17:15 〜 18:45
[MIS01-P19] Potential vegetation changes across Russia in warmer climates predicted from CMIP5 climate change scenarios by the end of the century
キーワード：Russia, vegetation, CMIP5 climate change scenarios , modeling
A large scale bioclimatic envelope model RuBCliM was used to predict the Russian boreal forest and its forest-forming coniferous species distributions under current and future climates. RuBCliM predicted a biome or a tree species from three orthogonal bioclimatic indices: Growing degree-days, base 5°C, Negative degree-days, base 0°C, and the annual moisture index AMI. An ensemble of 20 CMIP5 general circulation models and two greenhouse gas emissions scenarios (extreme and moderate) were employed to predict future climates in the 2080s (2070-2100). Model predictions for current climate were assessed using kappa-statistics against two independent maps of Russian forest biomes and dominant tree species: dark-needled (shade-tolerant) Pinus sibirica, Abies sibirica, Picea sibirica and Picea alba; and light-needled (shade intolerant) Pinus sylvestris, Larix sibirica and L. gmelini). Comparison between RuBCliM-modeled and actual vegetation map by Isachenko (1988) showed overall good match (kappa 0.46); good (0.43-0.55) for forest biomes (Dark- and Light-needled, Mixed and Broadleaved), but poor (0.28-0.34) for the ecotones Forest-tundra and Forest-steppe; good (0.44) and very good (0.57-0.67) for non-forest Tundra and Steppe/Semidesert respectively.
For future climates, our RuBCliM indicated dramatic shifts northwards in both biomes and dominant conifers and from conifers to broad or small-leaved deciduous species at the southern forest border for an extreme scenario and modest changes for a moderate scenario.
Our projections have profound implications for feedbacks between the energy budget, carbon cycle, and land cover.
Fig.1. Upper. Comparison between RuBCliM-modeled (left) and actual vegetation map by Isachenko (1988, right). Lower. Vegetation distribution in a warmer climate at the 2080s: under moderate RCP 2.6 (left) and extreme RCP 8.5 scenarios. Vegetation legend. 1. Tundra, 2. Forest-tundra, 3. Dark-needled taiga, 4. Light-needled taiga, 5. Mixed forest, 6. Broad-keaved deciduous forest, 7. Steppe, 8. Semidesert, 9. Forest-steppe.
For future climates, our RuBCliM indicated dramatic shifts northwards in both biomes and dominant conifers and from conifers to broad or small-leaved deciduous species at the southern forest border for an extreme scenario and modest changes for a moderate scenario.
Our projections have profound implications for feedbacks between the energy budget, carbon cycle, and land cover.
Fig.1. Upper. Comparison between RuBCliM-modeled (left) and actual vegetation map by Isachenko (1988, right). Lower. Vegetation distribution in a warmer climate at the 2080s: under moderate RCP 2.6 (left) and extreme RCP 8.5 scenarios. Vegetation legend. 1. Tundra, 2. Forest-tundra, 3. Dark-needled taiga, 4. Light-needled taiga, 5. Mixed forest, 6. Broad-keaved deciduous forest, 7. Steppe, 8. Semidesert, 9. Forest-steppe.