西南印度洋中脊热液产物稀土元素组成变化及其来源

对西南印度洋中脊热液区不同热液产物稀土元素(REE)进行了分析,探讨了热液产物形成过程中稀土元素组成变化及其来源。研究结果表明:不同热液产物稀土元素总量变化范围从3.47×10-7到4.80×10-5,轻重稀土比值(LREE/HREE)从2.06到6.16,表明轻重稀土有较大程度分异,δEu异常(δEu=0.86~3.88)和δCe异常(δCe=0.40~0.86)显示热液产物中REE呈Eu富集和Ce亏损特征。稀土元素球粒陨石标准化模式呈现两种类型:(1)呈轻微富集LREE的平坦模式,REE大于2×10-5;(2)呈显著富集LREE和正Eu异常模式,REE小于5×10-7。模式1类似于洋壳火山岩REE配分模式,而模式2与西南印度洋中脊黑烟囱REE模式相似,也与典型洋中脊热液喷口流体和硫化物LREE富集和正Eu异常模式类似。热液产物中稀土元素含量变化和模式特征以及Mg与LREE极强正相关关系可能反映了西南印度洋中脊硫化物形成在热液流体与海水混合沉淀的初始阶段,后期经历了广泛的热液流体再循环和海水蚀变过程。     

岭估计及其应用

最小二采估计是最小无偏估计。平差后,观测值的残差平方和为最小。在法方程系数矩阵A^TA接近于单位矩阵(矩阵状态良好)的情况下,未知参数的最小二乘估值是可靠的。但是,在法方程系数阵处于病态的情况下,观测值残差平方和最小并不能保证未知参数估值的方差也小。岭估计能缩短未知参数估值与其真值之间的距离,即减少未知参数估值的方差。     

Geochemistry of volcanic glass from Oligocene detrital sediments at DSDP Site 296, Kyushu Palau Ridge: Interpreting the magmatic evolution of the early northern Izu–Bonin–Mariana Island Arc

Understanding the petrologic and geochemical evolution of island arcs is important for interpreting the timing and impacts of subduction and processes leading to the formation of a continental crust. The Izu–Bonin–Mariana (IBM) Arc, western Pacific, is an outstanding location to study arc evolution. The IBM first arc (45–25 Ma) followed a period of forearc basalt and boninite formation associated with subduction initiation (52–45 Ma). In this study, we present new major and trace element data for the IBM first arc from detrital glass shards and clasts from DSDP Site 296, located on the northernmost Kyushu Palau Ridge (KPR). We synthesize these data with published literature for contemporaneous airfall ash and tephra from the Izu–Bonin forearc, dredge and piston core samples from the KPR, and plutonic rocks from the rifted eastern KPR escarpment, locations which lie within or correlate with KPR Segment 1 of Ishizuka, Taylor, Yuasa, and Ohara (2011). Our objective is to test ways in which petrologic and chemical data for diverse igneous materials can be used to construct a complete picture of this section of the Oligocene first arc and to draw conclusions about its evolution. Important findings reveal that widely varying primary magmas formed and differentiated at various depths at this location during this period. Changes in key trace element ratios such as La/Sm, Nb/Yb, and Ba/Th show that mantle sources varied in fertility and in the inputs of subducted sediment and fluids over time and space. Plutonic rocks appear to be related to early K‐poor dacitic liquids represented by glasses sampled both in the forearc and volcanic fronts. An interesting observation is that the variation in magma compositions in this relatively small segment encompasses that inferred for the IBM Arc as a whole, suggesting that sampling is a key factor in inferring temporal, across‐arc, and along‐strike geochemical trends.     

