Tree crashes onto school bus and cars, Perth Street

Tree crashes onto a car, Haiphong Road

Natural growth cycle of Trees

Trees go through a natural cycle of growth, aging, and death, and many of the trees planted in Hong Kong during the 1950s and 60s are now reaching the end of their lifespan. As such, it is important to reforest with native tree species that are better suited to current environmental conditions.

1950s Nathan Road | History pictures, Hong kong, Banyan tree

Senescence and Death

Adult Growth

Seedling Growth

Juvenile Growth

Forestree

Remote sensing and Forestry

Vegetation and Urban

Concrete jungle Hong Kong to get diverse array of plants on urban streets in drive to green the city

Trees provide a valuable contribution to the environment by helping to reduce the impact of urban heat islands, improving air quality, and providing shade and shelter to the public.

They can also help to support the goals of carbon neutrality and sustainable development, with the real estate sector increasingly using trees to create low-carbon gardens and green buildings.

The first zero carbon building in Hong Kong

Forestree

Remote sensing and Forestry

Why Tree really matters?

Toward the smart city development, why and how to manage our forest and trees?

Forestree

Remote sensing and Forestry

The way to manage trees

In recent months, there have been reports of stone walls and slope trees falling down one after another. The issue is compounded by the fact that the tree management department lack of united cooperation with other departments.

Forestree

Remote sensing and Forestry

Forestree developed an

Automated solution to manage trees in terrioriy-wide scale

Different types of forest in Hong Kong

Fung Shui Wood

Native woodland, usually a region of vegetation near a village.

Woodland and Shrublands

Tree covered areas which arose naturally

Secondary forest

regenerated through natural processes after human-caused disturbances

Plantation forest

established by planting and seeding, unnaturally

Forestree

Remote sensing and Forestry

Common ESG measures for smart city forestry

Evaluating by its age and species

assess the development area tree's age and species

Rearranging by plantation

logging first, forest plantation afterwards

Preserving

keeps the original wildlife habitat, such as Lok Ma Chau Wetland

Forestree

Remote sensing and Forestry

Vertical green wall

build and create a green wall and tree plantation

Development verse Preserving trees ?

Housing preseures leading to develop the secondary woodland

Developers are keen on developing the peripheries parts of the current country park, the government was forced to develop public housing to fulfill the needs and solve the inequalities.

Striking a balance between development and preserving the environment

A critical challenge for Hong Kong's next three decades of urban planning

Fung Shui land and Lowland Forest are the easiest part to be developed for housing, at the same time, the developers and NGOs are looking at the forest plantations that trying to reduce the impact on the environment.

Forestree

Remote sensing and Forestry

Challanges to strike a balance and preserve our forest

Limited availability

Constant pressure to develop housing and infrastructure, that leads the conversion of forested areas into land for development

Development needs and interest

Fanling Golf Club is a typical sample that shows the conflict between development and preserving the woodlands.

Histroical factors

Due to the complex land ownership in NT, such as the the permitted burial ground and government compensation issue for the woodland and development site selection.

Hillfires

Damage to develop, common method in NT landowner can deforestated their farmland, lower the ecological value to get a higher compentation.

Forestree

Remote sensing and Forestry

How Forestree toward ESG win-win situation with CSDI Open data

Forestree

Remote sensing and Forestry

Deciding and Optimisation

If the lands development need to deforestated/ cut off some of the vegtatation covers, optimisation and minimised the logging area can be with Forestree.

Monitoring and management

From aerail imageries, we are avaliable to retrive the tree health conditions and manage with remote sensing techniques to keep our forest safe.

Data analysation with CSDI and Our Systematic workflow

Output of the Fung Shui Woodland

Forestree

Remote sensing and Forestry

Our Auto-extraction application

Our Processing Status bar

Tree Invertory with Point Cloud GUI

Forestree

Remote sensing and Forestry

Post-procssing 1 Canopy-Slope intergated Simulation

Simulate and inspection across trees in man-made slope

Most of the tree collapse cases is caused by the tree root damage.

Roots may decays with infilterated water, so post-procssing the DSM, DTM into slope, flow directions and accumulation may filter in the potential root damage areas with GIS techniques.

Post-procssing 2 Health monitoring and prediction

Tree health map

LandsD has been continuously capturing aerial imagery of trees, which can be used to compute various vegetation indices such as the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), and Green Chlorophyll Index (GCI). These indices can be used to evaluate changes in the canopy and tree growth over time.

Base on the reflection and plant nature itself, the dead leaf has very low chlorophyll, anthocyanin content and reduced photosynthetic activity, stressed leaf compare has a relative higher content than the deadwood, the healthy leaf has the highest reflectance to the sensing platforms.

