Soil Moisture

Unit: Volumetric water content (m³/m³) — the volume of water per unit volume of soil

Soil moisture is the slow-moving background variable of fire weather. While temperature, humidity, and wind change hour to hour, soil moisture trends develop over weeks to months. It connects atmospheric conditions to the plants growing in the ground — and when soils dry, vegetation loses access to water, becomes stressed, and eventually becomes more flammable.

Why it matters for fire weather

Soil moisture influences fire danger through two main pathways:

Vegetation water stress. Plants extract water from the soil through their roots. When soil moisture drops, plants close their stomata to conserve water, reducing the moisture content of their leaves and stems. This live fuel moisture determines whether green vegetation can carry fire. Research has found that soil moisture can explain roughly two-thirds of the variability in live fuel moisture content (Qi et al., 2012).

Seasonal fire potential. Soil moisture at the start of summer is largely determined by winter and spring rainfall. A wet winter builds a moisture reserve that buffers vegetation through early dry spells; a dry winter means the fire season starts from a deficit. Fire seasons following anomalously dry winters are typically more severe because the background moisture deficit means fuels dry faster and fire danger builds more quickly.

How it works

The ECMWF IFS soil model

Wildflyer displays soil moisture from the ECMWF IFS weather model, which includes a land surface scheme called ECLand (evolved from HTESSEL). Rather than treating the ground as a single layer, ECLand models soil water movement across four depth layers:

Layer	Depth	Thickness	What it captures
1	0–7 cm	7 cm	Surface conditions — strongest coupling to recent weather
2	7–28 cm	21 cm	Shallow root zone — where grasses and crops extract water
3	28–100 cm	72 cm	Deep root zone — where shrubs and most trees access moisture
4	100–289 cm	189 cm	Deep reservoir — relevant to deep-rooted trees in extended drought

The four layers are arranged in an approximate geometric progression, designed to capture timescales from one day (the surface layer responding to a rain shower) to one year (the deep reservoir tracking seasonal and multi-year drought). This design provides “a reasonable compromise between computational cost and the ability to represent all timescales between one day and a year” (Balsamo et al., 2009).

Water enters the top layer through infiltration from rainfall and snowmelt. It leaves through surface evaporation, root extraction by vegetation, and drainage between layers. Free drainage is assumed at the bottom boundary.

Vegetation tiles and root depth

Each grid cell in the model is represented as a mosaic of up to eight surface tiles, including separate fractions for high vegetation (trees) and low vegetation (grasses, crops). Each vegetation type has a characteristic root depth distribution based on lookup tables from the BATS vegetation classification (Boussetta et al., 2021).

This is the mechanism through which soil moisture at different depths connects to different vegetation:

• Low vegetation (grasses, crops, herbaceous plants) has shallow roots concentrated in layers 1–2 (0–28 cm). These plants are the first to feel drought when the upper soil dries.
• High vegetation (broadleaf trees, needleleaf trees) has roots penetrating into layers 3–4 (down to 100–289 cm). Trees can access deeper water reserves that grasses cannot reach — but when extended drought depletes those reserves too, the consequences are severe.

During drought, shallow layers dry first, curing grasses and stressing fine vegetation before deeper-rooted trees are affected. Prolonged drought eventually depletes all layers.

Shallow vs. deep: what each depth tells you

The layered structure means shallow and deep soil moisture tell different stories about fire danger:

• Shallow soil moisture (0–28 cm) responds to recent rainfall and short-term weather. It correlates with dead fuel moisture and grass curing — the short-term, fine-fuel side of fire danger. When this layer dries, fine vegetation cures rapidly and surface fuels become available to burn (Krueger et al., 2023).

• Deep soil moisture (28–100+ cm) reflects longer-term drought trends. It relates to live fuel moisture in shrubs and trees — when deep soils dry, canopy water content drops and woody vegetation becomes more flammable. Multi-year drought that depletes deep layers increases the likelihood of crown fires and, in extreme cases, tree mortality.

The build-up / dry-down pattern

A distinctive soil moisture pattern precedes many large wildfires. O et al. (2020) analysed global fire records and found:

• In drier climates: Above-normal soil moisture appears approximately five months before a fire. These wet conditions promote vegetation growth, building up the fuel load. This is followed by a progressive drying that creates ignition conditions in that accumulated fuel. The wetter the pre-fire period, the larger the subsequent fires tended to be.

• In wetter climates: The pattern is simpler — fires are preceded by drier-than-normal soil in the weeks before ignition, creating suitable conditions in an otherwise too-wet environment.

This “wet-then-dry” sequence — sometimes called hydroclimate whiplash (Alizadeh et al., 2024) — highlights that fire danger is not just about current dryness. The history of soil moisture matters: wet periods grow the fuel, dry periods make it available to burn.

Key thresholds

Value (m³/m³)	Condition	Significance
< 0.15	Below wilting point	Plants cannot extract water — severe water stress
0.15–0.35	Between wilting point and field capacity	Normal range — vegetation actively transpiring
> 0.35	At or above field capacity	Saturated — recent heavy rainfall

Note

Wilting point and field capacity vary by soil type (sandy soils drain faster than clay). The values above are approximate and are used by Wildflyer as visualization reference lines.

