一种多因子加权平均温度模型的构建方法

针对在地基GNSS水汽反演的过程中,天顶湿延迟转换为大气可降水量时如何建立精确的大气加权平均温度(Tm)模型的问题,该文在建立Tm模型前全面考虑了对Tm有显著影响的变量并选择最优回归子集。但分析发现,最优回归子集中各变量之间存在较强的相关性,这将会导致变量之间存在多重共线性,从而影响模型的稳定性和可靠性。选择2013—2015年相关气象数据作为变量并应用岭回归的方法削弱变量之间的多重共线性,建立稳定的多因子Tm回归模型。并利用该模型分别预测2016年1—12月、2019年1—7月的Tm,均方根误差分别为2.3 K和2.0 K,预测精度较高,这将为高精度的水汽反演奠定较好的数据基础。     

地形因子与小微地震相关度的空间异质性分析

针对传统最小二乘回归未能顾及数据的空间特性,且无法度量模型自变量与因变量相关性的空间变异特性的问题,本文提出利用地理加权回归方法分析小微地震频次与地形因子相关度的空间异质性。以四川地区的地震监测资料、DEM为实验数据,选取地形复杂度、坡度变率、坡向变率和地面曲率为自变量,地震发生频次为因变量,构建地理加权回归模型,并进行回归系数的空间变异分析。实验分析发现,地震频次与地形因子具有一定的相关性:地形复杂度与地震频次相关性最强;坡度变率、沟壑密度、剖面曲率与地震频次的相关性依次减弱;不同空间位置的地形因子和地震频次的相关性具有较明显的空间异质性。实验结果表明,地理加权回归可以有效地度量分析地震频次与地形因子相关度的空间异质性,研究结果可为地震及次生灾害的分析与预报提供辅助决策参考。     

刍议《重庆历史地图集》第二卷的编制

以《重庆历史地图集》第二卷编制的具体情况为例,对图集的开本设计、内容结构设计、封面和版式设计、整体设计、创新源,以及图集编制取得的经验进行了论述,为以后类似图集的编制提供参考。     

利用不同方法反演冬小麦叶面积指数研究

基于高光谱数据的叶面积指数监测是快速获取冬小麦叶面积指数的重要方法。为了探究回归方法和高光谱数据变换对冬小麦叶面积指数反演精度的影响,采用逐步回归和偏最小二乘回归方法,分别建立基于冬小麦拔节期冠层高光谱数据、一阶导数光谱数据、二阶导数光谱数据和对数光谱数据的叶面积指数多元线性回归模型。结果显示,导数和对数变换能够提高冬小麦LAI反演精度,以蓝紫光、绿光、红光和近红外波段建立的一阶导数光谱数据逐步回归模型最优,建立回归模型的决定系数R2为0.974,交叉验证的RMSE为0.131,可为冬小麦LAI估算的方法选择和数据处理提供依据和参考。     

利用TMI资料估计西北太平洋热带气旋强度的客观方法

利用2007—2009年热带降雨测量卫星(TRMM)微波成像仪(TMI)观测的亮温资料,建立一种西北太平洋热带气旋强度(Tropical Cyclone,TC)的估计模型,对2010年热带气旋进行独立估计试验,并对估计误差进行分析。结果表明:该模型对强度小于强台风TC的拟合效果较好,均方根误差约为5 m/s,平均绝对误差约为4 m/s;对强台风和超强台风TC的拟合误差较大,均方根误差分别为9.65和6.60 m/s,平均绝对误差分别为7.76和5.49 m/s;对强台风及以上强度的TC,模型的拟合误差在日(夜)间减小(增大),误差最小(大)值为6.00 m/s(11.96 m/s),说明估计值在日(夜)间偏大(小)。     

Risks of shifting baselines highlighted by anecdotal accounts of New Zealand's snapper (Pagrus auratus) fishery