基于机器学习的华北气温多模式集合预报的订正方法

模式预报的订正是决定局地天气预报结果的一个重要步骤,基于机器学习的后处理模型近年来开始崭露头角。本文发展了基于岭回归（Ridge）、随机森林（Random Forest,RF）和深度学习（Deep Learning,DL）的3种后处理模型,基于中国气象局（CMA）的BABJ模式、欧洲中期天气预报中心（ECMWF）的ECMF模式、日本气象厅（JMA）的RJTD模式和NCEP的KWBC模式这4个数值天气预报模式2014年2月至2016年9月（训练期）近地面2 m气温预报和实况资料确定各模型参数,进而对2016年10月至2017年9月（预报期）华北地区（38°N~43°N,113°E~119°E）的逐日地面2 m气温预报进行了多模式集合预报分析。采用均方根误差对预报效果进行评估,这3种后处理模型的预报效果和4个数值天气预报模式以及通常的多模式集合平均（Ensemble Mean,EMN）的预报效果的对比表明:1）随着预报时长增加,4个数值预报模式及各种后处理模型的均方根误差均呈上升趋势;但区域平均而言,Ridge、RF和DL的预报效果在任何预报时长上都明显优于EMN和单个天气预报模式;特别是前几天的短期预报DL的预报效果更好,中后期预报Ridge的预报效果略好。2）华北地区的东南部均方根误差较小,其余格点上均方根误差较高,从空间分布而言,DL的订正预报效果最好,3种机器学习模型的误差在1.24~1.26℃之间,而EMN的误差达1.69℃。3）夏季各种方法的预报效果都较好,冬季预报效果都较差;但是Ridge、RF和DL的预报效果明显优于EMN,这3种模型预报的平均均方根误差在2.15~2.18℃之间,而EMN的平均均方根误差达2.45℃。     

投影寻踪回归技术在降水量预测中的应用

提出了应用投影寻踪回归（ＰＰＲ）技术预测降水量的建模思想及其ＳＭＡＲＴ算法的实现。三个地区应用ＰＰＲ进行降水量预测建模,模型的拟合及预留预测检验结果均好于逐步回归建模的相应结果。     

西南印度洋岩浆补给特征研究:来自洋壳厚度的证据

西南印度洋中脊为典型的超慢速扩张洋中脊,其岩浆补给具有不均匀分布的特征.洋壳厚度是洋中脊和热点岩浆补给的综合反映,因此反演洋壳厚度是研究大尺度洋中脊和洋盆岩浆补给过程的一种有效方法.本文通过对全球公开的自由空气重力异常、水深、沉积物厚度和洋壳年龄数据处理得到剩余地幔布格重力异常,并反演西南印度洋地区洋壳厚度,定量地分析了西南印度洋的洋壳厚度分布及其岩浆补给特征.研究发现,西南印度洋洋壳平均厚度7.5 km,但变化较大,标准差可达3.5 km,洋壳厚度的频率分布具有双峰式的混合偏态分布特征.通过分离双峰统计的结果,将西南印度洋洋壳厚度分为0~4.8 km的薄洋壳、4.8~9.8 km的正常洋壳和9.8~24 km的厚洋壳三种类型,洋中脊地区按洋壳厚度变化特征可划分为7个洋脊段.西南印度洋地区薄洋壳受转换断层控制明显,转换断层位移量越大,引起的洋壳减薄厚度越大,减薄范围与转换断层位移量不存在明显相关性.厚洋壳主要受控于该区众多的热点活动,其中布维热点、马里昂热点和克洛泽热点的影响范围分别约340 km,550 km和900 km.Andrew Bain转换断层北部外角形成厚的洋壳,具有与快速扩张洋中脊相似的转换断层厚洋壳特征.

