Skip to content

Quickstart — from rainfall series to hydrograph in 5 minutes

This guide walks through the minimum path to get a flood hydrograph starting from annual maxima of short-duration rainfall.

Want to try it without your own data?

Open the Info modal (top-right ? icon) → Resources tab → click "Cosenza — Metro_Cosenza". This loads a complete station with a real ~30-year series. The rest of the guide works the same.

1. Open the study

  1. Go to nx.geostru.ai/runofflab/.
  2. First time, the study is empty. You'll see Station metadata (name, code, basin, coordinates) and a list of durations below.
  3. The header shows * next to the name when there are unsaved changes, and Unsaved changes in small text just below.

2. Enter the rainfall series

For each duration (60′, 180′, 360′, 720′, 1440′) paste the annual maxima in millimetres.

  • One value per line. Decimals with dot or comma.
  • You need at least 5 values for a probabilistic analysis to make statistical sense; below 10-12 the results are quite uncertain.
  • Durations with fewer than 2 values are skipped during analysis.

Example

Duration 60′ (1 hour), years 1991-2020, Cosenza: 38.6, 41.2, 36.4, 52.1, … (one value per line).

Save as .hgstudy via File → Save As (or Ctrl+Shift+S). The file is a portable JSON you can email or share.

3. Run a probabilistic analysis

In the Analyses panel → "Add analysis" → pick a family. For the first study, start with:

  • Gumbel — Method of moments (simple, robust, default in many Italian annuals).

The software computes:

  • distribution parameters (\(\mu, \beta\) for Gumbel);
  • rainfall heights \(h(T)\) for standard return periods (5, 10, 50, 100, 200 years);
  • goodness-of-fit tests (KS and χ²).

To compare methods, also add a GEV (L-moments) or TCEV Level 1 (when in Italy and the VA.PI. dataset covers your region).

See Probabilistic analyses for method details.

4. Build the rainfall curve

In the Rainfall curves panel → "Add curve" → pick the source analysis and the return period (e.g. T = 100 years).

Log-log regression estimates the two parameters of

\[ h(t) = a \cdot t^{n} \]

with \(t\) in hours and \(h\) in mm. Typical southern-Italy values: \(n \approx 0.30\text{–}0.45\).

5. Synthetic hyetograph

From the curve, build a design hyetograph using Chicago or the alternating-block method:

  • Synthetic hyetographs panel → "Add hyetograph".
  • Set total duration (e.g. 60 min) and peak position \(r\) (typically 0.4).
  • The output is a time series of intensity \(i(t)\) — the rainfall "arriving" at the basin.

6. Flood hydrograph via SCS-CN

Hydrographs panel → "Add hydrograph". Enter:

  • Basin area (km²);
  • Curve Number CN (0-100, depends on soil + land cover);
  • Lag time Tlag (hours).

CN and Tlag wizards

Don't know the right CN? Open the integrated CN wizard — pick hydrologic soil type and land use; CN gets area-weighted automatically.

The tool applies SCS-CN (net rainfall) and convolves with the dimensionless SCS unit hydrograph, returning:

  • volume and duration of net rainfall;
  • peak discharge \(Q_p\) and time-to-peak;
  • the full \(Q(t)\) curve.

7. Export the report

File → Export PDF report produces a single document with:

  • station metadata and series;
  • parameter tables for all analyses and h(T);
  • rainfall curve plots;
  • hyetographs and hydrographs;
  • method and formulas for each section.

Next steps: Full workflow · Analyses in depth · AI Import