基于贝叶斯模型的器测、古气候重建与气候模拟数据的融合试验

气象部门馆藏的西部最早的器测气象资料始于20世纪30年代,不能满足建立20世纪以来中国气候变化序列的需求,而古气候重建或气候模拟资料则可以延伸到器测时代以前。为了探讨长序列多源气候资料序列融合方法,采用贝叶斯方法对中国北疆地区8条树轮气温重建资料、器测资料与国际耦合模式比较计划第5阶段（CMIP5）模式模拟资料进行了融合试验。首先利用器测资料对气温代用资料进行校验与网格重建,并以此作为贝叶斯模型的先验分布,然后,用泰勒图选出了该区域气候模拟效果最佳的几个模式;最后将网格重建和气候模拟序列用贝叶斯模型进行了融合试验。结果表明,贝叶斯融合模型能有效提取各种数据来源的有用信息进行融合,融合结果的长期变化（线性）趋势更接近器测气候序列,并在一定程度上提高了序列的精度,减小了结果的不确定性;并且,融合结果能够纠正先验分布及气候模拟数据的明显偏差,为长年代气候序列重建提供了一个可行的思路。      

Linking moisture and near-surface wind with winter temperature to reveal the Holocene climate evolution in arid Xinjiang region of China

An increasing number of palaeo-climatic records have been reported to identify the Holocene climate history in the arid Xinjiang region of northwest China. However, few studies have fully considered the internal linkages within the regional climate system, which may limit our understanding of the forcing mechanisms of Holocene climate change in this region. Here, we systematically consider three major issues of the moisture/precipitation, temperature and near-surface wind relevant to the Holocene climate history of Xinjiang. First, despite there still has debated for the Holocene moisture evolution in this region, more climatic reconstructions from lake sediments, loess, sand-dunes and peats support a long-term regional wetting trend. Second, temperature records from ice cores, peats and stalagmites demonstrate a long-term winter warming trend during the Holocene in middle- to high-latitudes of Asia. Third, recent studies of aeolian sedimentary sequences reveal that the near-surface winds in winter gradually weakened during the Holocene, whereas the winter mid-latitude Westerlies strengthened in the Tienshan Mountains. Based on this evidence, in the arid Xinjiang region we propose an early to middle Holocene relatively cold and dry interval, with strong near-surface winds; and a warmer, wetter interval with weaker near-surface winds in the middle to late Holocene during winter. Additionally, we develop a conceptual model to explain the pattern of Holocene climate changes in this region. From the early to the late Holocene, the increasing atmospheric CO₂ content and winter insolation, and the shrinking of high-latitude continental ice-sheets, resulted in increasing winter temperatures in middle to high latitudes in the Northern Hemisphere. Subsequently, the increased winter temperature strengthened the winter mid-latitude Westerlies and weakened the Siberian high-pressure system, which caused an increase in winter precipitation and a decrease in near-surface wind strength. This scenario is strongly supported by evidence from geological records, climate simulation results, and modern reanalysis data. Our hypothesis highlights the important contribution of winter temperature in driving the Holocene climatic evolution of the arid Xinjiang region, and it implies that the socio-economic development and water resources security of this region will face serious challenges presented by the increasing winter temperature in the future.     

边界层温度廓线遥感反演的本地化改进研究

采用MODIS资料和美国发展的MODIS大气温、湿度廓线统计反演算法,估算大气温度、湿度廓线作为初始场,应用101层快速透过率模式(PFAAST)估算了大气透过率,并采用Newton非线性迭代算法反演中国西北荒漠戈壁地区大气温度廓线。结果表明:该方法对边界层高度及以上部分的大气温度反演得比较好,误差基本都在2 K范围内,边界层范围内的温度反演误差较大,反演误差与气溶胶光学厚度增量和地表温度估算误差呈显著正相关关系,与大气水汽混合比的关系较差。文中从敏感性试验和理论分析角度阐述了地表温度和气溶胶光学厚度估算误差对大气温度反演误差的影响,发现不同光谱波段的地表温度权重均随地表温度的增加有不同程度增加,地表温度反演误差增加将增加地表温度权重,提高地表温度估算误差有助于提高地表温度权重的精度;荒漠戈壁地区大气边界层中气溶胶浓度较高,光学厚度较大,使边界层大气透过率降低,进而降低卫星红外遥感波段的地表温度权重和空气温度权重。由于该模式没有很好地考虑边界层中沙尘气溶胶的影响,使卫星反演的大气透过率偏高,以至于高估地表温度权重和大气温度权重,使得反演的表面温度和空气温度偏低。该研究结合太阳光度计获得的光学厚度资料,采用统计方法对气溶胶效应引起的大气透过率误差和表面温度估算误差进行校正,并对物理算法进行本地化改进,实现了边界层温度廓线的反演。     

山区年、月平均温度推算方法的研究

本文提出了山区年、月平均温度空间分布的推算方法,并提供了相应的小地形温度订正值。该方法在福建省沙溪流域山区实际应用的结果表明,能有效地揭示山区年、月平均温度空间分布的宏观与微观变化,效果满意,精度较高。     

