Free calculator · Rainfall

Design storm hyetograph generator

A design storm hyetograph spreads a design rainfall depth over time so a rainfall-runoff model can turn it into a flood hydrograph. The depth comes from the intensity-duration-frequency (IDF) curve for the return period; the temporal pattern decides when it falls, and for the same depth it can change the peak intensity by a factor of three. Studio's design storm generator builds hyetographs from SCS Type I to III, Huff, Chicago, alternating block and national patterns, and exports them to SWMM, HEC-HMS and InfoWorks ICM.

Design storm hyetograph generator

Pattern

Blocks from the IDF curve: the largest at mid-storm, the rest alternately right and left.

Storm
mm

For the return period and duration, from the IDF or DDF data.

h

90 min

min

Should divide D exactly.

IDF curve (Sherman)
mm/h
min

i = a / (t + b)c, with i in mm/h and t in minutes. Use the coefficients for the return period, and keep D within the durations they were fitted to.

Peak intensity
158.6mm/h
Depth in peak step
26.43mm
Peak step
40–50min

Hyetograph

020406080100120140160180010203040506070800102030405060708090Time (min)Intensity (mm/h)Cumulative depth (mm)
Intensity in each step (left axis)Cumulative depth (right axis)
Storm summary
Total depth, P71.15mm
Duration, D90 min in 9 steps of 10 min1.5h
Mean intensity, P/D47.43mm/h
Peak intensity158.6mm/h
Peak / mean intensity3.34
Depth in the peak step37.2 % of P26.43mm
Hyetograph centroid46.4min
IDF depth for DFrom the Sherman coefficients71.15mm

Rainfall increments

Depth, intensity and cumulative depth in each time step
Step (min)Depth (mm)Intensity (mm/h)Cumulative (mm)%
0–102.7116.22.713.8
10–203.4420.66.148.6
20–304.8028.810.9415.4
30–408.1548.919.0926.8
40–5026.43158.645.5364.0
50–6012.5775.458.1081.7
60–706.0336.264.1390.1
70–804.0024.068.1395.7
80–903.0218.171.15100.0

Keep this calculation

Studio accounts open soon. Join the waiting list and we'll e-mail you when you can save it in the full Design Storm Generator.

  • Save to a project
  • 390+ storm patterns, incl. SCS-SA
  • Loss models for rainfall excess
  • SWMM, HEC-HMS and ICM exports
  • Word and Excel report

Runs the same calculation engine as Delta Hydro Studio. Results are computed in your browser; nothing you type leaves it until you choose to save.

Using the calculator above

The calculator opens on the worked example below: a 90-minute alternating block storm at a 10-minute step from the Sherman coefficients a=1500a = 1500, b=10b = 10 and c=0.75c = 0.75, with a total depth of 71.15 mm.

  • Temporal pattern sets when the rain falls: the four SCS 24-hour types, the Huff quartiles, a triangle, a uniform block, or a storm built from the IDF curve (alternating block centred, front-loaded or back-loaded, and the Chicago storm with r=0.4r = 0.4).
  • Total depth, P is the design rainfall for the duration and return period.
  • Duration, D is in hours. Use 24 h for the SCS types.
  • Time step, Δt is in minutes and must divide the duration exactly.
  • a, b and c appear only for the IDF patterns, with ii in mm/h and tt in minutes. When P differs from the IDF depth for the duration, the calculator says so and offers the IDF depth.

The results give the peak intensity, the depth in the peak step, the hyetograph with its cumulative depth, and the depth in every step. Switch to US customary units to work in inches.

