ibul 2004
TREE-HEIGHT MEASURING CHARACTERISTICS OF URBAN FORESTS
BY LIDAR DATA DIFFERENT IN RESOLUTION
Y. Imai* *, M. Setojima?, Y. Yamagishi ", N. Fujiwara*
* Kokusai Kogyo Co., Ltd. , 5 Sanbancho, Chiyoda-ku, Tokyo, Japan - (yasuteru_imai, mseto)@kkc.co.jp
® Organization for Landscape and Urban Greenery Technology Development , 1-21-8 Toranomon,
Minato-ku, Tokyo, Japan - yamagishi@greentech.or.jp
* National Institute for Land and Infrastructure Management , 1 Asahi, Tsukuba-shi, Ibaraki, Japan -
fujiwara-n92db(à) nilim.go.jp
Commission VII, WG VII/4
KEY WORDS: LIDAR, Urban, Vegetation, Multiresolution, Measurement, Accuracy, Comparison
ABSTRACT:
Urban forests are important as valuable green resources to give moisture and tranquillity of mind in our living, requiring an efficient
control of forest increase/decrease and growth rate. In this study, we conducted tree-height measurements using LIDAR data
different in resolution at urban parks in the suburbs of Tokyo, and investigated the measuring characteristics. As a result, for conifers,
LIDAR data of high resolution (1 m) of high treetop irradiating probability showed higher measuring accuracy. For trimmed
broadleaf trees, on the other hand, the LIDAR data of low resolution (2 m) of low penetration probability between branches showed
higher measuring accuracy, indicating that the tree-height measuring accuracy by the LIDAR data does not necessarily depend on
the resolution.
1. INTRODUCTION
Urban forests are important as valuable green resources to give
moisture and tranquillity of mind in our living. In recent years,
much attention has been directed to the diversified functions of
urban forests such as the contribution to the prevention of
global warming through the absorption of CO, of plants,
mitigation of heat island phenomena by the evapotranspiration
of vegetation and disaster prevention effect such as the creation
of sheltering space and prevention of fire spread. Especially,
high expectation is accorded to the prevention of global
warming by CO, absorption function of plants in relation to the
Conference of the Parties to the United Nations Framework
Convention on Climate Change (COP) (Handa & Teshirogi,
2002).
The continuous monitoring of increase/decrease and volume of
growth of plants is necessary to promote the conservation and
creation of urban forest. To this end, LIDAR is effective means
in monitoring wide ranged area, and many studies have been
conducted in recent years as the measurement of tree height
using LIDAR provide basic parameter of CO, absorption
volume (Yamagata et al., 2002). However, the measurement of
tree height by LIDAR has been mostly conducted at mountain
forest, and the case studies at urban forest are rather scarce. In
addition, the platform of LIDAR is roughly classified into
rotary wing (helicopter) and fixed wing (Cessna) but the study
on the comparison of both platforms is very few.
In this study, we measured the tree height of urban forests using
the LIDAR data of different acquisition density obtained by
rotary wing and fixed wing, and studied their measuring
characteristics.
* Corresponding author.
513
2. MATERIAL
2.1 Study Site and Field Survey
Target area is set to Koganei Park in Tokyo, Japan (Figure 1).
The park comprised of area of 77 hectares containing variety of
trees with blend of the different combination and density. Study
sites area selected from this park because the park contains
different types of tree and the ground truths for the verification
of those areas are easily obtained by the field survey. Five sites,
named St.1, St.2, St.3, St.4 and St.5, are selected regarding to
the tree types, the height, and the density. Detail information
are shown in the Table |.
Field surveys are conducted in the all sites on November 5 and
6 of 2002. For each tree, the position, the height, the diameter at
breast height, and the species are obtained as the ground truth.
The positions and heights of every tree are acquired using the
differential GPS (GPS Pathfinder Pro/XR) and the laser
measurement instrument (LaserAce300). Information about the
diameter at breast height and the species are based on the actual
measurement.
2.2 Multispectral Satellite Image
In this study, we used IKONOS data to extract the vegetation
area. IKONOS data we used is that obtained on September 21,
2000 and the ortho-corrected multi-spectral (red, green, blue
and near infrared) image, and its spatial resolution is 4 m
(horizontal accuracy +/- 3.5 m) (Table 2).