CMIP5地球系统模式溶解氧模拟结果分析

本文基于常用的统计方法,通过与WOA09观测的海洋溶解氧浓度数据进行比较,定量地评估了9个CMIP5地球系统模式在历史排放试验中海洋溶解氧气候态特征的模拟能力。在海表,由于地球系统模式均能很好地模拟海表温度（SST）,模式模拟的海表溶解氧浓度分布与观测一致,模拟结果无论是全球平均浓度偏差还是均方根误差均接近0,空间相关系数与标准偏差接近1。在海洋中层以及深层这些重要水团所在的区域,各模式的模拟能力则差异较大,尤其在溶解氧低值区（OMZs）所在的500m到1000m,各模式均出现全球平均偏差、均方根误差的极大值以及空间相关系数的极小值。在海洋内部,模式偏差的原因比较复杂。经向翻转环流和颗粒有机碳通量均对模式的偏差有贡献。分析结果表明物理场偏差对溶解氧偏差的贡献较大。一些重要水团,比如北大西洋深水,南极底层水以及北太平洋中层水在极大程度上影响了溶解氧在这些海区的分布。需要指出的是,虽然在海洋内部各模式模拟的溶解氧浓度偏差较大,但是多模式平均结果却能表现出与观测较好的一致性。     

Prediction of Seaward Slope Recession in Berm Breakwaters Using M5' Machine Learning Approach

In the design process of berm breakwaters, their front slope recession has an inevitable rule in large number of model tests, and this parameter being studied. This research draws its data from Moghim’s and Shekari’s experiment results. These experiments consist of two different 2D model tests in two wave flumes, in which the berm recession to different sea state and structural parameters have been studied. Irregular waves with a JONSWAP spectrum were used in both test series. A total of 412 test results were used to cover the impact of sea state conditions such as wave height, wave period, storm duration and water depth at the toe of the structure, and structural parameters such as berm elevation from still water level, berm width and stone diameter on berm recession parameters. In this paper, a new set of equations for berm recession is derived using the M5'' model tree as a machine learning approach. A comparison is made between the estimations by the new formula and the formulae recently given by other researchers to show the preference of new M5'' approach.     

Intercomparison of CRA-Interim Precipitation Products with ERA5 and JRA-55

By using the hourly data from surface meteorological stations in China, the 3-hour precipitation data from CRA-Interim (Chinese Reanalysis-Interim), ERA5 (ECMWF Reanalysis 5) and JRA-55 (Japanese Reanalysis-55) are compared, both on the spatial-temporal distributions and on bias with observation precipitation in China. The results show that: (1) The three sets of reanalysis datasets can all reflect the basic spatial distribution characteristics of annual average precipitation in China. The simulation of topographic forced precipitation in complex terrain by CRA-interim is more detailed, while CRA-interim has larger negative bias in central and East China, and larger positive bias in southwest China. (2) In terms of seasonal precipitation, the three sets of reanalysis datasets overestimate the precipitation in the heavy rainfall zone of spring and summer, especially in southwest China. CRA interim’s location of the rain belt in the First Rainy Season in South China is west by south, the summer precipitation has positive bias in southwest and South China. (3) All of the reanalysis datasets can basically reflect the distribution difference of inter-annual variation of drought and flood, but the overall the CRA-Interim generally shows negative bias, while the ERA5 and JRA-55 exhibit positive bias. (4) For the diurnal variation of precipitation in summer, all the reanalysis datasets perform better in simulating the daytime precipitation than in the night, and bias of CRA-interim is less in southeast and northeast than elsewhere. (5) ERA5 generally performs the best on the evaluation of quantitative precipitation forecast, the JRA-55 is the next, followed by the CRA-Interim. CRA-Interim has higher missing rate and lower threat score for heavy rains; however, at the level of downpour, the CRA-Interim performs slightly better.     