Anecdotal data sources may constitute an important component of the information available about an exploited species, as record keeping may not have occurred until after exploitation began. Here, we aimed to fill any gaps in the exploitative history of the sparid snapper (Pagrus auratus), using social and historical research methods. Social research consisted of interviews with recreational fishers, focusing on the most and largest snapper they had caught. In addition, the diary‐logs of two recreational fishers were analysed. Historical research consisted of investigation of old books, photos, archives and unpublished sources unconventional to fishery science. Interviews with fishers demonstrated no or weak trends in snapper abundance or size, and were likely impeded by a lack of ability to detect change in a fish stock that may still be considered abundant. The fishers’ perception of change, however, largely reflected recent experiences (last c. 10 years), when biomass is understood to have increased, and mostly did not consider experiences before the 1980s. Alternatively, diary‐logs of fisher catch rates produced a pattern that matched formal stock assessments of snapper biomass, suggesting declines in abundance up until the 1990s and an increase in biomass after that time. Historical research, although more qualitative, had the ability to investigate periods where formal records were not kept and described a fishery vastly different from the current one. Snapper were easily caught, in great abundance and in unusual locations. Localised depletion of snapper was first noticed in the early 20th century, despite spectacular catches of snapper occurring after that time. Snapper behaviour was also likely different, with visual sightings of snapper by onlookers a common occurrence. Although predictions from stock assessment models are consistent with that of the anecdotes listed here (i.e., high biomass in the past), these anecdotes are valuable as they explain lost biomass in a perspective meaningful to all. This perspective may be valuable for managers trying to consider the non‐financial value of a shared fishery but, if unrecognised, represents a shifting baseline.     

川西低孔低渗储层参数评价

川西Y地区低孔低渗储层具有物性差、岩性多变、储集空间类型多样和孔隙结构复杂等特征,传统的储层参数测井评价方法在多方面受到限制,储层参数测井评价已经成为制约储层测井评价的主要因素。通过储层基本特征分析,基于储层物性、孔隙结构特征及概率理论,形成了一套系统的储层参数评价方法。实际应用表明,该套方法不仅大大提高了低孔低渗储层参数评价可靠性,而且扩大了模型的应用范围。     

Forced regressive shoreface sandstone from Himalayan foreland: Implications to early Himalayan tectonic evolution

Sedimentology and sequence stratigraphic analysis of the ∼ 31 Ma old marker White sandstone unit from the Subathu Sub-basin, NW Himalayan foreland, suggest it to be a forced regressive wedge (FRW) formed during the transition from the marine Subathu Formation to the continental Dagshai Formation. The FRW is bounded between the “Surf diastem” below and type 1 unconformity at the top and differs from RSME (regressive surface of marine erosion, occurring below) bounded FRWs described from other classical coastal/foreland settings. Correct identification of bounding surfaces of a FRW has an important implication to the estimation of rate of relative sea-level (RSL) fall. A faster rate of RSL fall, higher than the sedimentation rate, has been postulated for the erosion of the lower shoreface and RSME. Using the logged thickness of the Subathu/Dagshai transition zone including the White sandstone (bounded between the “Surf diastem” and unconformity), available chronology and eustatic sea-level fall (0.023 mm/year at 31 Ma), a higher RSL fall than the sedimentation rate (0.07 mm/year) has been inferred during the deposition of the White sandstone. Petrography of sandstones and their Sr and Nd isotopic compositions indicate a major provenance switch-over from dominant mafic/ultramafic to metamorphic source from White sandstone (∼ 31 Ma) onwards attesting the link between hinterland tectonics, provenance and forced regression. The provenance switch-over at 31 Ma was earlier inferred to be driven by proto-Himalayan thrust propagation in the foreland. Using a simple isostatic model, on the contrary, a mechanism of accelerated surface uplift (at a rate of > 0.10-0.15 mm/year) is suggested for both provenance change and forced regression.     

利用自然γ能谱测井资料确定粘土矿物的含量及其应用

根据粘土矿物含量的图版计算方法,计算了实际井段的粘土矿物含量,分析了高岭石、伊利石和蒙脱石与U、Th、K含量的相关关系,由此分析得到粘土矿物含量与放射性元素含量的多元线性关系,建立了粘土矿物含量计算的回归方程。通过对不同井段以及同一井段不同地层数据建立的回归方程分析,给出了具有代表性的线性回归方程模型。根据一个实际井段粘土矿物的计算,获得粘土矿物含量随深度的变化,其粘土矿物含量计算结果与原测井解释相吻合,同时与岩心分析误差在10%左右。     

基于GF5-AHSI遥感数据的横沙浅海水深反演

高光谱遥感水深反演是一种对传统水深测量方法的补充,具有方便、快捷、经济等突出优势。本文研究区位于上海横沙,属于典型滩涂浅水区,研究数据包括GF5-AHSI高光谱遥感数据和同时期的水深数据。通过数据变换和相关分析等方法提取建模参数,利用单波段比值模型、多元线性回归模型、最优标度回归模型和BP神经网络模型实现该区域水深反演,并对4种模型反演结果的准确性进行了验证和比较。研究发现：最优标度回归模型优于其他3种模型,R2达到了0.972,RMSE为0.47 m,适用于横沙浅海水深反演。