Hydrograph Generator

Open Hydrograph Generator

Generate a complete flow-versus-time hydrograph from a peak discharge, a design storm, or a direct unit hydrograph convolution. This guide covers the theory of synthetic unit hydrographs, the six supported methods (SCS dimensionless, SCS triangular, Snyder, Clark/ModClark, gamma, and triangular), input parameters, timestep selection, loss and baseflow treatment, validation checks, export formats, and a complete worked example for a 20 km² catchment.

Overview

A hydrograph is a plot of discharge versus time at a point in a river or drainage system. For design, we are interested in the hydrograph produced by a specified storm event — the design hydrograph — because it contains three pieces of information that a single peak-flow calculation cannot provide:

• Volume — the total runoff volume under the curve, required for detention basin and reservoir storage sizing.
• Shape — the rising limb, peak, and recession, which governs the time available for evacuation, the rate of stage rise downstream, and the attenuation performance of any storage element.
• Timing — the time to peak and time-to-centroid, required for unsteady hydraulic modelling, flood-wave routing, and coordinated operation of multi-reservoir systems.

The Hydrograph Generator produces this curve from either a peak discharge (combined with catchment properties to infer the shape) or from a full effective rainfall hyetograph convolved with a unit hydrograph. It is the essential bridge between peak-flow methods (Rational, RMF, Flood Frequency Analysis, QRT, TR-55) and the downstream hydraulic analyses that depend on them — reservoir routing, channel unsteady flow, floodplain inundation, and dam-break studies.

Peak-flow methods vs hydrograph methods

The Rational Method gives you a single number — the peak discharge. It cannot tell you how long the peak lasts, how much volume passes the outlet, or how the flood develops in time. The Hydrograph Generator takes that peak and, using a synthetic unit hydrograph appropriate to the catchment, expands it into a complete discharge-versus-time curve that downstream analyses can consume directly.

Hydrograph shape and terminology

A typical single-peaked storm hydrograph consists of three phases:

• Rising limb (concentration curve) — discharge rises from baseflow as overland flow, interflow, and eventually channel flow reach the outlet. The slope of the rising limb is controlled by the catchment’s time of concentration and the rainfall distribution in time.
• Peak — the maximum discharge. For a unit-hydrograph-derived hydrograph, the peak corresponds to the instant when the effective rainfall contribution from the entire catchment, routed through the basin response, is at its maximum.
• Recession limb (depletion curve) — discharge falls back towards baseflow. In natural catchments the recession is controlled by drainage from channel storage, then interflow depletion, then groundwater baseflow release.

Key hydrograph parameters:

• $Q_p$ — peak discharge (m³/s).
• $t_p$ — time to peak, measured from the start of effective rainfall.
• $T_p$ — rise time, measured from the start of the unit rainfall pulse to the peak ($T_p \approx t_p$ for a short pulse).
• $T_b$ — base time, the total duration over which direct runoff is non-zero.
• $t_L$ — lag time, from centroid of effective rainfall to the peak. The NRCS relation is $t_L = 0.6 \, T_c$, where $T_c$ is the time of concentration.
• PRF — Peak Rate Factor for SCS methods; a shape parameter linking peak, volume, and time.

Unit hydrograph theory

A unit hydrograph (UH) is the direct-runoff hydrograph resulting from one unit depth (1 mm in SI, 1 inch in US customary) of effective rainfall applied uniformly over the catchment over a specified duration $D$. The UH is a fundamental catchment response function: once determined, it can be convolved with any effective rainfall hyetograph to produce the corresponding direct-runoff hydrograph.

Core assumptions

The unit hydrograph concept rests on four Sherman (1932) assumptions:

1. Time invariance — the catchment response does not change between storms.
2. Linearity — doubling the effective rainfall doubles the runoff at every time step.
3. Superposition — the response to sequential rainfall pulses is the linear sum of the responses to each pulse individually.
4. Uniform spatial distribution — effective rainfall is applied uniformly over the catchment.

In practice these are approximations — catchments exhibit non-linearity (high-intensity storms produce proportionally higher peaks than UH linearity predicts), and spatial uniformity breaks down for catchments larger than the storm cell size. Nevertheless, the UH concept remains the dominant operational method for deriving design hydrographs worldwide.