How a design storm hyetograph is built

Every design storm combines a total depth PP for the chosen duration and return period with a temporal pattern. The pattern comes in one of three forms:

Pattern typeHow the hyetograph is madeExamples
Cumulative mass curveA dimensionless curve F(t/D)F(t/D) is scaled to the depth and durationSCS Types I, IA, II, III; SCS-SA Types 1 to 4; Huff quartiles; Florida DOT zones
IDF-basedIntensities come from the site’s IDF curve, so the shape changes with return periodChicago, alternating block, nested storm
Simple shapesA fixed geometric form holds the depthUniform block, triangular, trapezoidal

For a mass curve, the depth in time step ii is the difference in the cumulative curve across the step:

ΔPi=[F ⁣(ti+1D)−F ⁣(tiD)]P\Delta P_i = \left[F\!\left(\frac{t_{i+1}}{D}\right) - F\!\left(\frac{t_i}{D}\right)\right] P
SymbolMeaningUnits
PPTotal storm depth for the duration and return periodmm or in
DDStorm durationh
tit_iStart of time step iimin
FFCumulative fraction of the storm depth, 0 to 1dimensionless
ΔPi\Delta P_iDepth in time step ii; divide by the step length for intensitymm

SCS and SCS-SA distributions

The four NRCS distributions in TR-55 are 24-hour mass curves for different climates of the United States. Type II covers most of the continental US and has the sharpest central burst, with 38 % of the day’s rain between hours 11.5 and 12. Type III, for the Gulf of Mexico and Atlantic coasts, peaks at the same time with a broader burst. Types I and IA, for the Pacific coast, peak earlier and lower, and Type IA is the flattest of the four.

These curves are defined for a 24-hour storm, so use them at 24 hours. The generator scales whatever curve you pick to the duration you enter, which means a one-hour “Type II” storm is the whole 24-hour pattern compressed into an hour, not the most intense hour of a 24-hour storm.

South Africa has its own adaptation. The SCS-SA Types 1 to 4 of Schulze, Schmidt and Smithers (2004), adopted in the SANRAL Drainage Manual, run from the gentlest pattern in coastal and orographic regions (Type 1) to the sharpest in the interior convective regions (Type 4). Their central burst holds 8.4 % (Type 1) to 21 % (Type 4) of the storm in a single short interval at the midpoint. The generator centres its time steps on that burst so it falls in one step rather than being split across two, which would understate the peak.

Huff (1967) classified Illinois storms by the quartile of the storm that holds the heaviest rain. First-quartile curves suit short convective storms; fourth-quartile curves suit long frontal ones.

IDF-based storms: Chicago and alternating block

Both methods need the site’s IDF relationship, usually as Sherman coefficients:

i=a(t+b)ci = \frac{a}{(t + b)^{c}}

with ii the intensity (mm/h), tt the duration (min), and aa, bb, cc fitted to the return period. Keep the storm within the range of durations the coefficients were fitted to.

Alternating block. The depth for jj time steps is Pj=i(jΔt) jΔt/60P_j = i(j\Delta t)\, j\Delta t / 60. The increments Pj−Pj−1P_j - P_{j-1} are sorted, the largest is placed at the centre, and the rest alternate right and left. Studio also offers front-loaded and back-loaded arrangements.

Chicago storm. With the peak at tp=rDt_p = rD, the intensity at a time tbt_b before the peak is

i(tb)=a[(1−c)tbr+b](tbr+b)1+ci(t_b) = \frac{a\left[(1-c)\dfrac{t_b}{r} + b\right]}{\left(\dfrac{t_b}{r} + b\right)^{1+c}}

and after the peak the same expression applies with ta/(1−r)t_a/(1-r) in place of tb/rt_b/r.

Worked example: 90-minute alternating block storm

A site’s 1:50-year IDF curve is described by a=1500a = 1500, b=10b = 10, c=0.75c = 0.75 (illustrative values; use the coefficients for your own site). The catchment’s Tc is about 45 minutes, so a 90-minute storm at a 10-minute step is tested. The IDF depths are 26.43 mm for 10 minutes, 39.01 mm for 20, 47.15 mm for 30, 61.98 mm for 60 and 71.15 mm for 90, so the storm depth is 71.15 mm.

