Skip to content

Time patterns

A time pattern is a list of multipliers applied to a base value. On its own it does nothing. It only has an effect when something references it, and in SWMM that something is almost always a dry-weather flow entry or a direct inflow.

Open them at Model data ▸ Patterns.

TypeFactorsApplies to
HOURLY24Each hour of a weekday
WEEKEND24Each hour of a Saturday or Sunday
DAILY7Each day of the week, Sunday first
MONTHLY12Each month, January first

The classic is an HOURLY sanitary pattern: the morning peak as a town gets up, a dip overnight, and something near 1.0 through the middle of the day. That shape is what makes a dry-weather model look like a real sewer instead of a flat line.

This is the mistake worth naming. The factors multiply the base flow, so if their mean is 1.3, your average daily flow is 30% higher than the number you typed into the DWF box, and every calibration afterwards is chasing a discrepancy you created.

An hourly pattern that runs 0.4 at night and 1.6 at the morning peak is fine, as long as the 24 values average out near 1.0. The editor shows the factors as a bar chart so an obviously lopsided pattern is visible at a glance.

A DWF entry can carry up to four patterns, one of each type, and SWMM multiplies them together. Base flow x hourly x daily x monthly lets you express “the morning peak, lower at weekends, higher in winter” without inventing a single 8,760-value series.

That multiplication is also how patterns go wrong quietly: a monthly pattern averaging 1.2 combined with an hourly pattern averaging 1.1 gives a mean multiplier of 1.32, and neither one looks unreasonable on its own.

  • Dry-weather flow · what patterns are usually attached to
  • Curves · the other lookup object, which is a relationship rather than a schedule