Convolution

Given a unit hydrograph $U_j$ (ordinates for $j = 1 \ldots N_U$) and an effective rainfall hyetograph $P_i$ (pulses for $i = 1 \ldots N_P$), the direct runoff $Q_k$ at time step $k$ is obtained by discrete convolution:

$$Q_k \;=\; \sum_{i=1}^{\min(k,\, N_P)} P_i \cdot U_{k-i+1}$$

Discrete convolution of effective rainfall with a unit hydrograph

This is a sliding-weighted-sum operation: each rainfall pulse generates a scaled copy of the unit hydrograph starting at the pulse time, and the total runoff at any time is the sum of all contributing scaled UH ordinates. The total number of output ordinates is $N_Q = N_P + N_U - 1$.

When to convolve vs scale

If you only have a peak discharge and want a hydrograph consistent with it, the tool scales a unit hydrograph to produce a single-pulse response. If you have a design storm hyetograph (from the Design Storm tool) and a loss model, convolution produces a more realistic multi-peak hydrograph that preserves the temporal pattern of the storm. Convolution is preferred whenever the storm duration is much longer than the catchment time of concentration.

Supported methods

The tool implements six synthetic unit hydrograph methods. Choice of method depends on (a) what catchment parameters you have, (b) the geography of the catchment, and (c) the consistency required with peak-flow methods already applied.

1. SCS Dimensionless Unit Hydrograph

The NRCS/SCS method — originally documented in USDA National Engineering Handbook Part 630, Chapter 16 (1972, revised 2007) — is the workhorse synthetic UH method worldwide. A fixed dimensionless curve (Table 16-1, 51 points) expresses $Q/Q_p$ as a function of $t/T_p$. Given a catchment, you compute $T_p$ and $Q_p$; the dimensionless curve then scales to the dimensional UH.

$$T_p \;=\; \frac{D}{2} \;+\; t_L \;=\; \frac{D}{2} \;+\; 0.6 \, T_c$$

SCS time to peak (SI)

$$Q_p \;=\; \frac{\mathrm{PRF} \cdot A}{T_p}$$

SCS peak discharge (SI)

Where $D$ = rainfall pulse duration (hr), $T_c$ = time of concentration (hr), $A$ = area (km²), and $\mathrm{PRF}$ is the Peak Rate Factor in metric-consistent units (default 484 in US customary; 484/645.33 ≈ 0.75 in SI m³/s per km² per hr per mm). The full conversion is handled internally — enter $\mathrm{PRF}$ as the familiar 484.

Peak Rate Factor (PRF) controls the hydrograph shape. NEH-4 Table 16-5 and Appendix 16B give PRF values by physiography:

Physiographic setting	PRF
Very flat / swampy / coastal plains	100 – 300
Mixed flat to rolling (default)	484
Steep / mountainous	550 – 575

PRF is not just a knob

Lower PRF means a flatter, longer hydrograph with the same volume; higher PRF means a sharper peak with a shorter base. Do not adjust PRF to make results match an a-priori expectation — it is a physiographic parameter that should be set from catchment topography, channel density, and wetland extent. If you cannot justify a non-default value, leave PRF at 484.

2. SCS Triangular Unit Hydrograph

A simplified triangular approximation of the curvilinear SCS UH, preserving volume. The rising limb has duration $T_p$, the recession limb $1.67 \, T_p$, and the base time $T_b = 2.67 \, T_p$.

$$Q_p \;=\; \frac{2 \, V}{T_b} \;=\; \frac{0.75 \, A \cdot P_e}{T_p}$$

Triangular peak discharge — equivalent to SCS curvilinear volume

Where $V$ = total runoff volume (m³), $A$ = area (km²), $P_e$ = effective rainfall depth (mm). The 0.75 factor embeds both the 2/2.67 triangle ratio and the unit conversions.

Use the triangular form for quick hand checks, for teaching purposes, or when a single representative UH shape is adequate and the additional precision of the 51-point curvilinear table is not justified.