Interannual comparison

Raw soil moisture values are most useful when compared to what is normal for a given location and time of year:

• Is this year drier than last year at the same date?
• Is current soil moisture below the long-term average?

If both answers are yes, expect elevated fire danger even before hot, dry summer weather arrives.

Recovery after rain

After rainfall, soil moisture reveals how much water was actually retained. In drought conditions, soils can become hydrophobic (water-repellent), meaning rain runs off instead of infiltrating — a deceptive recovery where surface wetting masks continued deep drought.

How to read it in Wildflyer

Wildflyer shows soil moisture in two places:

• Fuel Conditions tab (Expert Weather View) — displays the 0–7 cm surface layer with reference lines for wilting point and field capacity. This gives a quick read on whether surface conditions are stressed.

• Interannual comparison — compare soil moisture trends across years to identify whether the current season is tracking drier or wetter than previous years at the same point in time.

Both views use data from the ECMWF IFS model (ECLand land surface scheme), reported as volumetric water content in m³/m³.

Going deeper

How ECLand calculates soil moisture

The ECLand scheme solves the Richards equation for vertical water flow through unsaturated soil. Each layer exchanges water with its neighbours based on hydraulic conductivity and matric potential, which depend on the soil’s texture (sand/clay fraction) and current water content.

The model tracks volumetric water content (m³/m³) in each layer, bounded between the permanent wilting point (below which roots cannot extract water) and porosity (total pore space). Between these bounds, evaporation efficiency from the soil surface drops linearly between field capacity and wilting point.

Root extraction follows the root distribution profile for each vegetation type. When soil moisture in a root-accessible layer drops below a stress threshold, the model reduces transpiration — the same mechanism that causes real plants to close stomata and wilt during drought.

Media opportunity

The ECMWF IFS documentation includes a diagram (Fig. 2.1.4.5-1) showing the four-layer soil structure with vegetation tiles, root zones, and heat/moisture fluxes. Licensed under CC-BY-4.0 — could be included with credit to ECMWF.

Soil moisture and fire danger rating systems

Current operational fire danger rating systems — including the Canadian FWI, the US NFDRS, and EFFIS — do not directly incorporate soil moisture measurements. They use proxy calculations from temperature, humidity, wind, and precipitation to estimate moisture in fuel layers of different depths. The FWI’s Drought Code (DC), for example, tracks deep organic layer drying using a simple water balance model.

Krueger et al. (2023) argue that direct soil moisture information could improve these systems by providing:

• Better estimates of dynamic fuel loads (how much vegetation is available to burn)
• More accurate live fuel moisture predictions for shrubs and trees
• Earlier warning of fire danger through the build-up pattern
• Better forecasts of fire occurrence and burned area

As soil moisture data becomes more available — through both weather models like ECMWF IFS and satellite missions like ESA’s SMOS and NASA’s SMAP — its integration into fire weather assessment is expected to grow. This is an active area of research.

VPD, soil moisture, and live fuel moisture

Vapour pressure deficit (VPD) and soil moisture interact to control live fuel moisture. VPD drives atmospheric demand — how aggressively the air pulls moisture from vegetation. Soil moisture determines the supply — how much water roots can deliver to replace what’s lost.

When VPD is high and soil moisture is low, plants face a double stress: high demand and low supply. Research by Griebel et al. (2023) found that VPD and specific leaf area are the dominant controls on live fuel moisture content. Resco de Dios et al. (2021) showed that climate change-driven increases in VPD may turn currently fire-free mountain forests into fire-prone ecosystems — a shift mediated partly through declining soil moisture.

Sources

• Alizadeh, M. R., et al. (2024). Land and atmosphere precursors to fuel loading, wildfire ignition and post-fire recovery. Geophysical Research Letters, 51, e2023GL105324.
• Balsamo, G., et al. (2009). A revised hydrology for the ECMWF model: Verification from field site to terrestrial water storage and impact in the Integrated Forecast System. Journal of Hydrometeorology, 10(3), 623–643.
• Boussetta, S., et al. (2021). ECLand: The ECMWF Land Surface Modelling System. Atmosphere, 12(6), 723.
• Griebel, A., et al. (2023). Specific leaf area and vapour pressure deficit control live fuel moisture content. Functional Ecology, 37, 1156–1170.
• Krueger, E. S., et al. (2023). Using soil moisture information to better understand and predict wildfire danger: a review of recent developments and outstanding questions. International Journal of Wildland Fire, 32(2), 111–132.
• O, S., Hou, X. & Orth, R. (2020). Observational evidence of wildfire-promoting soil moisture anomalies. Scientific Reports, 10, 11008.
• Qi, Y., et al. (2012). Monitoring live fuel moisture using soil moisture and remote sensing proxies. Fire Ecology, 8(3), 71–87.
• Resco de Dios, V., et al. (2021). Climate change induced declines in fuel moisture may turn currently fire-free Pyrenean mountain forests into fire-prone ecosystems. Science of the Total Environment, 797, 149104.