Lesht Barry M, Barbiero Richard P, Warren Glenn J
GDIT and Department of Earth and Environmental Sciences, University of Illinois at Chicago, 845 W. Taylor St., Chicago, IL 60607, USA.
GDIT, 1359 W. Elmdale Ave. Suite 2, Chicago, IL 60660, USA.
J Great Lakes Res. 2018 Aug;44(4):547-562. doi: 10.1016/j.jglr.2018.05.001.
The U.S. EPA's Great Lakes National Program Office (GLNPO) annual water quality survey (WQS) collects data at a relatively small number of stations in each lake. The survey was designed to measure conditions in the open-water regions of the lakes where an assumption of spatial homogeneity was thought likely to be met and the measured variables could be characterized by simple statistics. Here we use satellite observations to assess how well statistics based on samples collected in the GLNPO sampling network represent the lake-wide values of two variables, surface chlorophyll concentration and Secchi depth. We find strong linear relationships between the mean values calculated from the samples and the corresponding averages based on the subsets of the full satellite images. Although overall the means of the values from the sample locations agree well with means calculated from most of the non-coastal regions of the lakes, in terms of water depth, the GLNPO station averages best represent the regions of Lake Huron deeper than 30 m, of Lakes Michigan and Superior deeper than 90 m, and of Lake Ontario deeper than 60 m. When the lake regions are defined by distance offshore rather than by depth, the GLNPO station chlorophyll means in Lakes Huron, Ontario, and Superior are closest to the means for the area of the lakes > 10 km offshore. In Lake Michigan the closest correspondence is with the > 20 km offshore region. On a whole-lake basis in Lake Erie the GLNPO station chlorophyll averages are closest to the average calculated from the entire lake.
美国环境保护局(EPA)的五大湖国家项目办公室(GLNPO)年度水质调查(WQS)在每个湖泊中相对较少的站点收集数据。该调查旨在测量湖泊开阔水域的状况,在这些区域,空间均匀性的假设被认为可能成立,且测量变量可以用简单统计数据来表征。在这里,我们利用卫星观测来评估基于GLNPO采样网络中收集的样本的统计数据能多好地代表湖面叶绿素浓度和塞氏深度这两个变量的全湖值。我们发现,从样本计算得出的平均值与基于完整卫星图像子集的相应平均值之间存在很强的线性关系。尽管总体而言,样本位置的值的平均值与从湖泊大部分非沿海区域计算得出的平均值吻合良好,但就水深而言,GLNPO站点平均值最能代表休伦湖深度超过30米、密歇根湖和苏必利尔湖深度超过90米以及安大略湖深度超过60米的区域。当根据离岸距离而非深度来定义湖区时,休伦湖、安大略湖和苏必利尔湖的GLNPO站点叶绿素平均值最接近离岸>10公里的湖区平均值。在密歇根湖,最接近的对应区域是离岸>20公里的区域。在伊利湖全湖范围内,GLNPO站点叶绿素平均值最接近根据整个湖泊计算得出的值。