An accessible seismological dataset of 2021 Yangbi MS6.4 earthquake

A M_S6.4 earthquake occurred on 21 May 2021 in Yangbi county, Dali prefecture, Yunnan, China, at 21: 48 Beijing Time (13: 48 UTC). Earthquakes with an M3.0 or higher occurred before and after the main shock. Seismic data analysis is essential for the in-depth investigation of the 2021 Yangbi M_S6.4 earthquake sequence and the seismotectonics of northwestern Yunnan. Institute of Geophysics, China Earthquake Administration (CEA), has compiled a dataset of seismological observations from 157 broadband stations located within 500 km of the epicenter, and has made this dataset available to the earthquake science research community. The dataset (total file size: 329 GB) consists of event waveforms with a sampling frequency of 100 sps collected from 18 to 28 May 2021, 20-Hz and 100-Hz continuous waveforms collected from 12 to 31 May 2021, and seismic instrument response files. To promote data sharing, the dataset also includes the seismic event waveforms from 20 to 22 May 2021 recorded at 50 stations of the ongoing Binchuan Active Source Geophysical Observation Project, for which the data protection period has not expired. Sample waveforms of the main shock are included in the appendix of this article and can be downloaded from the Earthquake Science website. The event and continuous waveforms are available from the Earthquake Science Data Center website (www.esdc.ac.cn) on application.     

Prompt seismic data sharing for the 2021 Maduo earthquake in Qinghai province,China

An M_S7.4 earthquake struck west China in Maduo county, Guoluo prefecture, Qinghai province on May 22, 2021, at 2:04 Beijing time (18:04 UTC on May 21, 2021), which broke the quiet period of Chinese mainland for 1382 days without earthquakes of magnitude 7 or higher. The analysis of the seismic data sequence would play an important role in the in-depth study of the Maduo earthquake and the Bayan Har block. The Institute of Geophysics, China Earthquake Administration (CEA), compiled observation data recorded through 57 broadband seismometers within 500 km of the earthquake epicenter and intended to share for further researches in earthquake science community. The shared dataset included waveforms of the event and its sequence with magnitudes of 3.0 or higher that occurred between May 22–31, 2021 with a sampling rate of 100 sps along with the continuous waveforms of 20 Hz and 100 Hz. Additionally, the seismic instrument response files also were shared. The event and continuous waveform records could be downloaded by submitting a request through the web platform of the Earthquake Science Data Center of the Institute of Geophysics, CEA (www.esdc.ac.cn).     

CANOPEX: A Canadian hydrometeorological watershed database

Over the past few years, many international initiatives and collaborations were launched to improve and share knowledge in hydrology research. Large databases allowed finding patterns and relationships across regions and scales. This paper introduces the Canadian model parameter experiment (CANOPEX) database, which is adapted from the US MOPEX project data and methods. The CANOPEX database includes meteorological and hydrometric data as well as watershed boundaries for 698 basins. Two sets of basin‐averaged meteorological data (Maximum and minimum temperature and precipitation) are provided. The first dataset is directly taken from Environment Canada's weather stations whereas the second is extracted from the Natural Resources Canada gridded climate data product. Data are provided in MOPEX and Matlab formats. CANOPEX watersheds are well distributed over Canada, which allows investigating a variety of physiological and climatological conditions. The CANOPEX database can be used in a variety of hydrologic research projects such as climate change impact studies, model comparisons, multi‐modelling, ensemble streamflow prediction and model parameter estimation. CANOPEX could be used to generalize findings to other cold climate catchments as well as assess the robustness of research methodologies and procedures. Copyright © 2016 John Wiley & Sons, Ltd.     

格点降水资料在中国东部夏季降水变率研究中的适用性

本文使用1951~2010年PREC、CRU、APHRO和GPCC 4种格点降水资料,通过比较其与中国756站点观测降水资料在中国东部(105°E以东)夏季降水变率中的差异,检验和评估了它们的可靠性和适用性。结果表明:中国东部夏季降水变率的前3个主要模态分别是以江淮流域、长江流域和华北与东北南部为核心的经向多中心分布,有明显的年际和年代际变率特征,且干旱特征较洪涝更明显;长江流域夏季降水异常的主周期为3~7 a和20~50 a,而江淮流域和华北地区夏季降水异常的主周期则为准2 a和准10 a。另外,长江与江淮流域和华南地区分别在1970s末和1990s初发生了显著的年代际转变;4种格点降水资料都能很好地再现中国东部夏季降水的时空变率特征,但由于GPCC格点降水资料是基于更多的基站观测和更精细复杂的质量控制方案得到的,因此它具有更高的可靠性。     

一种基于场景合成和锚点约束的SAR目标检测网络

随着深度学习方法在计算机视觉领域的崛起,如何将其应用于具有全天时、全天候等优点的SAR图像也成为一大研究重点.相较于传统图像,SAR图像由于其难判读、应用人群较少等原因难以获得大量标注数据.本文提出一种基于场景合成和锚点约束的SAR图标检测方法.通过区域生长算法和阈值法对SAR车辆目标及其阴影进行分割,然后随机嵌入SAR复杂场景中的合理区域来合成目标检测数据集.针对SAR车辆目标的几何特性、图像分辨率参数,对Faster-RCNN中的锚点大小进行约束,减少不符合SAR车辆目标检测框尺寸的候选框,大量约简冗余计算,提升训练、测试效率及精度.     