模式内部变率引起的1.5℃和2℃升温阈值出现时间模拟的不确定性研究

模式内部变率是模拟结果不确定性的重要来源,然而它对于1.5℃和2℃升温阈值出现时间不确定性的影响尚不清楚。因此,基于耦合模式比较计划第五阶段(CMIP5)的多模式数据研究了模式内部变率对1.5℃和2℃升温阈值出现时间不确定性的影响以及对未来排放情景的敏感性。结果表明,模式内部变率对升温阈值出现时间模拟的影响与外强迫的影响相当,单个模式内部不同成员达到全球平均1.5℃或2℃增温的年份相差2~12年;其影响具有明显的空间差异,影响极大值出现在欧亚大陆以北洋面、白令海峡周围区域、北美东北部及其与格陵兰岛之间的海域、南半球高纬地区等;低排放情景下模式内部变率的影响大于高排放情景。     

Soil moisture retrieval from satellite images and its application to heavy rainfall simulation in eastern China

The soil water index (SWI) from satellite remote sensing and the observational soil moisture from agricultural meteorological stations in eastern China are used to retrieve soil moisture. The analysis of correlation coefficient (CORR), root-mean-square-error (RMSE) and bias (BIAS) shows that the retrieved soil moisture is convincible and close to the observation. The method can overcome the difficulties in soil moisture observation on a large scale and the retrieved soil moisture may reflect the distribution of the real soil moisture objectively. The retrieved soil moisture is used as an initial scheme to replace initial conditions of soil moisture (NCEP) in the model MM5V3 to simulate the heavy rainfall in 1998. Three heavy rainfall processes during 13–14 June, 18–22 June, and 21–26 July 1998 in the Yangtze River valley are analyzed. The first two processes show that the intensity and location of simulated precipitation from SWI are better than those from NCEP and closer to the observed values. The simulated heavy rainfall for 21–26 July shows that the update of soil moisture initial conditions can improve the model’s performance. The relationship between soil moisture and rainfall may explain that the stronger rainfall intensity for SWI in the Yangtze River valley is the result of the greater simulated soil moisture from SWI prior to the heavy rainfall date than that from NCEP, and leads to the decline of temperature in the corresponding area in the heavy rainfall days. Detailed analysis of the heavy rainfall on 13–14 June shows that both land-atmosphere interactions and atmospheric circulation were responsible for the heavy rainfall, and it shows how the SWI simulation improves the simulation. The development of mesoscale systems plays an important role in the simulation regarding the change of initial soil moisture for SWI.     

Future Changes in Extreme High Temperature over China at 1.5℃-5℃ Global Warming Based on CMIP6 Simulations

Extreme high temperature(EHT)events are among the most impact-related consequences related to climate change,especially for China,a nation with a large population that is vulnerable to the climate warming.Based on the latest Coupled Model Intercomparison Project Phase 6(CMIP6),this study assesses future EHT changes across China at five specific global warming thresholds(1.5℃-5℃).The results indicate that global mean temperature will increase by 1.5℃/2℃ before 2030/2050 relative to pre-industrial levels(1861-1900)under three future scenarios(SSP1-2.6,SSP2-4.5,and SSP5-8.5),and warming will occur faster under SSP5-8.5 compared to SSP1-2.6 and SSP2-4.5.Under SSP5-8.5,global warming will eventually exceed 5℃ by 2100,while under SSP1-2.6,it will stabilize around 2℃ after 2050.In China,most of the areas where warming exceeds global average levels will be located in Tibet and northern China(Northwest China,North China and Northeast China),covering 50%-70%of the country.Furthermore,about 0.19-0.44 billion people(accounting for 16%-41%of the national population)will experience warming above the global average.Compared to present-day(1995-2014),the warmest day(TXx)will increase most notably in northern China,while the number of warm days(TX90p)and warm spell duration indicator(WSDI)will increase most profoundly in southern China.For example,relative to the present-day,TXx will increase by 1℃-5℃ in northern China,and TX90p(WSDI)will increase by 25-150(10-80)days in southern China at 1.5℃-5℃ global warming.Compared to 2℃-5℃,limiting global warming to 1.5℃ will help avoid about 36%-87%of the EHT increases in China.     

