Content uploaded by Manat Srivanit
Author content
All content in this area was uploaded by Manat Srivanit on Sep 03, 2021
Content may be subject to copyright.
6
International Perspectives
A Case Study on the Visual Assessment of Tree Health Status and Risk ━ Useful GIS-
based Tool for Urban Tree Management in Thailand
By Suppawad Kaewkhow and Dr. Manat Srivanit
Faculty of Architecture and Planning, Thammasat University, Pathumthani, Thailand
Urban trees are very instrumental in
sustainable urban green infrastructure
planning as they are beneficial to the
ecosystem and city environment. This is
especially true for a country like Thailand,
which is located in the equatorial zone and
is facing the problems of water manage-
ment and stormwater runoff, urban heat
island, and air pollution. Therefore, the visual
assessment of tree health status and risk in
urban areas is a fundamental practice that
analyzers can get into, which involves
examining perennial plant conditions,
including the visible shoot and root systems
and their surroundings. After visual assess-
ment, the information gained can be
synthesized, classifying and identifying any
trees that have a high risk. Then, a
professional arborist may be called for an
advanced tree risk assessment.
After reviewing the literature, we found that
many different ways of visual tree assess-
ment or VTA have been developed for use
by organizations and cities, such as the
International Society of Arboriculture (ISA),
the Tree Hazard Evaluation Method, the US
Department of Agriculture Forestry Service
(USDAFS), and the Urban Tree Risk
Management - A Community Guide to
Program Design and Implementation
(Coelho-Duarte et al., 2021; Norris, 2007).
Additionally, many experts have seen that
research is necessary for studying and
increasing the knowledge of risk evaluation,
and to find a basic and suitable method that
is easy to understand for community
collaboration in perennial plant risk
management (Koeser & Smiley, 2017).
Nowadays, Geographic Information Systems
(GIS) are found to be important for many
cities because they combine efficient and
effective spatial information systems with big
data analysis. Therefore, this study aimed to
develop tree risk observation and the
geospatial analysis for understanding the
spatial patterns of tree health conditions and
risks. Another purpose was to support our
knowledge of urban plants and their
biodiversity. We expect that the GIS will be
an applicable tool for many cities in Thailand
to further efficient urban tree management
(Fig.1).
7
Figure 1. The research framework of the study.
This research was conducted in the area of Thammasat University, Rangsit Campus in
Pathumthani Province as the university has a strategic plan that aims to be a smart university and
a model of the sustainable campus. The area of the students and staff dormitories was selected
as the research location since it has model development and tree health condition and risk
assessment due to its crowded condition with a high frequency in occupation and buildings. We
began with a review of the previous studies on the visual assessment methods. Findings, more
than 35 variables have been studied in relation to assess tree health status and risk, to select the
key elements that affected the health conditions and risks of the trees from an interview with 3
Thai professional arborists. After that, the validity was calculated by analyzing the index of item
objective congruence (IOC). We found that 11 main variables have the IOC at a high level
(Kaewkhow & Srivanit, 2020), and we used these to design and create a survey prototype form
which was tested in the reliability and validity assessment by the prototype field testing (Fig. 2).
According to the results, the overall accuracy from the Kappa coefficient is 0.72 or 72% on
average, found by comparing the assessments, which were done by the professional arborists
and non-professional users, against the same assessment form to survey randomly in
8
approximately 10% of trees in the selected
area for study. Only 2 variables were found
to have a slightly different index of item
objective congruence or IOC, wind
turbulence effect and row spacings. The
coefficient is between 0.01 and 0.20 as per
the record of the Inter-Rater Reliability (IRR).
However, the experts had a suggestion to
adjust and add 2 more main variables,
girdling roots and the floor protruding roots,
which are defects that are affected and
caused by the trees in the research location,
to allow the risk assessment to be more
accurate. After that, an improved, more
practical version of the survey prototype form
was created.
Additionally, 442 trees in total were observed
by using some tools such as a mobile global
positioning system (Garmin GPSMAP 64s
Handheld), a distance measurement camera
(NIKON Forestry Pro II), and a laser
rangefinder. Every survey prototype form
was filled in by the tree positions and the data
were input as the point feature to the QGIS
Program Version 3.18 which has been the
Open Source Geographic Information
System (GIS). Then, these attributes of the
trees were used in the spatial pattern
analysis, to create a result of perennial plant
health condition and risk assessment
presentation (Fig. 3 and Fig. 4). According to
the study, most of the high-risk trees are
found in the road network and the paved
parts in the research area
.