The generator returns:

Block (min)0–1010–2020–3030–4040–5050–6060–7070–8080–90
Depth (mm)2.713.444.808.1526.4312.576.034.003.02
Intensity (mm/h)16.220.628.848.9158.675.436.224.018.1
Cumulative (mm)2.76.110.919.145.558.164.168.171.2

Check it against the IDF curve. The peak block, 158.6 mm/h, is the 10-minute IDF intensity. The two blocks from 40 to 60 minutes hold 39.01 mm, the 20-minute IDF depth, and the three blocks from 30 to 60 minutes hold 47.15 mm, the 30-minute depth. The nine blocks add up to the 90-minute depth, so nothing is rescaled.

The same 71.15 mm spread as a triangle peaks at 93.7 mm/h, and as a uniform block at 47.4 mm/h. Choosing the pattern changes the peak intensity by a factor of 3.3 at a fixed depth, which is why the pattern needs the same justification in a design report as the depth.

Choosing the duration and time step

  • Duration. At least Tc, from the time of concentration calculator, and tested either side to find the critical case.
  • Time step. About 5 minutes for storms up to 2 hours, 10 minutes for 2 to 6 hours, and 15 to 30 minutes for 6 to 24 hours. A step of Tc/5 to Tc/10 is a good start, and it should divide the duration exactly.
  • Depth. From the site’s IDF or depth-duration-frequency data, via design rainfall or a national atlas such as NOAA Atlas 14. The pattern only distributes the depth; a wrong depth spoils every result.
  • Area. For larger catchments, reduce the point depth with an areal reduction factor before building the storm.

Pattern library

The generator holds 391 named patterns for 135 countries. 171 have their own distribution data; the other 220 are national, regional or climate-change names that resolve to one of those base patterns. They include the NRCS SCS types, Huff quartiles, Florida DOT zones 1 to 5, NOAA Atlas 14 temporal, Denver UDFCD, USACE Standard Project Storm, the UK FSR profile, German DWA-A 531 and Euler types, the Desbordes double triangle, Australian Rainfall and Runoff patterns (ARR87 and ARR 2019), the SCS-SA Types 1 to 4, Simons (1976) and the South African Huff curves. Search by country, agency or tag.

Limitations

  • A design storm is a construct. The flood from a 1:50-year storm is not necessarily the 1:50-year flood: antecedent wetness and losses change the result.
  • Uniform rain. The hyetograph is a point pattern applied evenly over the catchment. Large catchments need an areal reduction factor, and very large ones need spatial rainfall.
  • Local patterns first. Patterns are fitted to the storms of one climate. Use the regional pattern your standard names.
  • IDF range. Sherman coefficients are valid only over the durations they were fitted to.

What a free Studio account adds

  • The whole pattern library: all 391 patterns, including the SCS-SA types, searchable by country, agency or tag.
  • Several return periods at once from a saved rainfall analysis, computed with one pattern and plotted together.
  • Loss models for rainfall excess: initial and continuing loss, proportional, SCS curve number, Green-Ampt and Horton.
  • Linked inputs: the depth and Sherman coefficients from a saved rainfall analysis, an IDF depth-duration table for the alternating block, and the duration from a saved Tc.
  • Exports to CSV, EPA SWMM (.dat), XP-SWMM, a complete SWMM .inp template, HEC-HMS and InfoWorks ICM.
  • Saving and reports: save each storm to a project and export Word and Excel reports.

For a peak discharge without building the hydrograph, the TR-55 calculator uses the same SCS Type I to III distributions.

