Stage-discharge relationships connect water level (called stage) to the volume of water flowing past a point per unit time (called discharge). This connection is what makes continuous streamflow monitoring practical: measuring water level is cheap and easy to automate, while measuring discharge directly is expensive and labor-intensive. By establishing a reliable relationship between the two, you can record stage around the clock and convert it to discharge whenever you need it.

These relationships are visualized through rating curves, which are the backbone of nearly every stream gauging station in the world. They support everything from reservoir operations to flood warning systems to long-term hydrologic research.

Stage-Discharge Relationships in Streamflow

The core idea is straightforward: at a given cross-section of a river, higher water levels generally correspond to higher discharge. If you can define that relationship mathematically, you only need to measure stage to estimate discharge.

Why this matters so much:

Cost-effective monitoring. Direct discharge measurements require field crews and specialized equipment. Stage can be recorded continuously with a simple pressure transducer or staff gauge.
Continuous data. Instead of occasional snapshots of discharge, you get an unbroken time series, which is far more useful for analysis.
Broad applications. Stage-discharge relationships feed into water allocation decisions, flood forecasting and emergency response, ecological flow assessments, and calibration of hydrologic models.

The relationship holds best at stable cross-sections where the channel geometry doesn't change much over time. That's why gauging stations are typically placed at locations with bedrock or otherwise resistant channel beds.

Development of Rating Curves

Building a rating curve requires collecting paired measurements of stage and discharge across a wide range of flow conditions, then fitting a mathematical relationship to those data.

Step 1: Collect stage-discharge pairs.

Measure stage using a staff gauge, pressure transducer, or other water level sensor.
Simultaneously measure discharge using velocity-area methods:
- Current meters (mechanical or electromagnetic) measure velocity at multiple points across the channel cross-section. You multiply velocity by the area of each subsection and sum them.
- Acoustic Doppler Current Profilers (ADCPs) measure velocity profiles across the entire cross-section more rapidly, and are now standard for medium-to-large rivers.

You need measurements spanning as much of the flow range as possible, from low base flows to high flood flows. This often takes years of field visits.

Step 2: Plot and fit the data.

Plot the paired measurements with stage on the y-axis and discharge on the x-axis. (Note: the hydrologic convention places discharge on the x-axis and stage on the y-axis, since stage is the variable you measure and discharge is what you derive. Some textbooks reverse this, so check your course's convention.)
Fit a curve through the data points. The relationship is typically non-linear, often following a power function:

$Q = a(h - h_0)^b$

where $Q$ is discharge, $h$ is stage, $h_0$ is the stage of zero flow, and $a$ and $b$ are fitted parameters. Taking the logarithm of both sides linearizes this, which is why rating curves often appear as straight lines on log-log plots.

Common fitting techniques include least squares regression on log-transformed data and segmented curves (using different equations for low, medium, and high flow ranges).

Step 3: Extrapolate beyond measured data.

You'll rarely have direct measurements at the most extreme high or low flows. To extend the curve, hydrologists use hydraulic equations such as Manning's equation or weir/flume equations to estimate what the relationship should look like outside the measured range.
Extrapolation introduces additional uncertainty, so discharge estimates at the extremes of the curve should be treated with more caution.

Limitations of Rating Curves

Rating curves assume a stable, unique relationship between stage and discharge at a cross-section. Several factors can violate that assumption:

Channel geometry changes. Erosion, sediment deposition, or channel migration can alter the cross-section over time, shifting the entire rating curve. A flood that scours the bed will produce a different stage-discharge relationship than existed before.
Vegetation growth. Seasonal vegetation in or near the channel increases hydraulic roughness, which raises stage for a given discharge.
Ice formation. Ice cover or ice jams create backwater effects that dramatically change the stage-discharge relationship, sometimes making the standard rating curve unusable during winter.
Hysteresis. During a flood event, the discharge at a given stage is often higher on the rising limb of the hydrograph than on the falling limb. This loop effect means a single curve can't perfectly represent rapidly changing conditions.
Measurement error. Both stage and discharge measurements carry uncertainty from instrument precision, turbulence, and human factors during field work.

How hydrologists address these limitations:

Update rating curves regularly (often annually or after major flood events) with new field measurements.
Use shift corrections or maintain separate rating curves for different seasons or flow regimes (e.g., one for ice-free conditions, another for winter).
Apply uncertainty analysis, including confidence intervals around discharge estimates and Bayesian methods that incorporate new data to refine the curve over time.
Document and report the uncertainty bounds on all published discharge data so that downstream users understand the reliability of the estimates.

Application of Rating Curves

Once a rating curve is established, converting stage to discharge is a simple lookup or calculation:

Record stage at the gauging station (typically logged automatically at 15-minute or hourly intervals).
Apply the rating curve equation to each stage value to compute discharge.
Compile the results into a continuous discharge time series, often called a hydrograph.

Interpreting the results requires context:

Look for temporal patterns such as seasonal high and low flows, storm-driven peaks, and long-term trends that might indicate land use change or climate shifts.
Compare discharge at different stations to understand spatial patterns, like how tributaries contribute flow or how discharge changes moving downstream.
Check estimated discharge against historical records to detect anomalies, against hydrologic model predictions to validate or calibrate those models, and against regulatory thresholds for compliance monitoring.

Always keep the rating curve's limitations in mind when interpreting results. Discharge estimates during extreme events or periods of channel instability carry wider uncertainty, and reporting that uncertainty is part of good hydrologic practice.

2,589 studying →