3. Snyder’s Synthetic Unit Hydrograph

Developed by Franklin Snyder in 1938 from 20 Appalachian catchments, Snyder’s method parameterises the UH in terms of two coefficients:

• $C_t$ — basin lag coefficient, 1.4 to 1.7 for mature Appalachian catchments; typically 1.35 to 1.65 for South African catchments.
• $C_p$ — peaking coefficient, 0.4 to 0.8.

Plus two geometric inputs: main-channel length $L$ (km) and distance from outlet to the catchment centroid along the main channel $L_{ca}$ (km).

$$t_L \;=\; C_t \cdot (L \cdot L_{ca})^{0.3}$$

Snyder basin lag time (hr)

$$Q_p \;=\; \frac{2.75 \, C_p \, A}{t_L}$$

Snyder standard-duration peak discharge (m^3/s)

The standard unit rainfall duration is $D_s = t_L / 5.5$. For an arbitrary duration $D$ the lag is adjusted: $t_L' = t_L + (D - D_s)/4$.

Width equations (width of hydrograph at 50% and 75% of peak) complete the shape:

$$W_f \;=\; \frac{K_w}{(Q_p / A)^{1.08}}$$

Snyder width at fraction f of peak (hr)

with $K_w = 5.87$ at $f = 0.5$ and $K_w = 3.35$ at $f = 0.75$.

When to prefer Snyder

Snyder’s method is most appropriate for medium-to-large catchments (50 to 5000 km²) with well-defined main channels where $L$ and $L_{ca}$ are meaningful. For very small urban or peri-urban catchments, the SCS methods are generally more suitable. Calibrate $C_t$ and $C_p$ against observed hydrographs where possible — regional values should be treated as starting points.

4. Clark (ModClark) IUH

The Clark instantaneous unit hydrograph (Clark 1945) couples a time-area histogram — representing translation of water through the catchment from remote to outlet — with a single linear reservoir representing basin storage attenuation. The linear reservoir routing equation:

$$\mathrm{IUH}_i \;=\; C_A \cdot I_i \;+\; C_B \cdot \mathrm{IUH}_{i-1}$$

Clark linear reservoir routing — two-parameter recursion

with coefficients

$$C_A \;=\; \frac{\Delta t / R}{1 + \Delta t / R}, \qquad C_B \;=\; \frac{1 - \Delta t / R}{1 + \Delta t / R}$$

Clark routing coefficients

where $R$ is the storage coefficient (hr) and $I_i$ is the time-area input ordinate at step $i$. The default $R \approx 0.7 \, T_c$ is a commonly-cited starting value; HEC-HMS documentation suggests $R / (T_c + R) = 0.5$ to $0.7$ for many US catchments.

The time-area relationship defaults to the USACE standard S-curve, but a user-defined distribution can be supplied for urbanised or irregularly-shaped catchments. The ModClark variant distributes the time-area histogram over a gridded representation of the basin so that spatially variable rainfall can be routed — useful for distributed radar-based precipitation inputs.

5. Gamma Distribution

A two-parameter gamma probability distribution fitted to the hydrograph ordinates (Aron and White, 1982):

$$\frac{Q(t)}{Q_p} \;=\; \left( \frac{t}{T_p} \right)^m \cdot \exp\!\left( m \cdot \left( 1 - \frac{t}{T_p} \right) \right)$$

Gamma unit hydrograph (NEH-4 Eq. 16-1 reformulation)

Where $m$ is the shape factor. The gamma form can reproduce any PRF from roughly 100 to 600 by varying $m$ — $m \approx 3.77$ corresponds to PRF = 484 (the standard SCS curvilinear UH).

The gamma hydrograph is attractive because it is smooth (infinitely differentiable), preserves volume exactly for any shape parameter, and is analytically tractable. It is the default in some European practice and in several machine-learning rainfall-runoff models that output UH parameters as regression targets.

6. Triangular Simple

A bare-bones triangle defined only by the rise time $T_r$ and a recession ratio $k$ (default 1.67). Neither $T_c$ nor area is required — the user supplies the peak $Q_p$ and the rise time directly.

Rising limb: linear from 0 to $Q_p$ over $T_r$. Falling limb: linear from $Q_p$ back to 0 over $k \cdot T_r$. Total base: $T_b = (1 + k) \, T_r$. Volume: $V = 0.5 \, Q_p \, T_b$.

