Home / CPHEEO / Chapter 2
CHAPTER 2

Rainfall Analysis and IDF Curves

Rainfall, IDF & Design Storm

Covers acquisition + analysis of rainfall data for design — IMD records, station selection, frequency analysis (Gumbel, log-Pearson III), Intensity-Duration-Frequency (IDF) curve construction, design storm hyetograph (uniform, triangular, Chicago method), areal reduction factor for catchments > 25 km², and city-specific rainfall intensities for major Indian cities.

🌧 Rainfall & IDF CurvesManual on Storm Water Drainage Systems1st Edition (2019), with AMRUT 2.0 + Smart Cities Mission updates referenced

Key formulas

  • IDF (Sherman) i = a / (t + b)^n where a, b, n are city + return period constants
  • Gumbel: x_T = mean + K_T × σ; K_T = -(√6/π)[0.5772 + ln(ln(T/(T-1)))]
  • Areal Reduction Factor (ARF) = 1 − 0.01 × (A in km²)^0.5 for A > 25 km² (USWB)
  • Design rainfall depth P = i × tc (mm), where i is intensity at duration = tc
  • Climate-adjusted i' = i × (1 + Δ), Δ = 0.10 to 0.30
  • Chicago method peak time t_p = r × td (r = 0.3-0.5 typical, td = storm duration)

Key values & thresholds

min rainfall record for design
25 years (preferred 30+)
imd data source
Pune IITM website + IMD Hydromet circle offices
delhi 60min 2yr intensity
45-55 mm/hr
delhi 60min 25yr intensity
75-90 mm/hr
mumbai 60min 2yr intensity
60-80 mm/hr
mumbai 60min 25yr intensity
120-150 mm/hr
chennai 60min 2yr intensity
55-70 mm/hr
chennai 60min 25yr intensity
100-130 mm/hr
bangalore 60min 2yr intensity
40-50 mm/hr
kolkata 60min 2yr intensity
55-70 mm/hr
areal reduction threshold
25 km² catchment

Clause-level requirements

  • Use minimum 25 years of rainfall record from a station within 10 km of the project, or interpolate from regional records.
  • For catchments < 25 km², no areal reduction factor — point rainfall = areal rainfall.
  • Design storm duration shall not be less than the time of concentration of the catchment.
  • Frequency analysis must use either Gumbel (extreme value type 1) or log-Pearson III; choose by goodness-of-fit (chi-square or KS test).
  • Climate-uplift factor of 10-30 % shall be applied to historical IDF before design unless local downscaled climate model data is available.
  • For drainage districts > 100 ha, use Chicago hyetograph (front-loaded triangular) for peak design, not uniform rainfall.

Practitioner notes — what goes wrong in the field

  • IMD daily rainfall is freely available; sub-daily (15-min, 60-min) often requires paid request to IMD Hydromet circle — budget ₹50K-2L per project.
  • For Mumbai, Chennai, Hyderabad, Pune — sub-daily data is also available via state government weather services + airport AWS records.
  • If station record is < 15 years, use regional IDF (state-level) + sensitivity check; do NOT design to short-record extrapolations.
  • Chicago hyetograph (asymmetric peak ≈ 1/3 into storm duration) is closer to real Indian convective storms than uniform rainfall.
  • Apply ARF for large catchments (> 25 km²) — point intensity overestimates the average areal intensity. Common mistake: using point IDF for whole municipal area.
  • Mumbai July 26, 2005 was 944 mm/24 hr — a > 100-year event. Many post-2005 designs explicitly include this as a design check.
  • Chennai Dec 2015 floods exceeded 100-year IDF — climate-uplift of 30 % is now standard for Chennai's drainage projects.
  • For peri-urban areas with poor data, use IMD's gridded daily rainfall (0.25° resolution) + downscale temporally using regional IDF ratios.
  • Document chosen IDF + adjustment factor in design report — auditors challenge designs with no IDF source.
  • Smart Cities + AMRUT projects increasingly fund automatic weather stations — push for AWS + telemetry as project deliverable.

