How We Calculate Your Wind Potential
Every number, formula and saving estimate on the Swales wind dashboard — explained simply.
Wind energy is one of the most misunderstood resources in renewable planning. A site that looks exposed and breezy can disappoint on paper; a sheltered valley can surprise you. This article walks through every calculation behind the Swales wind potential tool so you know exactly what the numbers mean, where they come from, and how confident you can be in them.
1. Where the Data Comes From
Swales pulls wind data from the Global Wind Atlas (GWA), a dataset produced by the Technical University of Denmark (DTU) and the World Bank. It combines decades of weather station records with numerical weather modelling to produce high-resolution wind maps for the entire planet.
For the monthly seasonality profile we use ERA5 reanalysis data served via Open-Meteo — the same satellite-era atmospheric reconstruction used by meteorologists worldwide.
When you click a location on the map, Swales draws a 3 km × 3 km square polygon around that point and queries the GWA API for every wind pixel inside that area. This avoids basing the estimate on a single pixel, which could sit on a hilltop or in a local hollow.
2. Mean Wind Speed — and Why We Don't Use the P90
The GWA returns a histogram of wind speed pixels across your polygon. Each entry looks like this:
{ val: "6.2", count: "120" } ← 120 pixels recorded 6.2 m/s
{ val: "7.0", count: "85" } ← 85 pixels recorded 7.0 m/s
We calculate the weighted mean across all pixels:
Mean wind speed = Σ (val × count) / Σ count
This gives the representative average across the whole area.
The GWA also provides a statistic called ninth_decile — the wind speed exceeded by the top 10 % of pixels. We deliberately do not use this. The P90 is an optimistic high-end estimate; it tells you what the windiest corner of your site might see, not what a turbine positioned anywhere on the site is likely to experience. The weighted mean is the honest representative value.
3. Hub Heights — Why Height Changes Everything
Wind speed increases with altitude. The ground creates friction that slows the lower air layers. At 10 metres above a rough landscape, wind might be 5 m/s; at 100 metres, the same location could see 8 m/s or more.
Swales fetches wind data at four hub heights matched to realistic turbine installations:
| Category | Hub Height | Typical Use Case | |------------|-----------|-----------------| | Home | 10 m | Micro-turbine on a mast or rooftop | | Farm | 50 m | Mid-size farm turbine | | Business | 100 m | Commercial-scale turbine | | Industrial | 150 m | Utility-scale wind turbine |
Each hub height gets its own separate GWA query, so switching between categories in the dashboard is not a rough approximation — it is a fresh measurement at that actual height.
4. Wind Power Density — the Physics Behind the Number
Wind power density (W/m²) is derived from the fundamental physics of a moving air mass:
P / A = ½ × ρ × v³
Where:
- P/A = power per unit of rotor swept area (W/m²)
- ρ = air density ≈ 1.225 kg/m³ at sea level, standard atmosphere
- v = wind speed (m/s)
The cube relationship is the most important thing to understand here. Double the wind speed and you get eight times more power. This is why 8 m/s is not just a bit better than 4 m/s — it is eight times more powerful. Small differences in mean wind speed have large consequences for energy output.
Swales fetches power density directly from GWA (which applies corrections for local air density and terrain) and applies the same weighted-mean calculation described above.
Traffic light thresholds for power density:
| Signal | Threshold | Meaning | |--------|-----------|---------| | Green | ≥ 500 W/m² | High density — strong turbine investment case | | Amber | ≥ 200 W/m² | Moderate — viable with right turbine and height | | Red | < 200 W/m² | Low — supplementary generation only |
5. Site Classification by Category
Not all categories are equal. A home turbine at 10 m hub height should require a lower wind speed to be "Excellent" than an industrial turbine at 150 m — because the industrial machine has much higher capital cost and needs stronger winds to justify it. Swales uses per-category wind speed thresholds:
| Category | Excellent | Good | Moderate | Poor | |------------|-----------|--------|----------|--------| | Home | ≥ 7.0 m/s | ≥ 5.0 | ≥ 3.5 | < 3.5 | | Farm | ≥ 8.0 m/s | ≥ 6.0 | ≥ 4.5 | < 4.5 | | Business | ≥ 9.0 m/s | ≥ 7.0 | ≥ 5.5 | < 5.5 | | Industrial | ≥ 10.0 m/s| ≥ 8.0 | ≥ 6.0 | < 6.0 |
These thresholds reflect real industry guidance. A 6 m/s annual mean is genuinely excellent for a residential 10 m mast; the same speed would make a 150 m industrial project economically marginal.