喀什站高空温度资料均一性检验与订正

利用加拿大环境部气候研究中心研发的RHtest均一化检验方法,结合历史台站沿革信息对喀什站各标准等压面高空温度资料进行了均一性检验和分析。结果表明:850～300 hPa共6个层次的高空日温度观测资料序列完整性较好。08时100 hPa月平均温度的缺测率3%,完整性较差,20时100 hPa的完整性最差,缺测率高达11.44%,其他层次的完整性则较好。喀什站温度资料的非均一性主要是由于台站位置变动、观测仪器换型、观测规范变更和辐射误差订正方法的变化引起的。在订正过程中发现,为了保证订正序列的可信度,需要充分考虑到原始观测序列的缺测率。08时年温度序列的断点数量要明显多于20时,08时序列的订正率达到了85.71%,而20时仅有100 h Pa一条序列做了订正。     

Spatio-temporal dynamics of land cover and land surface temperature in Ganges-Brahmaputra delta: A comparative analysis between India and Bangladesh

Land use and land cover (LULC) change in the Ganges-Brahmaputra delta (GBD) poses significant challenges towards future environmental sustainability of the region and requires regional scale monitoring of key bio-physical variables and changes in their inter-relationship over space and time. Focusing on the southern part of the lower GBD region along the international border of India and Bangladesh, this study examined the spatio-temporal variability of LULC change and its relationship with Land Surface Temperature (LST). Furthermore, LULC-LST relationships were compared between Indian and Bangladesh part and its trend in and around big cities (with more than 1 million population) and towns (with more than 100,000 population) was investigated. Results showed that LST changes were predominantly driven by LULC changes on both sides of the border. Urban growth is the dominant form of LULC change, and the rate of land change was faster in 2005–2010 time period than 1989–2005. Over the period of 21 years, mean January LST decreased by approximately 1.83 °C in Indian part and 1.85 °C in the Bangladesh part. Areas that changed from to rural from agricultural experienced decrease in mean LST, whereas those areas that changed to urban from either agriculture or rural, experienced increase in mean LST. The relationship between LULC and LST are same on both sides of the Indo-Bangladesh border. In bigger cities like Kolkata (in India) and Khulna (in Bangladesh), there is a high spatial variability in relationship between LULC and LST compared to large towns. The LULC-LST relationship in large towns in India was influenced by proximity to Kolkata and coastal areas, whereas in Bangladesh no such influence was evident. The results and the data produced in this study are crucial for monitoring LULC changes, for developing spatial decision support system, and thus will be helpful to address the current challenges of land management in the GBD region. Changes in the LULC and LST are important indicators of GBD's environmental health and access its vulnerability and thus the present findings serve as baseline information for future studies seeking to examine the impact of differential policies on the LULC change in the region.     

近百余年上海气温变化的若干特征

本文根据1873—1990年上海的年、冬、夏平均气温资料,并利用1956—1990年其郊区气温资料,对城市热岛效应对气温的影响作了处理,重建了近118a的上海气温序列。在此基础上对上海的气温变化特征进行分析,并与同期的北半球气温变化作了比较。结果表明,上海的气温变化有其明显的阶段性、跃变性和周期性。近百余年上海气温变化的总体特征与北半球相似,呈缓慢变暖趋势,但具体的变化过程和幅度有差别,尤其近20多年来的变化与北半球不一致,没有明显的增温现象。     

STRATIFIED MARINE LAKES OF PALAU (WESTERN CAROLINE ISLANDS)

The physical geography of the limestone islands of Palau permits permanent water-column stratification in 13 tropical sea-level marine lakes, each with unique water-column physics, chemistry, and biology. Embayments and lagoons amid the coral became isolated as marine lakes after Miocene uplifting. Surface mixing of lake water by wind is reduced by jungle-covered karst ridges. Surface tidal exchange through fissures in fenestrated karst is slow while midwater exchange through submarine tunnels is fast, but both produce damped, delayed tides with modest seawater exchange from the barrier-reef lagoon. Topographic protection from wind, heavy regular rain throughout the year, with precipitation exceeding evaporation, and modest tidal exchange produce stratified water columns with brackish waters above permanently anoxic saline hypolimnia. Permanent lake stratification is documented for 18 years; sediment cores (by others) show stratification for >100 years, and recent constant sea level implies ecosystem stability for thousands of years. Therefore, the marine lakes in Palau are small, closed, simple ecosystems that do not change over time–steady-state chemostats permitting replicate field measurements of biological and physical attributes from day to day, month to month, or decade to decade. [Key words: Palau, meromictic, marine lake, karst, lake stratification, tides, model ecosystem.]     

Riparian thermal conditions across a mixed rural and urban landscape

Riparian corridors have the potential to function as thermal refuges, moderating extremes of local temperature variation. However, although demonstrated at individual sites, and over short periods, the consistency of this effect at wider temporal and spatial scales is poorly understood. The aim of this study is to assess the temperature differences between riparian corridors and adjacent non-riparian habitats and to explore the influence of environmental characteristics on these differences. Air temperature was monitored hourly at 20 paired locations (riparian and non-riparian) for two consecutive years. Urban index and canopy cover were characterised by calculating the percentage of impervious surface area and tree canopy cover within a 100 m radius from the centre of each sampling site. Canopy cover reduced summer thermal stresses in both urban and rural areas whereas high urban index tended to increase the daily thermal indices. Rivers had a significant mitigating effect on the urban riparian thermal condition, particularly in extreme hot weather. Riparian corridors were generally 1 °C cooler than non-riparian locations in summer and could be up to 3 °C cooler at some sites in extreme hot weather. Furthermore, riparian corridors at some sites were warmer than non-riparian locations in winter. These findings suggest that the proximity of rivers can modify riparian thermal environments, potentially reducing the heat stress of riparian corridors across landscapes.