【目的】高时空分辨率的气温分布是准确模拟物种适宜分布区的基础。大熊猫分布区是全球35个生物多样性热点区之一,属于典型山地地形。【方法】为了有效消除海拔效应在气温空间插值中的误差,本研究基于线性递减率调整法(linear lapse rate adjustment, LLRA)对大熊猫分布区42个气象站点1960-2010年的日最高气温(daily maximum temperature,dT_(max))、日最低气温(daily minimum temperature,dT_(min))和日平均气温(daily mean temperature,dT_(mean))进行了空间插值,评估了LLRA插值法的准确度,并分析了近半个世纪大熊猫分布区气候变暖的格局。【结果】结果表明,使用LLRA法校正误差的局域递减率插值,准确度明显高于全球平均递减率和初步局域递减率;大熊猫分布区将近90%的区域dT_(max)、dT_(min)和dT_(mean)同时显著上升,上升最快的地区主要集中在秦岭和岷山西北部;dT_(max)、dT_(min)和dT_(mean)显著上升的中位速度分别为每100年上升1.68、2.32和1.77 ℃,超过3/4的区域dT_(min)上升速度大于dT_(max);在气象站点代表的低于3 500 m的海拔范围内,dT_(min)和dT_(mean)的变化速度随海拔上升表现出显著增加趋势,且dT_(min)的上升趋势大于dT_(mean)。【结论】本研究证明了LLRA法能有效消除海拔效应在气温空间插值时造成的误差,提高山地气温空间插值的准确度。研究发现,大熊猫分布区近半个世纪经历了显著变暖,由LLRA插值获得的逐日气温分布图可为进一步评估变暖对大熊猫分布区珍稀濒危物种的分布影响、制定相应的保护规划提供基础。
英文摘要:
[Objective] Distributions of air temperatures at high temporal and spatial resolution are the basis to model the suitable distribution range of species closely. The distribution range of giant panda is one of the 35 world's biodiversity hotspots, and the terrain of this region is heterogeneous and typical mountainous. [Method] To reduce the error induced by elevation effects on temperature interpolation effectively, we used the linear lapse rate adjustment method (LLRA) to interpolate the daily maximum temperature (dT_(max)), daily minimum temperature (dT_(min)), and daily mean temperature (dT_(mean)) of 42 meteorological stations throughout the distribution range of giant panda during 1960-2010. We assessed the accuracy of LLRA, and analyzed the pattern of warming in the distribution range of giant panda over the past half century. [Result] We found that the accuracy of interpolation using the local lapse rate adjusted by LLRA was higher than the global mean lapse rate and unadjusted local lapse rate. dT_(max), dT_(min), and dT_(mean) increased significantly simultaneously in nearly 90% of the distribution range of giant panda, and the area with the highest increasing rate mainly concentrated in the Qinling and northwestern Minshan Mountains. The median rate of significant increasing dT_(max), dT_(min) and dT_(mean) was 1.68, 2.32, and 1.77 every hundred years, respectively. More than 3/4 of the distribution range of giant panda was with a greater rate in dT_(min) than dT_(max). When the elevation was lower than 3 500 m, the representative elevations of the meteorological stations, the change rate of dT_(min) and dT_(mean) showed significant increasing trend with elevation, and the trend of dT_(min) was stronger than that of T_(mean). We confirmed that LLRA method can deal with elevation effects and improve the accuracy when interpolating temperature at high temporal and spatial resolution in mountainous areas. [Conclusion] The distribution range of giant panda experienced significant warming over the past half century, and the daily temperature maps from LLRA interpolation can be the basis for assessing the effects of warming on the distribution of rare and endangered species, and developing corresponding conservation plan.