     

Snow measurement by GPS interferometric reflectometry: an evaluation at Niwot Ridge,Colorado

Snow is a critical storage component in the hydrologic cycle, but current measurement networks are sparse. In addition, the heterogeneity of snow requires surveying larger areas to measure the areal average. We presented snow measurements using GPS interferometric reflectometry (GPS‐IR). GPS‐IR measures a large area (~100 m²), and existing GPS installations around the world have the potential to expand existing snow measurement networks. GPS‐IR uses a standard, geodetic GPS installation to measure the snow surface via the reflected component of the signal. We reported GPS‐IR snow depth measurements made at Niwot Ridge, Colorado, from October 2009 through June 2010. This site is in a topographic saddle at 3500 m elevation with a peak snow depth of 1.7 m near the GPS antenna. GPS‐IR measurements are compared with biweekly snow surveys, a continuously operating scanning laser system and an airborne light detection and ranging (LIDAR) measurement. The GPS‐IR measurement of peak snowpack (1.36–1.76 m) matches manual measurements (0.95–1.7 m) and the scanning laser (1.16 m). GPS‐IR has RMS error of 13 cm (bias = 10 cm) compared with the laser, although differences between the measurement locations make comparison imprecise. Over the melt season, when the snowpack is more homogenous, the difference between the GPS‐IR and the laser is reduced (RMS = 9 cm, bias = 6 cm). In other locations, the GPS and the LIDAR agree on which areas have more or less snow, but the GPS estimates more snow on the ground on tracks to the west (1.58 m) than the LIDAR (1.14 m). Copyright © 2011 John Wiley & Sons, Ltd.     

Mechanisms of magmatic gas loss along the Southeast Indian Ridge and the Amsterdam -St. Paul Plateau

New analyses of He, Ne, Ar and CO₂ trapped in basaltic glasses from the Southeast Indian Ridge (Amsterdam-St. Paul (ASP) region) show that ridge magmas degas by a Rayleigh distillation process. As a result, the absolute and relative noble gas abundances are highly fractionated with ⁴He/⁴⁰Ar* ratios as high as 620 compared to a production ratio of ∼3 (where ⁴⁰Ar* is ⁴⁰Ar corrected for atmospheric contamination). There is a good correlation between ⁴He/⁴⁰Ar* and the MgO content of the basalt, suggesting that the amount of gas lost from a particular magma is related to the degree of crystallization. Fractional crystallization forces oversaturation of CO₂ because CO₂ is an incompatible element. Therefore, crystallization will increase the fraction of gas lost from the magma. The He-Ar-CO₂-MgO-TiO₂ compositions of the ASP basalts are modeled as a combined fractional crystallization-fractional degassing process using experimentally determined noble gas and CO₂ solubilities and partition coefficients at reasonable magmatic pressures (2-4 kbar). The combined fractional crystallization-degassing model reproduces the basalt compositions well, although it is not possible to rule out depth of eruption as a potential additional control on the extent of degassing. The extent of degassing determines the relative noble gas abundances (⁴He/⁴⁰Ar*) and the ⁴⁰Ar*/CO₂ ratio but it cannot account for large (>factor 50) variations in He/CO₂, due to the similar solubilities of He and CO₂ in basaltic magmas. Instead, variations in CO₂/³He (≡C/³He) trapped in the vesicles must reflect similar variations in the primary magma. The controls on C/³He in mid-ocean ridge basalts (MORBs) are not known. There are no obvious correlated variations between C/³He and tracers of mantle heterogeneity (³He/⁴He, K/Ti etc.), implying that the variations in C/³He are not likely to be a feature of the mantle source to these basalts. Mixing between MORB-like sources and more enriched, high ³He/⁴He sources occurs on and near the ASP plateau, resulting in variable ³He/⁴He and K/Ti compositions (and many other tracers). Using ⁴He/⁴⁰Ar* to track degassing, we demonstrate that mixing systematics involving He isotopes are determined in large part by the extent of degassing. Relatively undegassed lavas (with low ⁴He/⁴⁰Ar*) are characterized by steep ³He/⁴He-K/Ti mixing curves, with high He/Ti ratios in the enriched magma (relative to He/Ti in the MORB magma). Degassed samples (high ⁴He/⁴⁰Ar*) on the other hand have roughly equal He/Ti ratios in both end-members, resulting in linear mixing trajectories involving He isotopes. Some degassing of ASP magmas must occur at depth, prior to magma mixing. As a result of degassing prior to mixing, mixing systematics of oceanic basalts that involve noble gas-lithophile pairs (e.g. ³He/⁴He vs. ⁸⁷Sr/⁸⁶Sr or ⁴⁰Ar/³⁶Ar vs. ²⁰⁶Pb/²⁰⁴Pb) are unlikely to reflect the noble gas composition of the mantle source to the basalts. Instead, the mixing curve will reflect the extent of gas loss from the magmas, which is in turn buffered by the pressure of combined crystallization-degassing and the initial CO₂ content.     