FAQ

Common questions

What is the SCS Type II rainfall distribution?
It is the NRCS 24-hour design storm for most of the continental United States, published in TR-55. Rain builds slowly, then an intense burst falls at the middle of the day: the cumulative fraction rises from 0.283 at hour 11.5 to 0.663 at hour 12, so 38 % of the 24-hour depth falls in that half hour. NRCS built it from the TP-40 rainfall frequency maps so that the depths for every duration from 5 minutes to 24 hours are nested around that central peak; the largest 1-hour depth is 45.4 % of the 24-hour depth.
How does the alternating block method work?
Read the IDF depth for durations of one, two, three and more time steps up to the storm duration. The differences between successive depths are the blocks. Put the largest block in the middle of the storm, then place the rest alternately right and left in decreasing order. Every window centred on the peak then holds the IDF depth for its duration, so the storm is consistent with the IDF curve at every duration.
What is a Chicago design storm?
Keifer and Chu (1957) built a continuous storm from the Sherman IDF equation, i = a / (t + b) to the power c. The intensity at each moment is set so that the depth in any window around the peak equals the IDF depth for that duration. The advance coefficient r, usually 0.3 to 0.5, places the peak at r times the storm duration, so values below 0.5 front-load the storm.
How long should a design storm be?
At least as long as the catchment's time of concentration, so the whole catchment contributes. Test durations either side of Tc to find the critical one: a longer storm carries more total depth but at a lower intensity, so it does not always give a higher peak. The SCS distributions are defined for 24 hours and should be used at that duration.
Which design storm pattern should I use?
The one your jurisdiction's design standard prescribes. In the US that is usually the SCS type mapped in TR-55 or a state distribution such as the Florida DOT zones; in South Africa the SANRAL Drainage Manual uses the SCS-SA Types 1 to 4; in Australia, Australian Rainfall and Runoff temporal patterns. A pattern borrowed from another climate can misrepresent where the rain falls in time.
Can the generator subtract losses to give effective rainfall?
Yes. Studio applies a loss model to the gross hyetograph: initial and continuing loss, a proportional loss, the SCS curve number, Green-Ampt or Horton infiltration. The result is the rainfall excess that a unit hydrograph or a runoff model turns into flow. Use the same loss model and parameters as the rest of the study so the runoff volume stays consistent.
Full documentation: Design Storm Generator
Every input, option and equation in the Studio tool, with its limitations.
Run it from your AI agent
This engine is open to Claude, ChatGPT, Cursor and other MCP clients as run_design_storm. The agent passes the inputs, the engine does the calculation, and the record is saved to your project for you to review.

Reference tables

References

  1. USDA Soil Conservation Service (1986) Urban Hydrology for Small Watersheds, Technical Release 55 (TR-55), 2nd ed. United States Department of Agriculture, Washington DC.
  2. USDA Natural Resources Conservation Service (2019) National Engineering Handbook, Part 630 Hydrology, Chapter 4: Storm Rainfall Depth and Distribution. 210-630-H, Amendment 88. Washington DC.
  3. Keifer, C.J. and Chu, H.H. (1957) Synthetic storm pattern for drainage design. Journal of the Hydraulics Division, ASCE, 83(4), 1-25.
  4. Huff, F.A. (1967) Time distribution of rainfall in heavy storms. Water Resources Research, 3(4), 1007-1019.
  5. Chow, V.T., Maidment, D.R. and Mays, L.W. (1988) Applied Hydrology. McGraw-Hill, New York.
  6. Pilgrim, D.H. and Cordery, I. (1975) Rainfall temporal patterns for design floods. Journal of the Hydraulics Division, ASCE, 101(1), 81-95.
  7. Schulze, R.E., Schmidt, E.J. and Smithers, J.C. (2004) Visual SCS-SA: User manual for the SCS-SA design flood estimation model. School of Bioresources Engineering and Environmental Hydrology, University of KwaZulu-Natal, Pietermaritzburg.
  8. SANRAL (2013) Drainage Manual, 6th ed. South African National Roads Agency SOC Ltd, Pretoria. Chapter 3: Hydrology.

Last reviewed by the Delta Hydro engineering team.

Keep the whole study in one place

Delta Hydro Studio puts this tool alongside catchment, rainfall, runoff and hydraulics tools, with projects, team sharing and Word and Excel reports. Run them by hand, or let your AI agent run them for you to review.