1982-2012年北半球荒漠草原过渡带植被物候特征及其与气候因子的关系

基于GIMMS（global inventory modeling and mapping studies）NDVI 3g数据,在提取北半球荒漠草原过渡带每年植被物候期的基础上,研究了1982-2012年物候期的时间演化趋势及空间分异特征,并结合全球气候再分析资料,探讨了物候变化的气候驱动因素。结果表明：在1998年之前,荒漠草原过渡带植被物候期变化地区间差异较大,而在1998年之后,北半球荒漠草原过渡带生长季结束期整体提前,平均提前0.41 d/a;同时,除萨赫勒以外的各地区植被生长季长度普遍缩短,平均缩短0.88 d/a。植被物候期与气候因子的相关分析发现,荒漠草原过渡带植被物候变化受气候变化影响显著,且空间差异明显。在中高纬度地区,气温是限制植被活动的关键因子,温度升高可以促进生长季开始期的提前,而降水增加则会妨碍植被生长;在较低纬度地区,水分是影响植被活动的关键因素,高温造成的水分亏缺会导致植被生长季缩短。从植被物候期对各气候因子响应的时滞性来看,荒漠草原过渡带植被的物候期对气温变化的响应最迅速,对蒸散的响应存在一定的滞后性,而对降水的响应不存在时滞差异。     

3种再分析资料的高空温度与中国探空温度资料的对比:年平均特征

利用中国105站的探空资料以及NCEP/NCAR、ERA和JRA 3种再分析资料,采用线性趋势、标准差、相关系数、EOF分析等多种统计分析方法,对再分析资料年平均的高空温度在中国区域的可信度进行了分析.研究表明:在数值上,3种再分析资料的高空温度均小于探空资料的高空温度,NCEP/NCAR资料在对流层上层更接近于探空资料,ERA和JRA资料则在对流层中下层与探空资料更为接近;在描述年际变化和长期变化趋势方面,ERA资料在我国北方的对流层上层的再现能力较好,NCEP/NCAR资料在我国南方的对流层上层的再现能力较好,而3种再分析资料在对流层中下层的再现能力相当;在时空变化特征方面,NCEP/NCAR和ERA资料能较好地表现高空温度的年代际变化特征,而ERA和JRA资料则能较好地表现年际变化特征.     

多种再分析资料中热带气旋潜在生成指数分析

根据7套再分析资料计算的热带气旋潜在生成指数(GPI),分析了GPI对西北太平洋区域台风盛季(7—9月)热带气旋生成的表征能力。结果表明,虽然这些再分析资料计算得到的GPI的空间分布与观测的热带气旋生成特征都比较一致。但是,在时间变化上7套再分析资料计算的GPI对观测热带气旋生成的表征能力差异较大,其中ERA-40(欧州中期天气预报中心再分析资料)和MERRA(美国国家航空和航天局研究和应用再分析资料)的GPI与观测的相关系数较高。进一步分析表明,各套资料GPI之间时间变化的差异主要来自相对湿度,而ERA-40和MERRA的GPI与观测值有较高的相关系数,也与相对湿度有密切关系。     

Changes in temperature extremes based on a 6-hourly dataset in China from 1961–2005

Changes in Chinese temperature extremes are presented based on a six-hourly surface air temperature dataset for the period 1961--2005. These temperature series are manually observed at 0200, 0800, 1400, and 2000 Beijing Time (LST), and percentile based extreme indices of these time series are chosen for analysis. Although there is a difference in time among the different time zones across China, as more than 80% of the stations are located in two adjacent time zones, these indices for all the stations are called warm (cold) nights (0200 LST), warm (cold) mornings (0800 LST), warm (cold) days (1400 LST), and warm (cold) evenings (2000 LST), respectively for convenience. The frequency of the annual warm extremes has generally increased, while the frequency of the annual cold extremes has decreased, and significant changes are mainly observed in northern China, the Tibetan Plateau, and the southernmost part of China. Based on the national average, annual warm (cold) nights increase (decrease) at a rate of 5.66 (-5.92) d (10 yr)^-1, annual warm (cold) days increase (decrease) at a rate of 3.97 (-2.98) d (10 yr)^-1, and the trends for the annual warm (cold) mornings and evenings are 4.35 (-4.96) and 5.95 (-4.35) d (10 yr)^-1, respectively. For China as a whole, the increasing rates for the occurrence of seasonal warm extremes are larger in the nighttime (0200, 2000 LST) than these in the daytime (0800, 1400 LST), the maximal increase occurs at 2000 LST except in the summer and the minimal increase occurs at 1400 LST except in autumn; the maximal decrease in the occurrence of seasonal cold extremes occurs at 0200 LST and the minimal decrease occurs at 1400 LST.     

NCEP/NCAR再分析资料所揭示的全球季风降水变化

大气模式是研究气候变化的重要工具,当前的大气模式在模拟季风降水时均存在较大偏差,目前尚不清楚该偏差是来自模式环流场还是模式物理过程.再分析资料由于同化了各类观测和卫星资料,其大气环流近似可被视作是“真实”的.再分析资料中的降水场是在基本真实的环流场强迫下,由当前最先进的数值预报模式计算输出的.因此,再分析资料的降水场能...