The geochemical controls on vent fluids from the Lucky Strike vent field, Mid-Atlantic Ridge

Hydrothermal vent fluids were collected from the Lucky Strike site at 37°17′N on the Mid-Atlantic Ridge in both 1993 and 1996. Seven vents were sampled with the DSV Alvin in 1993 and six vents were sampled in 1996 using the ROV Jason during the LUSTRE '96 Cruise. As three of the vents were sampled in both 1993 and in 1996, a time series of vent fluid chemistry is also reported. Measured temperatures ranged from 202 to 333°C at the 1618–1726 m depth of the vent field, which is located on Lucky Strike Seamount. These fluids are either equal to or less than the local bottom seawater in chlorinity. While the range in fluid compositions at Lucky Strike is generally within that observed elsewhere, the unusual aspects of the fluid chemistries are the relatively high pH and low Fe, Mn, Li and Zn. We attribute this, as well as an usually low Sr/Ca ratio, to reaction with a highly altered substrate. The high Si and Cu contents suggest a deep, as well as hot, source for these fluids. The fluid compositions therefore suggest formation by super-critical phase separation at a depth not less than 1300 m below the seafloor, and reaction with a relatively oxic, and previously altered, substrate. There is temporal variability in some of the vent fluid compositions as Li, K, Ca and Fe concentrations have increased in some of the vents, as has the Fe/Mn (molar) ratio, although the chlorinities have remained essentially constant from 1993 to 1996. While there is not a simple relationship between vent fluid compositions (or temperatures) and distance from the lava lake at the summit of the seamount, the vent fluids from many of the vents can be shown to be related to others, often at distances >200 m. The most southeasterly vents (Eiffel Tower and the Marker/Mounds vents) are distinct in chlorinity and other chemical parameters from the rest of the vents, although closely related to each other within the southeastern area. Similarly all of the vents not in this one area, appear closely related to each other. This suggests one or two source fluids for many of the vents, as is also inferred to be the case at TAG, but which is in contrast to observations on faster spreading ridges. This may suggest inherently different plumbing for hydrothermal systems at slower versus faster spreading ridges.     

Using saline tracers to evaluate preferential recharge in fractured rocks,Floyd County,Virginia, USA

Potassium chloride (KCl) and potassium bromide (KBr) tracers were used to explore the role of geologic structure on groundwater recharge and flow at the Fractured Rock Research Site in Floyd County, Virginia, USA. Tracer migration was monitored through soil, saprolite, and fractured crystalline bedrock for a period of 3 months with chemical, physical, and geophysical techniques. The tracers were applied at specific locations on the ground surface to directly test flow pathways in a shallow saprolite and deep fractured-rock aquifer. Tracer monitoring was accomplished with differential electrical resistivity, chemical sampling, and physical monitoring of water levels and spring discharge. KCl, applied at a concentration of 10,000 mg/L, traveled 160 m downgradient through the thrust fault aquifer to a spring outlet in 24 days. KBr, applied at a concentration of 5,000 mg/L, traveled 90 m downgradient through the saprolite aquifer in 19 days. Tracer breakthrough curves indicate diffuse flow through the saprolite aquifer and fracture flow through the crystalline thrust fault aquifer. Monitoring saline tracer migration through soil, saprolite, and fractured rock provided data on groundwater recharge that would not have been available using other traditional hydrologic methods. Travel times and flowpaths observed during this study support preferential groundwater recharge controlled by geologic structure.