Addition of nitrogen fertiliser increases net ecosystem carbon dioxide uptake and the loss of soil organic carbon in grassland growing in mesocosms

- We measured net ecosystem carbon exchange and its components in grassland mesocosms.
- Addition of high nitrogen increased net ecosystem carbon uptake.
- We used a non-disruptive technique to partition soil heterotrophic respiration.
- High nitrogen increased heterotrophic respiration.
- We conclude that increases in net carbon uptake can mask losses of soil carbon.

Maintaining and increasing soil organic carbon stocks in grasslands is essential for sustainable productivity and to offset anthropogenic carbon emissions. Direct measurements of net ecosystem carbon dioxide exchange, FN, can be used to detect whether an ecosystem is a net sink or a source of carbon. However partitioning heterotrophic respiration, RH, from ecosystem respiration, RE, is needed to determine the impacts of land-use and global change on soil organic carbon stocks. We extracted intact soil cores from an intensively grazed dairy farm to establish mesocosms growing in controlled conditions and we subjected them to low and high nitrogen treatments (100 and 400 kgN ha1 y− 1, respectively). After concurrent clipping and addition of nitrogen, we measured the timing for the recovery of FN and its components, photosynthesis, A, and ecosystem respiration, RE, by measuring them daily. Subsequently, we measured RH from the same mesocosms seven days after the treatments were applied, using a non-disruptive, natural abundance carbon isotope technique. To test the significance of the presence of living roots when measuring RH, we compared the results obtained from the isotopic approach to those obtained from a root exclusion technique, which involved removing the roots from the mesocosms. As the plants grew after clipping, FN decreased (increasing net CO2 uptake) exponentially to mean (± standard error) steady-state values of 1.11 ± 0.26 μmol m− 2 s− 1 (net source) and − 0.19 ± 0.33 μmol m− 2 s− 1 (near neutral) for the low and high nitrogen treatments, respectively. When measured using the isotopic approach, RH increased by 60%, from 1.26 ± 0.29 μmol m− 2 s− 1 in the low, to 2.06 ± 0.55 μmol m− 2 s− 1 in the high nitrogen treatment. Thus, addition of the high nitrogen resulted in an increase in soil organic carbon loss concurrently with an increase in net uptake of carbon by the ecosystem in the high nitrogen treatment compared with the low nitrogen treatment. When measured in the absence of living roots using the root exclusion technique, RH was overall much higher than the value obtained with the isotopic technique (4.34 ± 0.13 μmol m− 2 s− 1), indicating an apparent negative rhizosphere priming effect. Furthermore, when using the root exclusion technique, no difference was found between the nitrogen treatments, suggesting that the presence of roots mediated the response of heterotrophic respiration to addition of nitrogen. These results highlight the need to include measurements of changes in RH alongside measurements of FN in non-disturbed ecosystems to interpret the processes regulating the effects of management practices on long-term changes in soil organic carbon stocks.