• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on LiDAR Applications for Assessing Forest Ecosystems >An evaluation of the field sampling design of the first operational LiDAR based site quality survey of radiata pine plantations in South Australia.
【24h】

An evaluation of the field sampling design of the first operational LiDAR based site quality survey of radiata pine plantations in South Australia.

机译：南澳大利亚辐射松木种植园第一运行激光雷达现场采样设计的评价。

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The first operational LiDAR based site quality survey of radiata pine plantations was conducted in South Australia in 2009. All nine and ten year old plantations, 28 sites in all, covering 9,365 hectare scattered across an area of 10,000 km~2, were assessed. Area-based methods were applied and field data were collected to calibrate new prediction models for stand volume (i.e. the criterion of site quality). Decisions with regard to the field sampling design were to some extent guided by prior research, but were conservative so as to ensure robustness of the results. This study evaluated two key aspects of the sampling design - sampling intensity and sample selection method. It also posed the question whether the decision not to reuse pre-existing data and models was justified in light of the survey results. Firstly the precision of prediction models fitted to three independent datasets of different size were compared. Secondly a simulation experiment was conducted to test the sensitivity of model precision to a range of sample sizes and sample selection methods (random, stratified random and systematic sampling). This experiment was run with and without inclusion of pre-existing data. Results indicated that the sampling intensity applied in the survey could have been significantly reduced without significant loss of model accuracy. Models were only sensitive to the sample selection method when samples were small. The new prediction models proved to be significantly different from any pre-existing models. However, simulation results suggested that when new data are scarce multi-campaign models calibrated using new and pre-existing data may offer increased prediction accuracy. Given the high cost of field data collection these findings have practical importance for future surveys.
机译：2009年，南澳大利亚南澳大利亚的基于radiata杉木种植园的首次运行LIDAR的网站质量调查。所有九年和十岁的种植园都有28个，占地面积9,365公顷，占地面积在10,000公里〜2的面积。应用了基于地区的方法，收集了现场数据以校准用于架构的新预测模型（即站点质量的标准）。关于现场采样设计的决策在某种程度上是以先前的研究为引导的，但却是保守的，以确保结果的鲁棒性。本研究评估了采样设计 - 采样强度和样品选择方法的两个关键方面。它还提出了根据调查结果的决定不使用预先使用的数据和模型的决定。首先，比较了安装在不同大小的三个独立数据集的预测模型的精度。其次，进行了模拟实验，以测试模型精度到一系列样品尺寸和样品选择方法的敏感性（随机，分层随机和系统采样）。此实验与预先存在的数据一起运行和不包含预先存在。结果表明，在调查中应用的采样强度可能会显着减少，而无需显着损失模型准确性。当样品小时，模型对样品选择方法仅敏感。新预测模型被证明与任何预先存在的模型有很大差异。然而，仿真结果表明，当使用新的数据均匀校准的新数据时，使用新的和预先存在的数据可以提供更高的预测精度。鉴于现场数据收集的高成本，这些调查结果对于未来的调查具有实际重要性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号