The rapid expansion of the rail network in the 19th century created nearly 30,000 km of Technosol corridors across Great Britain (GB). Today, the GB railway estate covers over 51,000 ha and is managed by Network Rail Infrastructure Limited. A base line estimate of the soil organic carbon (SOC) stock is required to support Net Zero objectives. For this study 338 cores from 87 sites were collected from disused railway lines as an accessible proxy to the active network. Technosols are often excluded from soil carbon accounting and there are no estimates of railway soil carbon stocks.
Our analysis of soil cores revealed a mean (±SD) SOC concentration (SOCc) of 5.0 % (±3.7), corresponding to an average SOC density of 49.7 t ha−1 (±27.8) to a depth of 30 cm. Significant factors affecting SOCc included parent material, bulk density, moisture and soil texture while habitat and climate had less influence. Railway-specific factors such as structure, construction and abandonment dates had minimal impact. Mixed effects linear modelling explained 55 % of the SOCc variation (R2 = 0.55). With no soil data available for the working railways, a reduced-factor general linear model, incorporating underlying bedrock, adjacent soil type and habitat (R2 = 0.19), was used to produce an initial SOC density map for the active rail network This gave an average carbon density for the Network Rail estate of 29.7 t ha−1 and a total soil carbon stock of 1.52 million tonnes (±6430). This is significantly lower than natural soils and many other technosols and suggests that these immature soils have the potential to sequester more carbon, assisted by appropriate land and vegetation management.