FAQs

Where do I get IDF curves for my Indian city?
IMD publishes 'Atlas of Storm Rainfall' for major cities (free, dated 1933 with periodic updates). For sub-daily intensities, request IMD Hydromet circle (paid). Many state PWDs publish design rainfall — Mumbai has BMC Storm Water Drains Manual, Chennai has CMDA storm water guidelines. Smart Cities projects often commission new IDF studies.
How many years of rainfall data do I need?
Minimum 25 years for frequency analysis; 30+ preferred for stable estimates. Less than 15 years should not be used for extrapolation beyond 10-year return periods. Use regional IDF as fallback.
What's a Chicago hyetograph and why use it?
An asymmetric design hyetograph that front-loads peak intensity (peak around 1/3 into the storm duration) — better matches real Indian convective storms than uniform rainfall. Used in hydraulic models (SWMM, MIKE URBAN) for peak flow estimation in catchments > 100 ha.
Do I need to apply climate change uplift?
Yes — CPHEEO 2019 + AMRUT 2.0 strongly recommend 10-30 % uplift on historical design rainfall. Coastal + cyclone-prone cities (Mumbai, Chennai, Bhubaneswar, Vizag) use 25-30 %; inland cities use 10-15 %. Critical infrastructure designs use the higher end.
What if my catchment crosses multiple rainfall stations?
Use Thiessen polygons or arithmetic average of stations to compute areal rainfall. For catchments > 25 km², apply Areal Reduction Factor to convert point intensity → areal intensity per USWB equation.

Calculator

Design Rainfall Intensity (Sherman IDF + Climate Uplift)

Compute design rainfall intensity for a given duration + return period using Sherman's IDF equation i = a/(t+b)^n. Apply climate uplift per CPHEEO 2019 / AMRUT 2.0 (10–30 %). Constants a, b, n are city-specific — request from IMD Hydromet circle or use city PWD storm water manual.

Inputs
Sherman constant 'a'
Coefficient — typical 5000–9000 for Indian cities (per return period)
Sherman constant 'b'min
Time offset — typical 15–35 min
Sherman exponent 'n'
Decay exponent — typical 0.85–1.15
Storm duration (= time of concentration)min
Climate uplift%
10 % inland, 25–30 % coastal/cyclone-prone
Outputs
Design intensity (historical IDF)
118.2mm/hr
i = a / (t + b)^n
Climate-adjusted intensity
141.8mm/hr
i × (1 + uplift/100)
Total rainfall depth in duration
70.9mm
i × t / 60
CPHEEO Reference Values
Mumbai 60-min 25-yr (typical)120 – 150 mm/hr
Chennai 60-min 25-yr100 – 130 mm/hr
Delhi 60-min 25-yr75 – 90 mm/hr
Bangalore 60-min 25-yr55 – 70 mm/hr
Climate uplift inland10 – 15 %
Climate uplift coastal/cyclone25 – 30 %
Download the Excel version to keep a local copy with live formulas — change inputs in the sheet and outputs recompute automatically.

Cross-references

IMD Atlas of Storm Rainfall (1933, updated periodically)CWC Empirical Flood FormulaeIS 5477 (river discharge methods)CWPRS Pune (rainfall analysis services)USDA TR-55 (regional reference)BMTPC Vulnerability Atlas (cyclone/flood overlay)

Tags

IDF curvedesign stormGumbel distributionrainfall intensity IndiaIMD dataareal reductionChicago hyetographclimate uplift rainfallSherman equation

Engineer's notes

Rainfall analysis is where the entire drainage design starts — get the IDF wrong and every downstream pipe size, inlet spacing, and detention pond is wrong. Yet this is exactly where most Indian municipal drainage designs cut corners.

The classical approach: Take ≥ 25 years of station rainfall, fit a Gumbel or log-Pearson III distribution to the annual maximum series, derive intensity for design return period, build an IDF curve via Sherman's equation. Use it to compute design intensity at duration = time of concentration. Standard textbook procedure.

The practical reality in India: Quality rainfall data is patchy. IMD has good daily data; sub-daily is paid + bureaucratic. Many cities use IDF curves derived in the 1960s-80s that no longer reflect current climate. Mumbai's July 2005 event (944 mm in 24 hours) and Chennai's December 2015 floods (over 1000 mm in a week) demonstrated that historical IDF curves dramatically under-predict current extremes.

Climate uplift is now standard professional practice. CPHEEO 2019 recommends 10-30 % uplift; coastal cities push the upper end. Smart Cities Mission projects fund new sub-daily rainfall analysis using modern records — leverage these where available.

Areal reduction matters for catchments > 25 km² — point intensity at the rain gauge always overstates the average intensity over a large area. Skip ARF and you oversize the trunk drain.

Design hyetograph choice matters for hydraulic modeling — Chicago method (peak at 1/3 of storm duration) better matches Indian convective storms than uniform rainfall. For simple Rational-method designs, single intensity at tc duration suffices.

Where this chapter sits: rainfall is the boundary condition for the whole drainage system. The IDF + design storm decisions made here propagate through every chapter that follows.

Download full manual from MoHUA →
Manual on Storm Water Drainage Systems · 1st Edition (2019), with AMRUT 2.0 + Smart Cities Mission updates referenced · Central Public Health and Environmental Engineering Organisation (CPHEEO), Ministry of Housing and Urban Affairs, Government of India.
InfraLens provides chapter summaries for search — full manual is the authoritative reference.