Figure 2. (a) paper map of studied tree positions, (b) example of the preliminary version of tree
health status and risk assessment form, and (c) a survey by an expert in arboriculture and a
non-familiar area observer during prototype field testing.
9
The plants that have more defects and the
highest risk are 67 Pride of India
(Lagerstroemia speciosa (L.)Pers.), Pong
pong (Cerbera odollam Gaertn.), False
Mahogany (Swietenia macrophylla King.),
and Cock Tree (Millingtonia hortensis L.f.),
respectively.
Additionally, this study has presented a
guideline for the use of the recorded tree
information in the city ecosystem and
environment services evaluation.
It has provided an example of how it can be
applied in the plant information foundation
which is in the microclimate model of the
ENVI-met model program, to analyze the
tree effects on the outdoor thermal
environment and thermal comfort (Fig. 5).
Therefore, it is beneficial to the planners,
designers, and landscape architects to use it
as an assessment tool to find the effects of
their further climate-sensitive urban planning
and design.
Figure 3. Seven main types of tree defects above the ground.
10
Figure 4. The map shows the health rating of trees across the study area.
As per the above-mentioned result, the tree
health conditions and risks assessment form
has allowed us to identify the risks in
perennial plants exactly as an expert in
arboriculture. However, to be applied in
different urban areas, the form needs to be
adjusted by adding the specified variables of
the defects and conditions that affect tree
failure in that location. Moreover, a
determination of the importance of the
probability factors that affect tree failure in
each city life and property goals in
differences can result in a more accurate tree
risk level evaluation. Also, presenting the
mapping urban tree information that is based
on the geographic information system allows
the relevant persons can apply it in their
further planning and tree risk management.
Especially, the high-risk level trees require
basic aid and proper treatment to reduce the
risk. If the above-mentioned factors are well
done, the tree risks will be decreased.
Finally, since Thailand is located in the
tropical zone where perennials can
continuously grow throughout the year, their
risk assessment should not be done in only
one period per year. Observations should be
made during both drought and rainy seasons
as perennials might have a variety of
physical differences that can cause
assessment discrepancies.
11
Figure 5. An example of (a) creating a plant database of the ENVI-met microclimate model for
simulating the thermal effects of trees, and (b) distribution of the simulated outdoor thermal
comfort using the PET index.
References
Coelho-Duarte, A.P., Daniluk-Mosquera, G., Gravina, V., Vallejos-Barra, ´O., Ponce-Donoso, M.
(2021) Tree Risk Assessment: Component analysis of six visual methods applied in an urban
park, Montevideo, Uruguay. Urban Forestry & Urban Greening, 59, 127005.
Kaewkhow, S., Srivanit, M. (2020) Aggregation of Thai arborist judgments on urban tree hazard
inventories used to determine tree health at the single-tree level. IOP Conference Series:
Materials Science and Engineering, 910: 012023, 1-5.
Koeser, A., Smiley, E. (2017) Impact of Assessor on Tree Risk Assessment Ratings and
Prescribed Mitigation Measures. Urban Forestry & Urban Greening, 24, 109–115.
Norris, M. (2007) Tree risk assessments – what works – what does not – can we tell? A review
of a range of existing tree risk assessment methods, In ISAAC Conference Proceedings Perth,
p 29.
12
Acknowledgments
This work is kindly supported by the Thammasat University Research Fund (Contact No. TUGT
2/2562). We would like to express a heartfelt thanks to the involved professional arborists
including Taradon Tunduan, Bunchong Somboonchai, and Dr.Ponthep Meunpong, for their
participation and contribution to elicit which tree inventory parameters reflected the tree health
and risk at the single-tree level.
Suppawad Kaewkhow is an assistant professor in Landscape Architecture and
is also involved in several consultancy projects related to arboriculture such as
urban tree risk assessment and tree health management.
Dr. Manat Srivanit is currently working as an assistant professor in the area of
Urban Environmental Planning and Development, who specializes in
Geographic Information Science and Climate-Sensitive Urban Planning.