Use this method when only the peak and approximate timing are known — for example, when a QRT calculation gives you a peak but no catchment delineation, or for scoping-level detention basin sizing before a full hydrological model is available.

Input parameters

The tool accepts the following inputs, many of which can be linked directly from saved HydroDesign calculations:

Input	Methods using it	Source
Peak discharge $Q_p$	All (scaling) / derived from convolution	Rational Method, RMF, FFA, QRT, TR-55
Catchment area $A$ (km²)	SCS, Snyder, Clark, Gamma	Watershed Delineation
Time of concentration $T_c$	SCS, Clark (for $R$ default), Snyder lag	Tc Calculator
Longest watercourse $L$	Snyder	Watercourse Profile
$L_{ca}$ (outlet to centroid)	Snyder	Measured on map / GIS
Storm hyetograph	Convolution mode (any method)	Design Storm
Effective rainfall depth $P_e$	Single-pulse mode	SCS CN Method or φ-index
Baseflow $Q_b$	All	Gauged recession analysis / default zero

Input linking and provenance

When you select a linked calculation (for example, the peak discharge from a saved Rational Method result), the tool extracts the relevant value and stores the linked calculation ID alongside your hydrograph. This preserves the provenance chain: anyone opening your saved hydrograph can trace back to the exact Rational calculation and its inputs (catchment delineation, Tc method, runoff coefficient, IDF value). For regulatory submissions this is often required — preserving the audit trail is not merely convenient.

Timestep selection

The simulation timestep $\Delta t$ must be small enough to resolve the rising limb of the hydrograph without numerical smoothing. The widely-cited rule of thumb:

$$\Delta t \;\le\; \frac{T_p}{5} \quad \text{or equivalently} \quad \Delta t \;\le\; 0.25 \, T_r$$

Timestep rule (Ponce, 1989; HEC-HMS guidance)

Where $T_p$ is time to peak and $T_r$ is rise time. For a catchment with $T_c = 2$ hr and $T_p \approx 1.2$ hr, a timestep of 15 minutes is typically adequate; for $T_c = 30$ min and $T_p \approx 18$ min, use 3 to 5 minutes.

Time of concentration	Recommended $\Delta t$
< 15 min	1 min
15 – 60 min	2 – 5 min
1 – 3 hr	10 – 15 min
3 – 12 hr	15 – 30 min
> 12 hr	30 – 60 min

Too-coarse timestep is non-conservative

A coarse timestep artificially truncates the peak because the rising limb is resolved with too few ordinates. The peak error is typically 2 to 10 percent when $\Delta t = T_p / 5$, and can exceed 20 percent when $\Delta t > T_p / 2$. For detention basin sizing this is non-conservative because the inflow peak drives the required storage. Always verify with a sensitivity run at half the selected timestep.

Loss methods and effective rainfall

When running the tool in convolution mode with a storm hyetograph, you must separate total rainfall $P$ into an abstraction (losses — interception, depression storage, infiltration) and effective rainfall $P_e$ that becomes runoff. Three loss models are supported:

SCS Curve Number

The SCS-CN method (NEH-4 Chapter 10; see SCS Method) computes effective rainfall via:

$$P_e \;=\; \frac{(P - I_a)^2}{P - I_a + S} \qquad P > I_a$$

SCS-CN effective rainfall

with $S = 25400 / \mathrm{CN} - 254$ (mm) and $I_a = 0.2 \, S$ (default) or $0.05 \, S$ (modern research-based value).

The method is applied cumulatively to the storm hyetograph, so that the loss fraction decreases as the storm progresses and soils saturate — producing the characteristic late-storm intensification of effective rainfall.

Phi-index (constant loss rate)

A constant loss rate $\phi$ (mm/hr) is subtracted from each rainfall pulse:

$$P_{e,i} \;=\; \max(0, \; P_i - \phi \cdot \Delta t)$$

Phi-index effective rainfall

Simple, robust, and appropriate when a gauged event has been used to derive $\phi$ by volume balance. Less physically realistic than SCS-CN for long-duration storms.

