印度洋-太平洋暖池海域表层水温分析

利用1951~2003年HadISST资料集的表层海水温度（SST）资料,讨论了印度洋-西太平洋暖池（IPWP）海域,尤其是印度尼西亚贯穿流（ITF）及其周边海域SST的季节及年际变化的时空特征.研究结果表明,整个研究海域SST的年际变化均与ENSO相关,但印度洋与南海的响应特征与西太平洋的相反且不同步.前者海温变化滞后Nio3指数3~6个月,而热带太平洋西边界和ITF流经海域海温则超前1~3个月.沿ITF及其东印度洋出口,SST的年际变化规律不同于热带印度洋而与太平洋的相似,分析表明其在较大程度上受到ITF海洋桥的影响.在季节尺度上,印度洋和太平洋赤道海域SST的波动规律也有明显不同.以巽他岛弧（苏门答腊、爪哇和小巽他群岛）为界,从赤道西太平洋向西沿ITF流径,太平洋一侧SST的季节变化以0.5a周期的波动占主导,印度洋一侧则以1a周期占主导.     

无居民海岛开发适宜性评价指标体系的构建和在厦门海域的应用

无居民海岛开发适宜程度取决于对其生态重要性与开发可行性的综合评价.本研究从表征生态重要性的物种、生境和自然历史文化遗迹,以及反映开发可行性的自然、资源和社会经济条件构建无居民海岛开发适宜性评价指标体系;采用专家咨询法和层次分析法确定指标权重,在评价指标量化与标准化处理的基础上,建立生态重要性-开发可行性互斥矩阵确定无居民海岛的开发适宜程度.对厦门海域的实际应用研究结果表明,其18个无居民海岛的开发适宜性总体偏低,中度适宜开发的海岛共有7个,低适宜开发的海岛有8个,不适宜开发的海岛有3个,评价结果与国家海洋局2011年发布的可供开发利用无居民海岛名录中厦门海域部分基本吻合.作者建议对厦门海域中度适宜开发的无居民海岛可据其资源特征进行适度开发;对低适宜开发的无居民海岛应优先保护海岛生态,维护海岛自然属性;对不适宜开发的无居民海岛应重点加强海岛生态保护,必要时进行生境的重建和修复.     

苯并芘对褐菖鲉肝细胞DNA交联的影响

以褐菖鲉（Sebastiscus marmoratus）为实验材料,设定苯并芘[B（a）P]的注射剂量梯度分别为0.00、0.05、0.50、5.00 mg/kg,分别于实验的第0、3、6天对每一条鱼进行腹腔注射,于第1、4、7天取样,测定胆汁中的B（a）P及其代谢物含量.采用荧光检测法检测B（a）P染毒后褐菖鲉肝细胞DNA-DNA交联（DDC）交联系数的变化,并用KC l-SDS沉淀法检测DNA蛋-白质交联（DPC）交联系数的变化.结果表明：（1）经注射染毒后鱼体胆汁中的B（a）P及其代谢物含量蓄积程度存在着明显的剂量效应和时间效应;（2）B（a）P的注射剂量分别为0.05、0.50 mg/kg时各时间点褐菖鲉肝细胞DDC交联系数均无显著增大（p〉0.05）,在5.00 mg/kg剂量组,随着时间延长DDC交联系数逐渐增大,第4天即有显著差异（p〈0.05）,第7天差异极显著（p〈0.01）;（3）B（a）P诱导DPC的趋势与DDC类似,但0.5 mg/kg剂量组在第7天DPC交联系数即出现显著增大（p〈0.05）.这表明B（a）P对DPC的诱导高于DDC.这为进一步研究B（a）P的遗传毒性提供了依据,并为探讨B（a）P致癌的作用机理提供了新的信息.     

基于动态极坐标参数化的遥感影像匹配方法

图像匹配作为三维重建至关重要的环节,其精度直接影响了平差优化、正射校正等模块的精度。对于城镇、农场等特征密集型区域,特征距离小,相似性强,易于匹配图像;而针对草地、沙漠等特征不明显区域,特征距离大,如果使用特征点匹配的方法,严格阈值下难以获得足够数量的匹配对,放宽阈值又将引入较多误匹配对,这也是导致稀疏点云不够均匀的原因之一。在此场景下,本文提出了基于动态极坐标参数化的无人机正视影像匹配算法,首先对图像做极坐标参数变化,采用动态策略解决极轴方向采样不均匀的问题,使用最小二乘法对得到的极坐标影像对做位移方向上的匹配,匹配后得到的旋转量和平移量,将该结果和SIFT算法的结果做比较。本文设计了2组实验,即参数已知的解算实验和参数未知的解算实验,且每组实验进行3次。在同等配置的计算机上,对两张7360像素×5400像素,32位的影像,本文方法的位姿解算时间相比SIFT的时间减少约57%,二者求得的位姿差通常小于1%。结论表明二者的结果在精度上表现相当,在时间上明显优于SIFT算法,具有实际的应用价值。     

一种基于M估计的Robust-ELM滑坡位移预测方法

采用传统ELM算法进行滑坡位移预测时,其网络输出权值由最小二乘估计得出,导致ELM抗差能力较差,从而造成网络训练参数不准确。为此,将M估计与ELM相结合,提出一种基于M估计的RobustELM滑坡变形预测方法。该方法利用加权最小二乘方法来取代最小二乘法计算ELM输出权值,以减少滑坡监测数据中粗差对ELM预测的干扰。分别以链子崖、古树屋滑坡体为例,将Robust-ELM进行了单维、多维粗差的抵御性验证。结果表明,该方法能够有效降低粗差对预测的影响,具有良好的抗差能力。     

