城镇绿地植被固碳量遥感测算模型的设计
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金资助项目(40671177)


A design of carbon-sink model of urban landscape vegetation driven by remote sensing
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    城镇绿地是增加城市碳汇的积极因素之一。但城镇绿地植被结构和分布的极高空间异质性、影响植物生产力的环境压力因子的极高空间异质性等,使城镇绿地碳汇水平估算存在很大的不确定性。为此,提出了完全由遥感数据驱动的城镇绿地植被固碳量测算模型。它以分树种单株测算来适应城镇绿地植被结构和分布等的高异质性;以小尺度提取影响植物生产力的主要环境压力和管养模式因子,来适应这些因子空间分布的高异质性。该模型可以用于自动测算城镇绿地植被地上干生物量和地上净第一生产力,它的提出对于评价城镇绿地植被的碳捕获能力和储量分布、碳汇水平、以及由此产生的对城镇地区碳循环和生态承载力等的定量影响具有重要意义。为了论证该技术框架中一些关键技术的可行性,进行了局部试验,并取得了一些进展。

    Abstract:

    Urban landscape vegetation is one of the positive factors in absorbing carbon in urban area. However, there remain significant uncertainties for estimating carbon sink in relation to urban landscape vegetation due to extremely high heterogeneities in structure and distribution of both urban landscape vegetation and environmental stresses on plant productivity. In this paper, we tend to discuss the necessities and difficulties of measuring captured carbon by urban landscape vegetation and estimating plant carbon sink by means of the technique of remote sensing. Then, a model driven by remote sensing named MMCC (Model of measuring captured carbon) is proposed. With MMCC, it is most likely that both the above-ground vegetation biomass (AVB) and the above-ground net primary productivity (ANPP) were measured automatically. Here is the summary about the solutions by MMCC to some main difficulties while assessing the urban plant carbon sink.① both the present sum and annual increase of captured carbon by urban landscape vegetation could be obtained with the algorithms of measuring AVB and ANPP respectively; ② the continuous quantification in timeliness could be fulfilled by making annual change of ANPP in the coming years predictable. The prediction is supported by such data as certain plant species and plant canopy diameters because they can be used as proxies for a certain plant′s age and its ANPP in this age when the stress over the plant from its surroundings is considered; ③ considering the high heterogeneity of plant species, distribution and structure of urban landscape vegetation, the basic object measured by MMCC is an individual plant rather than a plot of ecological community, as in the most existing models of assessing plant carbon sink. To adapt to the same high heterogeneity of spatial distribution of those environmental stress factors having their effects on ANPP, a small-scale (25m×25m) block is scheduled to sampling these factors; ④ with the development of MMCC fully driven by remote sensing and GIS data, the continuous quantification in spatiality can be actualized. The quick quantification can also be actualized by making all the parameters to drive MMCC acquirable by computer. Namely, all the parameters, including plant and environmental stress factors need to be collected in the field only in the modeling phase, but can be produced automatically by image recognition in its running phase. In order to demonstrate feasibilities of some key technologies in MMCC, parts of the most difficulties were tested with some progresses achieved. For example, in identifying the nature of plant species, 14 new descriptors inductive of spectrum, texture and shape signatures have been designed. These descriptors should be up to such requirements as possessing a true physical implication relating to geometric or ecological significance, having a relatively steady segmentation threshold and being less sensitive to the differences among image types or the environmental conditions during image acquisition, and so on. This study uses decision tree with four descriptors to identify plant species and yields an error rate no more than 5.8% while comparing 25.9% by using traditional properties. Moreover, a new conception of Normalized Difference Umbra Index (NDUI) has been achieved in the present study. It is demonstrated that NDUI is a robust index for extracting pixels of trees planted at darken area and thus helpful for repairing their brightness. Another notable test is to get diameter of plant canopies with image recognition. As a result, it is initially determined that the area covered by vegetable can be plotted out into individual canopies and the average diameter of these canopies in a 25m×25m sampling block can be calculated as the image resolution is better than one meter and plant canopy density is not saturated yet. The correlation between the ratio of perimeter to area and diameter of plant canopy can be used in the calculation. In short, MMCC of the present study is proved to be applicable for assessing quantity and distribution of captured carbon by urban landscape vegetation and its quantitative effect on carbon circulation and ecological capacity in a city.

    参考文献
    相似文献
    引证文献
引用本文

周坚华*,胡永红,周一凡,俞立中.城镇绿地植被固碳量遥感测算模型的设计.生态学报,2010,30(20):5653~5665

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数: