Subjects: Geosciences >> Marine Sciences submitted time 2023-05-31
Abstract： Quantifying eddy mixing in the ocean is a hot and tough problem in the area of physical oceanography. Based on the theory of effective diffusivity, the present study investigated the stirring effects of geostrophic turbulence that lead to stretching, distorting, deforming, and folding of tracer contours. These changes are then related to the efficiency of turbulent mixing. Results show that under the stirring effect of geostrophic turbulence, the length of tracer contour can be quickly elongated and fine-scale tracer filaments and fronts are also generated. This fractal elongation of tracer contour, about 10-20 times longer than the original length, is the dominant contributor to the mixing efficiency, whereas the gradient enhancement associated with filament and front generations only plays a secondary role. On the other hand, fine-scale features are smoothed out by small-scale diffusivity which eventually suppresses the increase of contour length and the generation of tracer filaments. This imposes an upper bound of the mixing efficiency when the stirring and smoothing effects are in a dynamical balance. Through a ‘box-counting’ method, the fractal dimension of tracer contour is also found between 1.4~1.6, indicating a geometric dimension lies somewhere between 1D and 2D. Due to the limitation of data resolution, contour length and thus mixing efficiency may be underestimated. Finally, the present study made an empirical relation between the fractal dimension and mixing efficiency, providing an opportunity for estimating mixing efficiency through a well-developed pattern recognition technique in remote sensing, and a new way of diagnosing ocean mixing and its parameterization.
Subjects: Geosciences >> Geography submitted time 2020-01-06 Cooperative journals: 《干旱区地理》
Abstract：青土湖区域属于绿洲—荒漠过渡带，生态系统脆弱，极易发展为荒漠。梭梭、白刺和芦苇为青土湖区域的植被优势种，对其生态系统稳定与健康发展起着关键作用。以青土湖区域梭梭、白刺和芦苇为研究对象，利用空间分辨率为0.5 m的高分辨率遥感影像Worldview-2，采用辅以纹理特征的面向对象分类方法，提取梭梭和白刺的冠幅面积以及芦苇的分布面积；根据野外试验数据，建立梭梭和白刺地上生物量与冠幅面积、芦苇地上生物量与分布面积关系模型。利用关系模型、冠幅面积以及分布面积对青土湖区域植被优势种地上生物量进行了估算，实现了植被优势种地上生物量估算由“点”到“面”的转换。结果表明：（1） 采取辅以纹理特征的面向对象分类方法取得了较高的分类精度，总体Kappa系数为87.9%，总体精度达到91.3%。（2） 研究区植被优势种地上生物量总量为3.17×103 t，其中梭梭地上生物量为0.54×103 t，白刺地上生物量为0.90×103 t，芦苇地上生物量为1.73×103 t，地上生物量芦苇>白刺>梭梭。该研究可以为深入研究青土湖区域生态恢复与碳储量提供参考。
Subjects: Geosciences >> Other Disciplines of Geosciences submitted time 2019-09-11 Cooperative journals: 《干旱区地理》
Abstract：基于甘肃省28个气象站点1967—2017年的逐日气象数据，采用Penman-Monteith 公式和作物系数计算了小麦的需水量，结合美国农业部土壤保持局推荐的方法计算了有效降水量，进一步得出小麦缺水量，并分析了小麦需水量与缺水量的变化趋势和空间分布特征，探讨了小麦需水量影响因子重要程度。结果表明：1967—2017年甘肃省春、冬小麦全生育期的年平均需水量分别为517.03 mm和436.70 mm，年平均缺水量分别为468.24 mm和301.54 mm；在时间上，51 a来春小麦种植区内的需水量与缺水量整体上无明显的趋势变化，而冬小麦种植区内的需水量与缺水量整体上呈明显上升趋势；在空间上，春小麦种植区内需水量和缺水量大致由西北向东南递减，冬小麦种植区内的需水量和缺水量大致由西向东递减。甘肃省小麦生育期内日照时数和日平均气温是影响小麦需水量的主要因素。
Subjects: Geosciences >> Other Disciplines of Geosciences submitted time 2018-10-23 Cooperative journals: 《干旱区地理》
Abstract： 纹理特征作为一种非光谱信息能够增强地物之间的特征差异，这对于高分辨率遥感影像的地物提取有着重要意义。本文以青土湖为研究区，以Worldview-2影像为数据源，通过引入权重因子定义联合概率函数来确定最佳窗口尺度，利用灰度共生矩阵提取最佳窗口尺度下的纹理特征，将其与原始遥感影像合成，采用最小二乘支持向量机（LSSVM）进行地物提取，将提取结果与仅利用光谱信息的支持向量机（SVM）提取结果、辅以纹理特征的SVM提取结果对比分析。结果表明，本文方法可以更加快速准确地提取青土湖地物，精度高达85.86%，优于仅利用光谱信息的SVM的 65.13%，辅以纹理特征的SVM的73.45%，可为地物破碎的干旱区高分辨率遥感影像地物提取提供有益借鉴。
Hits
YES
Hits
Downloads
Comment