鲤鱼（Cyprinuscarpio L．）体重和体长QTL的定位

利用265个AFLP标记、127个微卫星分子标记、37个EST-SSR标记和16个RAPD标记（共445个标记）对大头鲤／荷包红鲤抗寒品系的F2代雌核发育群体44个个体进行基因型检测,构建鲤鱼（CyprinuscarpioL.）遗传连锁图谱;利用软件WinQTLCart2.5采用复合区间作图法对体重和体长两个性状进行了QTL定位分析。结果共检测到与两个与体重性状相关的QTL,分别定位到LG24连锁群（qBWh-24-1）和LG20连锁群（qBWh-20-1）上,可解释的表型变异分别为7.98％和20.05％;检测到两个与体长性状相关的QTL,分别位于LG2连锁群（qFS-2-1）和LG20连锁群（qFS-20-1）,可解释的表型变异分别为5.69％和12.69％,4个QTL的加性效应均为负值。     

Urban Construction Land Suitability Evaluation Based on Improved Multi-criteria Evaluation Based on GIS （MCE-GIS）： Case of New Hefei City,China

Urban construction land suitability evaluation （UCLSE） is a complex system engineering and the basis for rational use of the limited urban land resources in China. It has an important practical value on urban construction land use planning and management from the angle of methodology. As a widely used technique, traditional multi-criteria evaluation based on GIS （MCE-GIS）, is not suitable for UCLSE. This study develops an improved MCE-GIS method which could be more suitable for UCLSE based on urban complex ecological system theory and the summary of the shortcomings of traditional MCE-GIS. The improvements include three aspects： a composite evaluation index system rather than natural indexes alone, an index weight calculated by using fuzzy Analytic Hierarchy Process （AHP） method rather than the common AHP method, and the integrated overlay rule, which includes selecting the minimum value, weighted linear combination （WLC） and simple summation. The main advantage of this improved technique is that it can make UCLSE more comprehensive, more operational and more reasonable. It can provide a scientific basis for decision making in the planning and management of urban construction land use. The improved MCE-GIS system has been adopted in the New Hefei City, Anhui Province, China. Based on the results of UCLSE in New Hefei, three functional areas including construction-appropriate areas, construction-restricted areas and construction-forbidden areas could be worked out, in which 36.90% of the total study area could be developed as urban construction land and the remaining 63.10% should be protected as reserves land or as ecological land. Furthermore, the results can orovide scientific decision suooort for spatial planning and eco-environment nrotection in New Hefei.     

Estimation of regional evapotranspiration in alpine area and its response to land use change: A case study in Three-River Headwaters region of Qinghai-Tibet Plateau, China

Three-River Headwaters(TRH) region involved in this paper refers to the source region of the Changjiang(Yangtze) River,the Huanghe(Yellow) River and the Lancang River in China.Taking the TRH region of the Qinghai-Tibet Plateau as a case,the annual evapotranspiration(ET) model developed by Zhang et al.(2001) was applied to evaluate mean annual ET in the alpine area,and the response of annual ET to land use change was analyzed.The plant-available water coefficient(w) of Zhang’s model was revised by using vegetation-temperature condition index(VTCI) before annual ET was calculated in alpine area.The future land use scenario,an input of ET model,was spatially simulated by using the conversion of land use and its effects at small regional extent(CLUE-S) to study the response of ET to land use change.Results show that the relative errors between the simulated ET and that calculated by using water balance equation were 3.81% and the index of agreement was 0.69.This indicates that Zhang’s ET model based on revised plant-available water coefficient is a scientific and practical tool to estimate the annual ET in the alpine area.The annual ET in 2000 in the study area was 221.2 mm,11.6 mm more than that in 1980.Average annual ET decreased from southeast to northwest,but the change of annual ET between 1980 and 2000 increased from southeast to northwest.As a vast and sparsely populated area,the population in the TRH region was extremely unbalanced and land use change was concentrated in very small regions.Thus,land use change had little effect on total annual ET in the study area but a great impact on its spatial distribution,and the effect of land use change on ET decreased with increasing precipitation.ET was most sensitive to the interconversion between forest and unused land,and was least sensitive to the interconversion between cropland and low-covered grassland.     

基于CMONOC建立和评估中国大陆地壳运动速度场模型

针对现有速度场模型采用数据少、现势性不强的问题,详细论述了速度场模型建立的数据获取与处理方法,利用中国大陆地壳运动监测网络中260个连续运行基准站2011~2015年的连续观测数据,采用GAMIT/GLOBK与GMT软件,获得高精度的定位及速度成果,并基于上述速度成果采用克里金插值法构建了中国大陆地壳运动速度场1°×1°格网模型。最后利用IERS公布的中国区域7个IGS站实测速度以及全国范围内均匀分布的陆态网络区域站实测速度,对模型内插结果进行检核,验证了模型的现势性、精确性和可靠性。     

基于贝叶斯正则化BP神经网络的DEM趋势面逼近

趋势面从宏观上揭示了研究对象的特性,在各领域发挥着重要作用。BP神经网络可以对复杂系统进行无限逼近,进而进行预测。建立了基于贝叶斯正则化BP神经网络的数字高程模型趋势面,与二次多项式建立的数字高程模型趋势面进行比较分析,证明了该方法的可行性和有效性。