Soil temperature synchronisation improves representation of diel variability of ecosystem respiration in Sphagnum peatlands

Abstract The temperature dependence of Ecosystem Respiration (ER) is often assessed based on the temperature of one specific layer. Air temperature or temperatures in the first ten centimetres of the soil profile are the most frequently used temperatures in models. However, previous studies showed that the relationship between ER and temperature is depth dependent, making depth selection for temperature measurements an important issue, especially at short time-scales. The present study explores one possible way to assess this relationship by synchronising the ER and temperature signals and to test if the relationship between ER and temperature differs between daytime and nighttime. To do so, ER measurements were undertaken in 2013 in four Sphagnum -peatlands across France using the closed chamber method. The ER fluxes were measured hourly during 72 h in each of four replicates in each site. Synchronisations between ER and T signal were determined for each depth (from surface to 30 cm depth) by selecting the time-delay leading to the best correlation between ER and soil temperatures and ER was then modelled. Our results showed that: (i) the delay between ER and soil temperature is greater in peat than in mineral soils; (ii) at a daily time-scale synchronisation can improve the model representation using soil temperatures.