Post-procssing 3 Deep learning with Local Tree species

Tree can be study and identifies with Deep learning

Such as the Multi-trucks (Co-dominated), by the intensity value we can derive the multi-trucks from it, Delonix regia

Moreover, the tree leafs structures like Casurina equisetifiolia (needle like) leafs can identifies by the point density

Integrated-processing Soundscape Ecology for EIA

Tree-Bird Automatic Classification and invertory builder

Forestree now intergate with the soundscpae to automatically classifies bird species across the terriories, this process coorperate with AK IN KK bring the 1D sound wave georeference and correalted with the 3D forest space.

This builder foster the process of the EIA data acquisition and field observation process.

The Honourable sound recordings Artist

AK IN KK - Nature Field Recording HK

Mr. Kan Hei Chun (Field recordist)

Development timeline and sustabiable plan

Forestree

Remote sensing and Forestry

Our important dates and keep milestones:

Prototype

Study and revise the idea

Development

Application programming, trail and error testing for auto tree inventory

Deep leanring

Future development in tree species identifcation

Application Revise

Revise, improve and enhance the Forestree idea toward the social needs

Team of Forestree

Ho Chun Kai, Data Scientist

Lam Yik, Data Scientist

Lau Sze Chun, Geospatial-Analyst

Lo Sze Chai, Geospatial-Analyst

Yu Kai Him Otto, Project Coordinator

Some of the on ground validation trip:

Group of student from Land surveying and Geo-informatics, PolyU

Keen on GIScience, Remote sensing, Photogrammetry and Forestry.

For Forestree project and application details, please contact kai-him-otto.yu@connect.polyu.hk (Otto Yu).

Forestree

Remote sensing and Forestry

Reference and acknowledgement

We are grateful to express our appreciation to our mentor Ms. Kimmy Ng from HKSTP, Dr. Yan Wai Yeung and Dr. Sissi Chen from PolyU, for their advice, support and guidance throughout the CSDI Award 2023.

Agnew, M. (2018). Hong Kong housing crisis: government will develop country parks unless we stop them as task force is for show, says Greenpeace. South China Morning Post, Web site: https://www.scmp.com/sport/outdoor/trail-running/article/2155169/hong-kong-government-will-develop-country-parks-unless

Ahamed, T. (2022). Remote Sensing Application : Regional Perspectives in Agriculture and Forestry (T. Ahamed, Ed.; 1st ed. 2022.). Springer Nature Singapore. https://doi.org/10.1007/978-981-19-0213-0

Bryan C. Pijanowski, Almo Farina, Stuart H. Gage, Bernie Krause, Nadia B. M. Oliveira, Soundscape Ecology: The Science of Sound in the Landscape, Landscape Ecology, November 2011, Volume 26, Issue 10, pp 1213–1232, DOI: 10.1007/s10980–011–9600–8.

Choi, B. (2013). Airborne LiDAR Survey with SatRef Data Services. Civil Engineering and Development Department, Survey Division, Website: https://www.geodetic.gov.hk/common/satref_launch2013/airborne%20lidar%20survey%20with%20satref%20data%20services.pdf

Liu, S., Zhang, M., Kadam, P., Kuo, C.-C. J., & Kuo, C.C. J. Chung C. J. (2021). 3D point cloud analysis : traditional, deep learning, and explainable machine learning methods. Springer. https://doi.org/10.1007/978-3-030-89180-0

Ng, N. (2017). Hong Kong pushes on with controversial proposal to build housing on fringes of country parks. South China Morning Post, Web site: https://www.scmp.com/news/hong-kong/health-environment/article/2094696/hong-kong-housing-society-study-feasibility

Pijanowski, Bryan & Villanueva-Rivera, Luis & Dumyahn, Sarah & Farina, Almo & Krause, Bernie & Napoletano, Brian & Gage, Stuart & Pieretti, Nadia. (2011). Soundscape Ecology: The Science of Sound in the Landscape. BioScience. 61. 10.1525/bio.2011.61.3.6.

Planning Department (2021). Land Utilization in Hong Kong. Planning Department, Website: https://www.pland.gov.hk/pland_en/info_serv/open_data/landu/index.html#!

Stereńczak, Krzysztof & Kraszewski, Bartłomiej & Mielcarek, Miłosz & Piasecka, Żaneta & Lisiewicz, Maciej & Heurich, Marco. (2020). Mapping individual trees with airborne laser scanning data in an European lowland forest using a self-calibration algorithm. International Journal of Applied Earth Observation and Geoinformation. 93. 102191. 10.1016/j.jag.2020.102191.

Forestree

Remote sensing and Forestry