ISWC OpenIR  > 水保所2018--2019届毕业生论文
水蚀风蚀交错区典型植被叶面积指数动态变化及其对冠层截留量的影响
牛小桃
Subtype硕士
2019-05-24
Degree Grantor中国科学院大学
Place of Conferral北京
Degree Name工程硕士
Keyword叶面积指数 鱼眼摄像机 农作物 灌木 降雨再分配
Abstract

叶面积指数(LAI)的自动准确测量可以为相关生态过程研究提供关键参数。本研究首先采用光学鱼眼相机(DHP)和LAI-2200测量了黄土高原北部典型植被柠条、沙柳、紫花苜蓿、长芒草、大豆和玉米不同生长时期冠层LAI,通过比较两者测量结果去验证光学鱼眼镜头的准确性。其次通过在玉米、大豆和柠条样地固定安装鱼眼摄像机定时拍摄冠层照片,并运用图像处理软件CAN-EYE分析获取LAI动态变化,同期使用LAI-2200LI-3000ADHP仪器以每周一次的频率测量LAI,对比校正摄像机数据。最后将鱼眼摄像机法测量LAI应用于灌木冠层截留影响机制的研究,以柠条和沙柳群落为研究对象,通过测量2017-2018年自然生长状态下两样地的降雨再分配模式,明确两种灌木截留量的主控因素。并通过人工控制沙柳样地LAI大小,对比相同气象因素条件下LAI对冠层截留量的影响。主要研究结果如下:

1DHP获取的多种植被类型的LAILAI-2200测量结果存在显著线性相关关系(R2=0.85P0.05,均方根误差为0.256。拍照时的太阳辐射值直接影响专业软件处理照片时的关键参数,镜头朝下时,计算LAI的关键参数-绿色指数随着拍摄照片时太阳辐射强度的增加而增加;但是镜头朝上时,亮度参数随着太阳辐射值的增加而减少。通过关键参数的调整,玉米、大豆和柠条样地LAI变化与LAI-2200测量结果一致,很好地反映了这些植物生长期的LAI动态变化,鱼眼摄像机可以用于定位监测LAI动态变化。

2)利用鱼眼摄像机可获取阔叶作物(玉米和大豆)和旱生灌木(柠条)的每日LAI,它在测量植被LAI的动态方面具有较大的潜力。本研究首次将玉米地生长后期拍摄的鱼眼照片中的衰老部分去掉,降低了光学仪器在测量玉米等阔叶作物生长后期时较大的不确定性(y=0.24xRMSE=0.79),这与破坏法测量LAI值产生了较好的一致性(y=0.75xRMSE=0.37)。朝下安装的鱼眼摄像机获取的LAI具有更好的代表性,适合于低矮植被,如大豆、苜蓿和长芒草等;而朝上安装的鱼眼摄像机由于处理照片时分类像素的不确定性较小,更适合较高的植被,如高杆作物(玉米)、灌木和乔木等。使用CAN-EYE软件处理柠条地朝上照片时,通过选择不同的分类像素,可估计木质部分面积指数(WAI)。鱼眼摄像机可以精确、廉价、简单和快速的测量冠层结构变量LAICI。应该对该方法进行进一步的评估,特别是对于密集的冠层。

3)柠条林冠层截留量随降雨量增加而逐渐增加,稳定截留率约为15%。沙柳林冠层截留率相对较低,稳定截留率约为10%。沙柳样地LAI高水平(1.7)时,穿透雨最少,树干茎流居中,截留量最多;LAI中水平(1.3)时,穿透雨居中,树干茎流量最多,截留量最少;LAI低水平(1.0)时,穿透雨最多,树干茎流量最少,截留量居中。这表明沙柳截留量受树干茎流量的影响较大,LAI低水平时树干茎流量最少,进而推断叶片对茎流起积极作用,而相比LAI高水平,LAI中水平时树干茎流量更大,叶片似乎对茎流量又起消极作用。适宜的LAI可减少截留蒸发,增加树干茎流。

