...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Environmental Science & Technology >Including Source Uncertainty and Prior Information in the Analysis of Stable Isotope Mixing Models
【24h】

Including Source Uncertainty and Prior Information in the Analysis of Stable Isotope Mixing Models

机译：在稳定同位素混合模型的分析中包括源不确定性和先验信息

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Stable isotope mixing models offer a statistical framework for estimating the contribution of multiple sources (such as prey) to a mixture distribution. Recent advances in these models have estimated the source proportions using Bayesian methods, but have not explicitly accounted for uncertainty in the mean and variance of sources. We demonstrate that treating these quantities as unknown parameters can reduce bias in the estimated source contributions, although model complexity is increased (thereby increasing the variance of estimates). The advantages of this fuly Bayesian approach are particularly apparent when the source geometry is poor or sample sizes are small. A second benefit to treating source quantities as parameters is that prior source information can be included. We present findings from 9 lake food-webs, where the consumer of interest (fish) has a diet composed of 5 sources: aquatic insects, snails, zooplankton, amphipods, and terrestrial insects. We compared the traditional Bayesian stable isotope mixing model with fixed source parameters to our fully 8ayesian model-with and without an informative prior. The informative prior has much less impact than the choice of modd-the traditional mixing model with fixed source parameters estimates the diet to be dominated by aquatic insects, while the fully Bayesian model estimates the diet to be more balanced but with greater importance of zooplankton. The findings from this example demonstrate that there can be stark differences in inference between the two model approaches, particularly when the source geometry of the mixing model is poor. These analyses also emphasize the importance of investing substantial effort toward characterizing the variation in the isotopic characteristics of source pools to appropriately quantify uncertainties in their contributions to consumers in food webs.
机译：稳定的同位素混合模型提供了一个统计框架，用于估算多种来源（例如猎物）对混合物分布的贡献。这些模型的最新进展已使用贝叶斯方法估计了源比例，但并未明确考虑源均值和方差的不确定性。我们证明，尽管模型复杂性增加了（从而增加了估计的方差），但是将这些数量视为未知参数可以减少估计的源贡献中的偏差。当源几何形状较差或样本量较小时，这种完整的贝叶斯方法的优点尤其明显。将源数量视为参数的第二个好处是可以包含先前的源信息。我们提供9种湖泊食物网的发现，其中感兴趣的消费者（鱼）的饮食由5种来源组成：水生昆虫，蜗牛，浮游动物，两栖动物和陆生昆虫。我们将带有固定源参数的传统贝叶斯稳定同位素混合模型与带有或不带有先验信息的完全8ayesian模型进行了比较。信息性先验的影响远小于modd的选择-具有固定源参数的传统混合模型估计饮食以水生昆虫为主导，而完全的贝叶斯模型则估计饮食更为平衡，但对浮游动物的重要性更大。该示例的发现表明，两种模型方法之间的推断可能存在明显差异，尤其是在混合模型的源几何形状较差时。这些分析还强调了进行大量工作以表征源库同位素特征变化的重要性，以适当地量化其对食物网中消费者的贡献的不确定性。

著录项

  • 来源
    《Environmental Science & Technology》 |2010年第12期|P.4645-4650|共6页
  • 作者

    EROC J. WARD; rnBRICE X. SEMMENS; DANIEL E. SCHINDLER;

  • 作者单位

    Northwest Fisheries Science Center, National Marine Fisheries Service, National Oceanic & Atmospheric Administration, Seattle, Washington;

    rnNorthwest Fisheries Science Center, National Marine Fisheries Service, National Oceanic & Atmospheric Administration, Seattle, Washington;

    rnSchool of Aquatic and Fishery Sciences, University of Washington, Seattle, Washington;

  • 收录信息 美国《科学引文索引》(SCI);美国《工程索引》(EI);美国《生物学医学文摘》(MEDLINE);美国《化学文摘》(CA);
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

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