Initial-and-constant

A fixed initial abstraction $I_a$ (mm) is removed first, then a constant loss rate $f_c$ (mm/hr) is applied thereafter. This is equivalent to the Green-Ampt-like structure in HEC-HMS “Initial and Constant Loss” and is useful for Mediterranean-climate catchments where depression storage and interception are a substantial fraction of short-duration storms.

Volume balance check

Whichever loss method you use, verify after the fact that the runoff volume under your hydrograph equals $P_e \times A$ to within a fraction of a percent. Systematic volume errors greater than 1 to 2 percent indicate a bug in the hyetograph specification, a timestep problem, or UH truncation on the recession.

Baseflow treatment

The unit hydrograph produces direct runoff only. Total streamflow includes a baseflow component from groundwater discharge. For design purposes baseflow is often small compared to the storm peak and can be ignored, but for longer-duration storms, large catchments, or wet-antecedent conditions it can be significant.

Supported baseflow options:

• Zero baseflow — appropriate for small urban catchments, very dry antecedent conditions, or when only the direct-runoff component is required for subsequent routing.
• Constant baseflow — $Q_b$ added to every ordinate. Simple and adequate for most design studies on perennial streams.
• Recession baseflow — an exponential recession $Q_b(t) = Q_{b,0} \cdot k^{t}$ where $k$ is a daily recession constant (typically 0.90 to 0.99 for South African catchments). Useful for long-duration simulation.
• Straight-line separation — from start of rise to the end of the recession, consistent with classical hydrograph-separation techniques.

Validation and quality checks

The tool reports the following diagnostic quantities after every run:

• Peak discharge $Q_p$ and time to peak $t_p$.
• Runoff volume $V$ (m³) — integrated area under the curve excluding baseflow.
• Volume error — $(V - P_e \cdot A) / (P_e \cdot A)$; should be < 1 %.
• Mass balance check — $\sum_i P_{e,i} \cdot A = \sum_k Q_k \cdot \Delta t$ (baseflow excluded).
• PRF back-calculation — $\mathrm{PRF}_\text{check} = Q_p \cdot T_p / A$; useful for confirming that the produced hydrograph is internally consistent with its declared method parameters.

Check your recession tail

Unit hydrographs have finite base times. If the storm is long enough that effective rainfall is still occurring after the UH has fully returned to zero from the first pulse, convolution is fine — but verify that the final recession is not being truncated at the simulation end time. The tool pads the simulation by $N_U \cdot \Delta t$ beyond the last rainfall pulse by default, but if you have shortened this window to manage output size, volume errors will appear.

Export formats

Generated hydrographs can be exported in formats compatible with common hydraulic and hydrologic modelling packages:

Format	File type	Use case
CSV	.csv	General purpose, spreadsheet import, custom processing
EPA SWMM	.dat	[INFLOWS] + [TIMESERIES] blocks for direct node inflow import
HEC-HMS time-series gage	.gage	Source / sink hydrograph import in a Basin Model
HEC-RAS unsteady boundary	.csv	Upstream / lateral boundary condition for unsteady 1D or 2D flow
InfoWorks ICM	.csv	DateTime / Flow format for Innovyze / Autodesk InfoWorks ICM
InfoSWMM / PCSWMM	.csv	Inflow hydrograph import
WORD / Excel report	.docx / .xlsx	Formatted deliverable with metadata, chart, and ordinates table

The CSV format includes a header block recording: method used, catchment parameters, timestep, peak, volume, mass-balance error, and ISO-8601 generation timestamp — so that downstream modellers have full provenance.

Worked example — 1:100 year hydrograph for a 20 km² rural catchment

The following example walks through generating a 1:100 year design hydrograph for a rural catchment in KwaZulu-Natal using the SCS Dimensionless Unit Hydrograph method with convolution.

Given information

• Catchment: 20 km², rolling topography (foothills of the Drakensberg)
• Time of concentration $T_c$ = 2.5 hr (from Tc Calculator, Kirpich method)
• Longest watercourse: 8.5 km, $S_\text{avg}$ = 0.015 m/m
• SCS Curve Number: 78 (AMC II, cultivated land on hydrologic soil group C)
• Design storm: 24-hour SCS Type II distribution, 1:100 year total depth = 150 mm (from Design Rainfall)
• PRF: 484 (default, consistent with rolling physiography)
• Target timestep: 15 minutes

Step 1 — Pulse duration and time to peak

Pulse duration $D$ = 15 min = 0.25 hr. Lag time $t_L = 0.6 \cdot T_c = 0.6 \cdot 2.5 = 1.5$ hr. Time to peak $T_p = D/2 + t_L = 0.125 + 1.5 = 1.625$ hr.

