all 600 L of entered demand over three days and ends with 220 L. A smaller 300 L tank would leave 100 L unmet on the dry day.
Good to know
Bucket bookkeeping only, not drinking-water treatment, roof engineering or a forecast. Row timing matters; no leakage, evaporation or water-quality inference. Initial volume must fit every candidate.
Reference: Texas Water Development Board: rainwater harvesting guidance (opens in a new tab) ↗
One rainfall depth becomes a volume
One millimetre over one square metre is one litre before losses. The entered runoff fraction then represents the portion retained under your collection assumption.
First flush is entered as rainfall depth lost separately for every row. If several rows belong to one continuous event, decide whether this repeated-loss convention matches the intended data.
Daily ordering can change storage results
The model harvests, spills and then serves demand. A large storm can therefore overflow a full tank before the day’s use creates space.
A finer time series could produce different results when use and rain occur throughout a day. Do not interpret a daily bucket result as an exact subdaily hydraulic simulation.
Compare candidates without adding incompatible rows
Candidate-total rows summarize complete schedules. The following daily rows describe only the first entered tank, so they must not be added to those totals again.
A larger tank can reduce modeled overflow or shortage under the same schedule, but the page does not choose an economic optimum or certify collection and treatment equipment.
Use the result with the next measurement
The crop water ET0 depletion schedule calculator tracks root-zone depletion, not roof-tank storage. Its entered effective rain and crop coefficients should not be copied directly from a roof runoff model.
Bucket bookkeeping only, not drinking-water treatment, roof engineering or a forecast. Row timing matters; no leakage, evaporation or water-quality inference. Initial volume must fit every candidate.
A few questions, answered
Is rainfall a forecast?
No, all rows are entered observations or scenarios.
Does demand happen before overflow?
No, harvest and overflow happen first.
Can I sum daily and candidate-total records?
No, they are distinct record types.