Temperature and precipitation projections for the Antarctic Peninsula over the next two decades: contrasting global and regional climate model simulations
Temperature and precipitation projections for the Antarctic Peninsula over the next two decades: contrasting global and regional climate model simulations
复制标题
DOI：
10.1007/s00382-021-05667-2
复制
发表时间：
2021-02
期刊：
影响因子：
4.6
通讯作者：
D. Bozkurt;D. Bromwich;Jorge Carrasco;R. Rondanelli
中科院分区：
文献类型：
--
作者：
D. Bozkurt;D. Bromwich;Jorge Carrasco;R. Rondanelli
本文研究了高排放情景（RCP8.5）下南极半岛近未来（2020-2044年）的温度和降水变化。我们利用了参与耦合模式比对项目第5阶段（CMIP5）的19个全球气候模式（GCMs）的历史和预估模拟。我们将GCMs预估与两组区域气候模式模拟（RCMs）进行了比较和对比：(1)利用偏校正后的NCAR-CESM1 （NC-CORR）强迫Polar-WRF模式在南极半岛进行的高分辨率（15公里）模拟，(2)利用corde - antarctica获得的EC- earth （EC）强迫KNMI-RACMO21P进行的中分辨率（50公里）模拟。对ERA5再分析的历史模拟（1981-2005）的进一步比较也包括环流型和近地表温度气候学。总的来说，两个RCM边界条件都很好地代表了历史时期的主要环流模式。尽管如此，在EC和NC-CORR的预估中，阿蒙森海低压分别显著加深和减弱。预计整个半岛的年平均近地表温度将上升约0.5 - 1.5摄氏度。秋冬两季气温上升幅度较大（2℃）。随着相反的环流型变化，EC和NC-CORR均表现出不同的升温速率，表明自然年代际变率可能继续存在。尽管总体上显示相似的温度变化，但与各自驱动场相比，RCM预估显示拉森冰架变暖较小，融化日数增加较小。关于降水，模拟结果基本一致，表明年平均降水增加（5% ~ 10%）。然而，RCMs在拉森冰架上显示出一些显著的差异，其中总降水量减少（对于RACMO），降雨频率略有增加。我们得出的结论是，在两种RCM边界条件下，从gcm得到南极半岛一致的预估似乎仍然很困难。此外，边界条件的主导和共同变化在RCM模拟中非常明显。我们认为，RCM预估的附加值是由RCM引入的更精细的局部细节和不同物理方案所形成的过程驱动的，特别是在拉森冰架上空。
This study presents near future (2020–2044) temperature and precipitation changes over the Antarctic Peninsula under the high-emission scenario (RCP8.5). We make use of historical and projected simulations from 19 global climate models (GCMs) participating in Coupled Model Intercomparison Project phase 5 (CMIP5). We compare and contrast GCMs projections with two groups of regional climate model simulations (RCMs): (1) high resolution (15-km) simulations performed with Polar-WRF model forced with bias-corrected NCAR-CESM1 (NC-CORR) over the Antarctic Peninsula, (2) medium resolution (50-km) simulations of KNMI-RACMO21P forced with EC-EARTH (EC) obtained from the CORDEX-Antarctica. A further comparison of historical simulations (1981–2005) with respect to ERA5 reanalysis is also included for circulation patterns and near-surface temperature climatology. In general, both RCM boundary conditions represent well the main circulation patterns of the historical period. Nonetheless, there are important differences in projections such as a notable deepening and weakening of the Amundsen Sea Low in EC and NC-CORR, respectively. Mean annual near-surface temperatures are projected to increase by about 0.5–1.5C across the entire peninsula. Temperature increase is more substantial in autumn and winter (2C). Following opposite circulation pattern changes, both EC and NC-CORR exhibit different warming rates, indicating a possible continuation of natural decadal variability. Although generally showing similar temperature changes, RCM projections show less warming and a smaller increase in melt days in the Larsen Ice Shelf compared to their respective driving fields. Regarding precipitation, there is a broad agreement among the simulations, indicating an increase in mean annual precipitation (5 to 10%). However, RCMs show some notable differences over the Larsen Ice Shelf where total precipitation decreases (for RACMO) and shows a small increase in rain frequency. We conclude that it seems still difficult to get consistent projections from GCMs for the Antarctic Peninsula as depicted in both RCM boundary conditions. In addition, dominant and common changes from the boundary conditions are largely evident in the RCM simulations. We argue that added value of RCM projections is driven by processes shaped by finer local details and different physics schemes that are introduced by RCMs, particularly over the Larsen Ice Shelf.