Step 2 — Unit hydrograph peak

Apply the SCS peak equation (SI):

$$q_p \;=\; \frac{C \cdot \mathrm{PRF} \cdot A}{T_p}$$

Using the metric-consistent constant $C = 1/645.33$ for PRF in its conventional units, and inserting $A = 20$ km² and $T_p = 1.625$ hr, the unit-depth UH peak is:

$$q_p \;=\; \frac{1/645.33 \cdot 484 \cdot 20}{1.625} \;\approx\; 9.23 \ \mathrm{m^3/s \ per \ mm}$$

This is the peak of the UH for 1 mm of effective rainfall applied over a 15-minute pulse.

Step 3 — Loss model: SCS-CN abstraction

With CN = 78, $S = 25400 / 78 - 254 = 71.6$ mm and $I_a = 0.2 \cdot S = 14.3$ mm.

Applying the SCS-CN equation to the 150 mm total:

$$P_e \;=\; \frac{(150 - 14.3)^2}{150 - 14.3 + 71.6} \;=\; \frac{18420}{207.3} \;\approx\; 88.9 \ \mathrm{mm}$$

Effective runoff fraction: 88.9 / 150 = 0.59. The loss model is applied cumulatively to the SCS Type II hyetograph so that each 15-minute pulse receives the correct incremental effective depth.

Step 4 — Convolution

The 24-hour storm contains 96 pulses of 15 minutes; the effective hyetograph (after CN abstraction) is convolved with the SCS dimensionless UH. The peak of the resulting direct-runoff hydrograph falls during the hour of peak effective rainfall intensity.

Step 5 — Expected results (tool output)

Typical output for this catchment:

Quantity	Value
Peak discharge $Q_p$	265 m³/s
Time to peak (from start of storm)	12.2 hr
Runoff volume	1.78 × 10⁶ m³
Base time	≈ 20 hr
Effective depth check	88.9 mm ✓
Mass balance error	0.2 % ✓

Step 6 — Interpretation

The hydrograph shape shows the classic Type II storm signature: a long, shallow rise until hour 11, a sharp rise through hour 12 driven by the intense central storm burst, and a long recession extending past hour 20. The peak of 265 m³/s at a specific volume of ~89 mm over 20 km² is consistent with regional experience for 1:100 year events on similar catchments.

For a downstream detention basin sized to attenuate this peak, the inflow hydrograph would be routed through the basin using the Flood Routing tool — the total volume (1.78 × 10⁶ m³) sets a lower bound on basin storage if any material attenuation is required.

Sensitivity analysis

Always run a sensitivity suite: vary $T_c$ by ±25 %, CN by ±5 units, and PRF by ±20 % to bracket the result. For this catchment, peak discharge typically varies by ±30 % across that parameter envelope — a reminder that the hydrograph is a model-based estimate, not a measurement.

Limitations

Synthetic unit hydrograph methods embed important assumptions that should be understood:

• Linearity — doubling effective rainfall exactly doubles the runoff peak. Real catchments are mildly non-linear; observed peaks from large events can be 10 to 30 percent higher than linearity predicts, because of channel overbank spreading, reduced roughness at high flows, and contribution from normally-dry areas.
• Spatial uniformity — the UH assumes that effective rainfall is applied uniformly over the catchment. For catchments larger than the storm-cell size (typically 100 to 500 km² in South African convective storms), this is a poor assumption. Use spatially-distributed radar-based hyetographs or coupled rainfall-runoff models for large basins.
• Stationary response — the UH is assumed to be a fixed function of the catchment. Urbanisation, deforestation, and large-scale land-use change will shift the UH; regional regression-based synthetic UHs that use land-use descriptors are appropriate when the catchment has changed materially since calibration.
• Single-peak response — the fundamental UH is a single-peak function. Multi-peak hydrographs arise only from multi-peak effective rainfall distributions. Catchments with bi-modal response (fast surface runoff + slow interflow/baseflow) are better represented by conceptual models (e.g. HBV, Sacramento) than by a single UH.
• Time invariance — the UH is assumed constant between events. In practice, antecedent moisture, frozen ground, and snow-cover states all modify the effective UH; the tool does not attempt to model these explicitly but the CN loss model captures moisture state approximately through AMC selection.
• Input uncertainty dominates — the largest source of error in most design hydrographs is not the UH method but the design rainfall depth, the CN value, and the time of concentration. Spend your uncertainty budget on those inputs before arguing about UH method selection.

