Soil carbon loss by experimental warming in a tropical forest

Abstract

Tropical soils contain a third of global soil carbon 1 , so destabilization of soil organic matter caused by the approximate 4°C warming predicted for tropical regions this century could accelerate climate change by releasing additional carbon dioxide (CO2) to the atmosphere 2-5 . Theory predicts that warming should cause only modest carbon loss in tropical soils relative to those at higher latitudes 4,6 , but there have been no warming experiments in tropical forests to test this prediction 7 . Here we show that in situ experimental warming of a lowland tropical forest soil on Barro Colorado Island, Panama, caused an unexpectedly large increase in soil CO2 emissions. Two years of warming of the whole soil profile by 4 o C increased CO2 emission by 55% compared to soils at ambient temperature. The additional CO2 originated from heterotrophic rather than autotrophic sources and equated to a loss of 8.2 ± 4.2 (± 1 SE) Mg C ha -1 yr -1 from the breakdown of soil organic matter. During this time, we detected no acclimation of respiration rates, no thermal compensation or change in temperature sensitivity of enzyme activities, and no change in microbial carbon-use efficiency. These results demonstrate a high sensitivity of soil carbon in tropical forests to warming, which represents a potentially substantial positive feedback to climate change.

Reply on RC2

“…Bai et al, 2019;Fierer et al, 2003a;Lin et al, 2018). The latter indicate that subsoil respiration is equally or more sensitive to warming compared to topsoils (Bai et al, 2019;Fierer et al, 2003a;Lin et al, 2018), consistent with in situ observations of field warming experiments (Hicks Pries et al, 2017;Nottingham et al, 2020). The underlying microbial processes in subsoils are largely unexplored.…”

“…So far only a few field studies have been conducted to test the temperature sensitivity of subsoil microbial activity in situ (Hicks Pries et al, 2017;Nottingham et al, 2020), whereas there have been many incubation and mesocosm studies (e.g. Bai et al, 2019;Fierer et al, 2003a;Lin et al, 2018).…”

Influences of temperature and moisture on abiotic and biotic soil CO2 emission from a subtropical forest

“…This may be associated with relatively higher temperature sensitivity of autotrophic than heterotrophic respiration [ 33 ] or/and thermal acclimation of SOC decomposition, i.e., higher temperature sensitivity under low than high temperatures [ 19 , 30 , 34 ]. Air temperature in the study site had been increasing by 1.0 ± 0.1 ℃ during past five decades (1954–2009) [ 35 ] and the lasting increase in temperature may have consequently reduced the magnitude of C sink [ 36 ], due to extra soil C loss induced by warming [ 37 ].…”

Large seasonal variation of soil respiration in a secondary tropical moist forest in Puerto Rico

“…Despite small month‐to‐month variation in temperature (<4°C), seasonal patterns of daily and monthly R s were strongly correlated with air temperature, pointing to the importance of temperature as a driver of R s even in warm tropical forest ecosystems (Nottingham et al., 2020; Schwendenmann & Veldkamp, 2006; Schwendenmann et al., 2003; Sotta et al., 2004; Sotta et al., 2006; Wood et al., 2013). The significant positive linear relationship between R s and air temperature in this forest, suggests that soil CO 2 emissions in tropical forests may increase with projected warming as long as moisture and substrate are not limiting (Nottingham et al., 2020; Stocker et al., 2013; Townsend et al., 1992). The observed relationship between temperature and R s may also be related to seasonal variation in rates of primary productivity, which can affect belowground C investment and R s (Giardina & Ryan, 2000).…”

“…Regardless, observed patterns suggest that temperature plays an important role in regulating the temporal variation of R s in this forest. Whether the effects of temperature are direct (i.e., increasing rates of heterotrophic respiration) or indirect (i.e., increasing primary productivity) should be further explored using methods that can partition sources of R s (e.g., Nottingham et al., 2020).…”