The gauge on the hero card fills to a level proportional to the wind speed against the "Excellent" threshold for the selected category. The fill formula is:
Fill % = (mean wind speed / (Excellent threshold × 1.3)) × 100
The 1.3 multiplier means the gauge can reach 100 % only at very windy sites, so it never falsely shows a "full" reading for a merely average site.
6. Reliability Score — Seasonal Consistency
A turbine that generates well in winter but produces almost nothing in summer creates problems: you still need grid backup for half the year. The reliability score measures how consistent generation is across months.
Swales uses the Weibull distribution (shape parameter k = 2, also known as the Rayleigh distribution) — the standard model for wind speed probability in energy planning. The higher the k value, the more consistent the wind. With k = 2, the distribution captures the typical variability seen at most land-based sites.
The reliability score is calculated from:
- The mean wind speed (higher mean → more reliable above cut-in speed)
- The Weibull k (shape of the speed distribution)
- The monthly wind speed profile (seasonal spread from peak to trough)
A high reliability (≥ 70) means the site generates meaningful wind power in most months with little seasonal swing — ideal for year-round energy planning. A low reliability means generation is strongly seasonal and a backup strategy (grid import, battery, or solar pairing) is needed for the quiet months.
Traffic light thresholds for reliability:
| Signal | Threshold | Meaning | |--------|-----------|---------| | Green | ≥ 70 | Stable — consistent year-round generation | | Amber | ≥ 50 | Variable — seasonal storage or backup needed | | Red | < 50 | Unreliable — significant seasonal dependence |
7. Turbine Selection
The dashboard selects a reference turbine size based on the category and the site's mean wind speed. This is not a purchase recommendation — it is a realistic benchmark for the calculations. The size brackets used are:
| Category | Turbine Range | Benchmark Logic | |------------|----------------|-----------------| | Home | 5 – 15 kW | Micro-turbine; scaled to residential demand | | Farm | 50 – 250 kW | Farm-scale; covers pumping and operational loads | | Business | ~500 kW | Commercial C&I turbine | | Industrial | 2 – 4 MW | Utility turbine; scaled to power density |
8. Annual Energy Yield
Annual energy output is estimated using:
Annual Energy (kWh) = Turbine Capacity (kW)
× Capacity Factor
× 8,760 hours/year
The capacity factor is the most important adjustment. A turbine rated at 100 kW never runs at 100 kW for every hour of the year — wind speed varies continuously. The capacity factor (typically 25–45 % for land-based wind) represents the fraction of nameplate capacity actually delivered over a year.
Swales derives the capacity factor from the mean wind speed, wind power density, and hub height using the wind energy industry's standard power curve approach — higher wind speeds and higher hub heights both improve the capacity factor.
9. Savings Estimates per Category
Once annual energy yield is known, Swales translates it into a practical number relevant to each use case.
Home — Months of Bills Offset
Months offset = min( round( (Annual kWh / 3,600) × 12 ), 12 )
A typical UK home uses approximately 3,600 kWh per year. Dividing your turbine's annual output by this and multiplying by 12 gives the number of months in which the turbine could cover your full electricity consumption. The result is capped at 12 — a turbine cannot offset more than a full year's worth of bills.
Farm — Irrigation Pump Hours Powered
Pump hours = round( Annual kWh / 7.5 )
Based on a 7.5 kW irrigation pump — a standard mid-size pump used for field irrigation. This shows how many hours per year the turbine runs that pump at zero electricity cost, directly reducing operating expenditure.