Related tools and cross-references

• Design Storm — build the rainfall hyetograph that drives convolution.
• Design Rainfall — extract 1:T year rainfall depths by duration and location.
• Rational Method — peak discharge for small urban catchments; produces $Q_p$ you can feed into the tool’s scaled-hydrograph mode.
• SCS Method — the full SCS design flood method including Curve Number abstraction.
• Unit Hydrograph — the regionalised SA synthetic unit hydrograph (Midgley, HRU 1/72) as a dedicated peak-flow method.
• DRH Method — Design Rainfall Hydrograph, the SCS-equivalent SA workflow end-to-end.
• Flood Routing — route the hydrograph through a basin or reservoir.
• Tc Calculator — obtain $T_c$ needed by SCS, Snyder, and Clark.
• Watershed Delineation — obtain $A$, $L$, and slope.

References

• Chow, V.T., Maidment, D.R. & Mays, L.W. (1988). Applied Hydrology. McGraw-Hill, New York. Chapter 7 — Unit Hydrograph.
• USDA NRCS. (2007). National Engineering Handbook, Part 630, Chapter 16 — Hydrographs. United States Department of Agriculture, Natural Resources Conservation Service. (The authoritative reference for the SCS Dimensionless UH and PRF.)
• USDA SCS. (1972). National Engineering Handbook, Section 4 — Hydrology. Chapter 16 in the original NEH-4 series.
• SANRAL. (2013). Drainage Manual (6th ed.). South African National Roads Agency, Pretoria. Chapter 3 — Hydrology; Synthetic Unit Hydrograph methods.
• Midgley, D.C., Pitman, W.V. & Middleton, B.J. (1994). Surface Water Resources of South Africa 1990 (WR90), Water Research Commission, Pretoria. Report 298/1/94 — regional UH parameters.
• Hydrological Research Unit (HRU). (1972). Design Flood Determination in South Africa. Report 1/72, University of the Witwatersrand, Johannesburg.
• Sherman, L.K. (1932). Streamflow from rainfall by the unit-graph method. Engineering News-Record, 108, 501 – 505. (Original UH concept.)
• Snyder, F.F. (1938). Synthetic unit-graphs. Transactions, American Geophysical Union, 19(1), 447 – 454. (Snyder’s synthetic UH.)
• Clark, C.O. (1945). Storage and the unit hydrograph. Transactions, American Society of Civil Engineers, 110, 1419 – 1446. (Clark IUH and linear reservoir routing.)
• Aron, G. & White, E.L. (1982). Fitting a gamma distribution over a synthetic unit hydrograph. Water Resources Bulletin, 18(1), 95 – 98.
• USACE. (1994). Flood-Runoff Analysis. Engineer Manual EM 1110-2-1417, US Army Corps of Engineers, Washington DC. Chapters 6 – 8 — unit hydrograph theory and synthetic UH methods.
• USACE HEC. (2023). HEC-HMS Technical Reference Manual. US Army Corps of Engineers Hydrologic Engineering Center, Davis, California. ModClark and Clark IUH implementation details.
• Ponce, V.M. (1989). Engineering Hydrology — Principles and Practices. Prentice Hall, Englewood Cliffs, New Jersey. Chapter 5 — Hydrograph analysis; timestep guidance.
• Gericke, O.J. & Smithers, J.C. (2014). Review of methods used to estimate catchment response time for the purpose of peak discharge estimation. Hydrological Sciences Journal, 59(11), 1935 – 1971.

---

Open Hydrograph Generator