波段选择协同表达高光谱异常探测算法

高光谱遥感影像具有光谱分辨率极高的特点,承载了大量可区分不同类型地物的诊断性光谱信息以及区分亚类相似地物之间细微差别的光谱信息,在目标探测领域具有独特的优势。与此同时,高光谱遥感影像也带来了数据维数高、邻近波段之间存在大量冗余信息的问题,高维度的数据结构往往使得高光谱影像异常目标类和背景类之间的可分性降低。为了缓解上述问题,本文提出了一种基于波段选择的协同表达高光谱异常探测算法。首先,使用最优聚类框架对高光谱波段进行选择,获得一组波段子集来表示原有的全部波段,使得高光谱影像异常目标类与背景类之间的可分性增强。然后使用协同表达对影像上的像元进行重建,由于异常目标类和背景类之间的可分性增强,对异常目标像元进行协同表达时将会得到更大的残差,异常目标像元的输出值增大,可以更好地实现异常目标和背景类的分离。本文使用了3组高光谱影像数据进行异常目标探测实验,实验结果表明,该方法与其他现有高光谱异常目标探测算法对比,曲线下面积AUC(Area Under Curve)值更高,可以更好地实现异常目标与背景分离,能够更有效地对高光谱影像进行异常目标探测。     

基于多分辨率分析的GF-5和GF-1遥感影像空—谱融合

针对空间分辨率比率较大尺度差异下的高分五号(GF-5)与高分一号(GF-1)卫星影像的空—谱融合问题,提出多传感器影像融合策略:一方面,通过现有空—谱融合方法的分步融合得到融合影像;另一方面,在分步融合理论基础上,推导得出一体化融合基础框架,并进一步提出基于多分辨率分析的多传感器一体化融合方法,缓解现有方法因空间分辨率比率过大导致影像空、谱互补信息难以有效集成的问题。其中,提出的一体化融合方法基于调制传递函数MTF (Modulation Transfer Function)滤波对多传感器影像空间(高频)和光谱(低频)分量进行分解提取,并充分考虑多传感器高空间分辨率影像与高光谱分辨率影像之间的关系,以及高光谱分辨率影像波段间关系,设计合理的融合权重,最终可得到具有最高空间分辨率和最高光谱分辨率的融合影像。通过GF-1全色影像、GF-1多光谱影像、GF-5高光谱影像数据对提出方法进行实验验证,结果表明:本文方法可有效集成多传感器影像间的空、谱互补信息,得到较优融合结果。     

顾及水陆差异的高分五号影像条带去除

高分五号搭载的可见光短波红外高光谱相机,能获取精细的地物光谱信息,具有十分广泛的应用前景。但高光谱卫星遥感数据往往无法避免条带噪声的干扰,进行条带去除是数据预处理中不可缺少的步骤。传统方法往往对地物的异质性与丰富的细节纹理考虑不足,导致条带不能被彻底地消除。为此,本文提出了一种顾及水陆差异的影像条带去除方法,采用水体与陆地区别统计的策略,解决条带噪声在异质区域的统计特征差异问题,并结合优化统计和一维变分滤波技术实现水陆区域参考统计特征的精确估计,最终基于矩匹配方法分别实现条带去除。实验结果表明:无论是真实实验还是模拟实验,本文提出的算法相较于传统条带去除算法,能更加稳健地去除数据条带噪声,还原地表真实辐射信息;在模拟实验中,本文算法处理结果的峰值信噪比(PSNR)达到46.58,且平均绝对误差(MAE)仅有11.56,均明显优于用于比较的3种传统算法,且算法执行效率也具备优势,能够更好地适用于高分五号大数据量的处理需求。     

高光谱遥感技术在基岩区区域地质调查填图中的应用

高光谱遥感以其超高的光谱维数据优势,使对地物的精细识别和区分能力较传统多光谱遥感数据有质的提升.以HyMap高光谱数据和高分五号高光谱数据为数据源,选择中国西部基岩区区域开展了高光谱遥感岩性-构造解译工作.通过图像增强处理后,对研究区地层单元、岩体/脉、构造等进行了遥感地质解译.对比已有的地质调查结果,发现高光谱遥感数...