The contribution of land-use change versus climate variability to the1940s CO 2 plateau: Former Soviet Union as a test case
Ana Bastos,Anna Peregon,Érico A. Gani,Sergey Khudyaev,Chao Yue,Wei Li,Célia M. Gouveia,Philippe Ciais
DOI: https://doi.org/10.5194/bg-2017-267
2017-01-01
Biogeosciences Discussions
Abstract:Abstract. According to the ice-core record, atmospheric CO 2 growth rate ( plateau ) stalled during the 1940s, in spite of maintained anthropogenic emissions from fossil fuel burning and land-use change. Bastos et al. (2016) have shown that the state-of-the-art reconstructions of CO 2 sources and sinks do not allow closing the global CO 2 budget during this period. Their study indicates that even considering an enhancement of the ocean sink, still a gap sink of 0.4–1.5 PgC.yr −1 in terrestrial ecosystems is needed to explain the CO 2 stabilization. They hypothesised that (i) the major socioeconomic and demographic disruptions during World War II (WWII) may have led to massive land-abandonment, resulting in an additional sink from regrowing natural vegetation which is not accounted for in most reconstructions and/or (ii) the warming registered at the same time, especially in the high-latitudes, might have led to increased vegetation growth and an enhancement of the natural sink. Here, we test the different contributions of these two factors in the Former Soviet Union (FSU), motivated by several reasons. On the one hand, the territory of the FSU encompasses 15 % of the terrestrial surface, 20 % of the global soil organic carbon pool and is responsible for a considerable fraction of the present-day terrestrial CO 2 sink. On the other hand, heavy economic and demographic losses have been registered in FSU during WWII, together with likely decrease in farmland due to occupation, destruction of infrastructure and shortages of manpower. Here we present a newly compiled dataset of annual agricultural area in FSU, which better matches other socioeconomic indicators and reports a decrease in cropland of ca. 62 Mha between 1940–1943. We use an updated version of the land-surface model ORCHIDEE, ORCHIDEE-MICT, which is specifically developed to better represent high-latitude processes to simulate the carbon fluxes in terrestrial ecosystems over the 20th century. Using our new cropland dataset, we test the different contributions of the land-use change and the decadal warming reported in the 1940s to explain the plateau . As reference, we compare our results with the gap sink estimated by the group of land-surface models in Bastos et al. (2016): 0.7 PgC/yr. We find that the massive cropland decrease between 1940–1943, even if short-termed, could result in an additional decadal sink of 0.04–0.07 PgC/yr, i.e. 6–10 % of the gap sink required to explain the plateau . The ORCHIDEE-MICT simulations also indicate a very strong enhancement of the terrestrial sink by 0.4 PgC/yr, explaining about 60 % of the gap sink from the TRENDYv4 models. This enhancement is mainly explained by tree-growth in high-latitudes coincident with strongest warming sustained over the 1940–1949 decade, which is not captured by any of the other land-surface models. Even if land-abandonment during WWII might contribute to a relatively small fraction of the sink required to explain the plateau , it is still non-negligible, especially since such events have likely been registered in other regions. The vegetation growth in high-latitudes simulated by ORCHIDEE-MICT and absent in other models appears to be supported by tree-ring records, highlighting the relevance of improving the representation of high-latitude hydrological and soil processes in order to better capture decadal variability in the terrestrial CO 2 sink.