Other Abstract

The automatic and accurate measurement of LAI is of great importance for the research of ecological processes. In the study, first, the digital hemispherical photography (DHP) and LAI-2200 were used to measure the canopy LAI of different growth stages of typical land in the northern Loess Plateau, including C. korshinskii, S. psammophila, alfalfa, S. bungeana, soybean and maize. And the accuracy of the optical fisheye lens is verified by comparing the two measurements. Second, photos were daily taken by fisheye webcam and processed by CAN-EYE software to obtain the dynamics of LAI in soybean, maize and C. korshinskii fields. Meanwhile, these measurements were compared with data obtained using the LAI-2200, (DHP), and destructive sampling measurements (LI-3000A). Finally, the fisheye webcam method is used to measure the influence mechanism of LAI on the canopy interception of shrubs. Taking C. korshinskii and S. psammophila fields as research objects, the rainfall redistribution patterns of the two fields in the natural growth state of 2017-2018 were measured to clarify the main controlling factors of the two shrub interceptions. And different LAI levels were set by artificially for the three S. psammophila fields, and the effect of LAI on canopy interception under the same meteorological conditions was compared. Main findings of this study are listed as follows:

(1) The results showed that a linear correlation existed between the LAI measured by DHP and LAI-2200. The coefficient of determination (R2) was 0.85 (P0.05) and root mean square error (RMSE) was 0.256. The key parameters of professional software were affected by the solar radiation when taking pictures. When the downward webcam was used, the green index was the key parameter which increased with the increase of solar radiation. However, the brightness index decreased with the increase of solar radiation when the webcam was upward. Through the adjustment of the key parameters, the results of LAI of maize, soybean, and C. korshinskii were consistent with the LAI-2200 results, well reflecting LAI dynamics during the plant growth. The fisheye webcam could be used for monitoring the dynamic LAI of different vegetations.

(2) In this study, an approach was proposed and evaluated to continuously monitor daily LAI in broadleaf crops (maize and soybean) and xerophytic shrub (C. korshinskii) using fisheye webcam. Fisheye webcam shows great potential for continuous LAI measurement in agricultural crop and shrub fields. As the upward-pointing fisheye webcams originally derived PAIeff contains YAI is usually unreliable for maize during the senescent period (y=0.24xRMSE=0.79) , remove senescent parts from fisheye photos was first reported in this study, which yielded a good agreement with LAI values from destructive method (y=0.75xRMSE=0.37) . The downward webcam allows a more wider spatial representativeness and is suitable for low vegetation, such as soybean, alfalfa and S. bungeana. In contrast, the upward webcam is more suitable for tall vegetation, such as long-stalked crops (maize), shrub and arbor, because of the small classification uncertainties of the upward photos. Potentially, the wood area index (WAI) can be estimated from the upward fisheye photos by selecting different classify pixels using CAN-EYE software. The fisheye webcam can achieve monitor canopy structural variables such as LAI, GF, and CI in an accurate, inexpensive, easy and fast manner. Further evaluation of this method should be made, in particular for dense canopies.

(3) The interception increased with the increase of rainfall for C. korshinskii tree stand, and the steady interception rate was about 15%. Comparatively, the interception rate was relative low for S. psammophila tree stand and the steady interception rate was about 10%. For S. psammophila tree stand, when the LAI level is high (1.7), the throuhfall is the least, the stemflow is in the middle, and the interception is the most; when the LAI is at the middle level (1.3), the throuhfall is centered, the stemflow is the most, and the interception is the least; when the LAI is low (1.0), the throuhfall is the most, the stemflow is in the least, and the interception is the centered. This indicates that the interception of S. psammophila is greatly affected by the stemflow. When the LAI level is low, the stemflow is the least, and it is inferred that the leaf has a positive effect on the stem flow, while compared with the high level of LAI, the stemflow is larger at the level of LAI, and the leaf seems to have a negative effect on the stemflow. The appropriate LAI could reduce the interception evaporation and increase the stemflow.

MOST Discipline Catalogue工学 ; 工学::环境科学与工程(可授工学、理学、农学学位)
Language中文
Document Type学位论文
Identifierhttp://ir.iswc.ac.cn/handle/361005/8808
Collection水保所2018--2019届毕业生论文
Recommended Citation
GB/T 7714
牛小桃. 水蚀风蚀交错区典型植被叶面积指数动态变化及其对冠层截留量的影响[D]. 北京. 中国科学院大学,2019.
Files in This Item:
File Name/Size DocType Version Access License
牛小桃-毕业论文.pdf(2566KB)学位论文 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[牛小桃]'s Articles
Baidu academic
Similar articles in Baidu academic
[牛小桃]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[牛小桃]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.