Business — Annual Bill Reduction
Annual saving (k£) = round( Annual kWh × £0.28 / 1,000 )
Calculated at the UK commercial electricity rate of £0.28 per kWh (as of 2024; this rate fluctuates with energy markets). The result is shown in thousands of pounds. This is a gross saving before financing costs — use it as the starting point for your payback period calculation.
Industrial — Grid Offset Potential
Grid offset (GWh) = Annual kWh / 1,000,000
At industrial scale, the relevant unit is gigawatt-hours (GWh). Each GWh of on-site wind generation displaces approximately 200 tonnes of CO₂ compared to the UK grid average. This metric is most useful when negotiating Power Purchase Agreements or reporting Scope 2 emissions.
10. Monthly Wind Profile
The monthly chart shows the average wind speed for each calendar month at the selected hub height, drawn from GWA's temporal analysis layer (ERA5-derived).
The chart is deliberately scaled to this location's own monthly range rather than a fixed global scale. This makes seasonal patterns visible even at consistently windy sites — the chart answers "which months are windiest here?" not "is this an overall good site?" (that question is answered by the hero gauge and verdict card above).
Colour coding is relative to the site's own range:
- Green — windiest months at this location
- Orange — average months
- Red — calmest months
Bar heights are also relative: the tallest bar is always the windiest month, the shortest is always the calmest. The peak and calmest month names and speeds are shown in the legend below the chart.
11. Wind Direction Rose
The wind rose shows the frequency of wind from each compass direction at 100 m hub height. Longer petals mean wind blows from that direction more often. This is important for:
- Turbine siting — position the turbine so the dominant wind direction faces an open fetch with no obstructions
- Obstacle avoidance — trees, buildings and ridges upwind of the dominant direction cause turbulence that reduces output and increases mechanical fatigue
- Planning — local authorities often require wind rose data as part of a turbine planning application
The rose is fetched at 100 m and shared across all categories, since wind direction at a location does not change significantly with height.
12. What These Estimates Do Not Account For
Every number on the dashboard is a model estimate based on large-area reanalysis data. The following factors can significantly affect real-world output and are not captured:
- Local terrain — a hilltop turbine will outperform the model; a turbine in a sheltered hollow will underperform
- Turbulence — caused by nearby trees, buildings, and complex terrain; reduces both output and turbine lifespan
- Wake effects — if other turbines are nearby, their wakes reduce wind speed for downwind machines
- Turbine model specifics — actual power curves vary between manufacturers and models
- Planning and grid connection — in many areas, the connection cost or planning constraints are the binding constraint, not the wind resource itself
- Maintenance — real-world availability (typically 95–98 % for modern turbines) is not factored into the capacity factor estimate
Always commission a professional wind resource assessment — including on-site anemometry data for at least 12 months — before committing to a turbine investment.
Summary of Key Numbers
| Parameter | Source | Formula | |-----------|--------|---------| | Mean wind speed | Global Wind Atlas — weighted pixel mean | Σ(v × count) / Σ count | | Wind power density | Global Wind Atlas — weighted pixel mean | ½ × 1.225 × v³ | | Hub heights | Fixed per category | 10 / 50 / 100 / 150 m | | Gauge fill | Category-adjusted normalisation | (v / (v_excellent × 1.3)) × 100 | | Annual yield | Capacity factor model | Capacity (kW) × CF × 8,760 h | | Home saving | UK average home consumption | (kWh / 3,600) × 12 months | | Farm saving | 7.5 kW reference pump | kWh / 7.5 | | Business saving | UK commercial rate £0.28/kWh | kWh × 0.28 / 1,000 | | Industrial saving | GWh conversion | kWh / 1,000,000 | | Reliability | Weibull k=2 (Rayleigh) model | Based on mean speed + monthly spread |
Data sources: Global Wind Atlas v3 (DTU / World Bank), ERA5 reanalysis via Open-Meteo. Electricity rate: Ofgem commercial benchmark 2024. CO₂ factor: UK DESNZ grid intensity 2024.