Forest health and habitat

Project F1: Monitoring dieback on state forests

• Monitoring canopy dieback in NSW North Coast forests - Research note
• Monitoring of sudden canopy dieback in coastal forests - Final report

Dieback is a term used to describe gradual deterioration of tree health sometimes leading to tree death. There are many potential causes of dieback including pests, disease, drought, fire and management actions. The Coastal IFOA recognises Bell Miner Associated Dieback as one form of dieback and requires FCNSW to implement management actions to mitigate its further spread.

The program engaged Professor Brendan Choat at the Hawkesbury Institute for the Environment, Western Sydney University to develop a repeatable, scientifically robust and low-cost method to monitor landscape-scale forest canopy dieback and identify principal drivers of dieback. The Western Sydney University team collaborated with remote sensing research scientists from NSW Department of Climate Change, Energy, the Environment and Water (DCCEEW).

The approach

The method was tested for forests impacted by the 2023 ‘flash drought’ in the mid-north coast region of NSW. High-resolution PlanetScope satellite imagery was used to manually classify canopy conditions into three categories: dead, partially dead, and live. This classification was then extended across the broader pilot study area using publicly available Sentinel-2 satellite imagery and a machine learning approach.

A further analysis was done to assess the influence of topography, fire history, and land management on canopy dieback patterns. The researchers also collected on-ground-data at selected sites to assess longer-term impacts on tree health for different classes of canopy dieback.

Key findings

The pilot study showed that the method was able to successfully identify canopy dieback and could be cost-effectively extended to other forest regions. Within the pilot study region the research found:

• canopy dieback was most extensive on ridges and north-facing slopes
• the area affected by Bell Miner Associated Dieback was small compared to other factors associated with canopy dieback
• time series analysis from 2019 to 2023 showed multiple peaks in canopy dieback associated with drought and the 2019-20 fires
• the highest probability of full canopy dieback was found in National Parks and other protected areas outside State Forests; however , the highest probability of partial canopy dieback was found in State Forests available for harvesting. Further analysis is needed to explore reasons for these differences, which could reflect forest types and site conditions or differences in management.

Spatial data for the pilot study region will be made available on the SEED portal.

This project adds insights to the wider forest dieback research program overseen by the Commission.

Project F2: Assessing change in forest structure on state forests

• Stage 1 Report – Monitoring forestry outcomes using Airborne LiDAR (October 2023)
• Stage 2 Report – Retrospective analysis of forest structure change (October 2024)
• Research note - Investigating LIDAR for forest monitoring (October 2024)

Researchers at the University of Newcastle analysed variation in forest structure on state forests using airborne light detection and ranging (LiDAR) data and other spatial and non-spatial data. The researchers assessed the influence of natural and anthropogenic activities including management (harvesting, prescribed fire, areas managed for conservation), natural disturbance such as wildfire, topographic position and forest type. Building on this work, researchers also analysed structural complexity, canopy height and canopy foliage density.

Worked examples have been provided for several areas of NSW State Forests across the Coastal IFOA region exploring remote sensing metrics. These examples examine the impacts of terrain, fire severity in 2019/20, historical harvesting and differences across NSW State Forest management zones on forest structural attributes.

The researchers contributed to a stage 1 report on developing the methods and remote sensing metrics to assess change in forest structure and a stage 2 report testing these methods in case studies and reporting the findings. We produced a research note to outline the research and key findings.

Key findings

The research demonstrated LiDAR metrics offer a means to summarise change in forest height and canopy coverage. Modelling canopy top height recovery and incorporating slope types, harvesting practices and Forest Extent and Severity Mapping (FESM) classes can describe the rate of change across these different factors.

This research found:

• canopy top height and canopy coverage recovers after harvest events across a range of slope classes
• comparable rates of canopy regrowth over time are observed despite variations in harvesting intensity
• similarities are seen in canopy structure between areas managed for timber production and areas managed for conservation at the landscape scale
• differences in canopy top heights due to different fire severities were observed via LiDAR
• harvesting influences vertical and horizontal distribution of biomass, but this returns to pre-harvest levels within a short period in selectively harvested forests.

Explore the data

You can explore the LiDAR-derived data and reference layers for the seven case study regions across NSW in web maps on the NSW Spatial Collaboration Portal.

The LiDAR point cloud data and 1m Digital Elevation Model for each study region are available on Geoscince Australia's ELVIS portal.

Watch our webinar

On 15 November 2024, Professor Scott Brown from the University of Newcastle led a webinar presenting findings and interactive webmaps from this research exploring changes in forest structure in NSW state forests.

Following the presentation there was a Q&A session with a panel of forest monitoring experts, including Professor Brown (University of Newcastle), Professor Patrick Baker (University of Melbourne) and Dr Sam Hislop (FLINTPro).

Watch the webinar on the new LiDAR-based forest monitoring.

Project F3: Assessing change in tree composition on state forests

• Research Note - Investigating the effect of disturbance on tree species composition in coastal state forests (July 2025)
• Report - The effect of timber harvesting and fire on tree species composition in eucalypt forests with a focus on arboreal marsupial feed trees (May 2025)

The program engaged researchers at the Forest Science Unit, DPIRD to assess how tree composition has changed over time on coastal state forests. Timber harvesting has the potential to alter tree species composition and thus the habitat and nutritional quality available for many species that rely on them. The researchers accessed permanent growth plot measurements from 1979 to 2023 held by FCNSW to undertake the analysis. The investigation also assessed change in selected focal feed trees for koalas, greater gliders and yellow-bellied gliders (e.g. tallowwood, grey gum, manna/ribbon gum) and important timber species (e.g. blackbutt, spotted gum).

Key findings

The researchers found:

• overall, timber harvesting or fire disturbance have minor and transient effects on tree species composition measured by basal area, with lower intensity, selective harvesting having less effect than more intensive harvesting or multiple disturbances
• this conclusion also applies to many key feed tree species for koala, yellow-bellied glider and southern greater glider, although the response to harvesting was variable between tree species, regions, and harvest intensity
• there were no significant changes in basal area favouring preferred commercial species following harvesting
• together with the substantial body of previous research on arboreal marsupials, the tree composition changes identified in this study are unlikely to be a severe risk to arboreal marsupials under the conditions assessed.

The researchers note there were a range of data limitations including limited historic disturbance records, limited measurements for some tree species, change in plot size over time, fewer plots for regions outside the North Coast and fewer plots measured after 2020. Further data and work are needed to better understand habitat recovery and the influence of other disturbances and climate on tree species composition.

Project F4: Post-fire forest recovery in Coastal IFOA region

• Research note - Forest recovery in NSW following 2019-20 wildfires (April 2025)
• Report - Review of forest recovery in the Coastal IFOA region following 2019-20 wildfires (August 2024)

Major fires throughout the Coastal IFOA region during 2019-20 affected around 40 percent of native forests on public and private land.

Researchers from Western Sydney University investigated the impacts of these fires, the preceding drought and subsequent high rainfall on forest survival and structural recovery in coastal NSW forests. They analysed rates of tree mortality, topkill (the loss or death of tree crowns) and seedling recruitment across a range of fire severity, forest types and substrates (the layer below topsoil). The study used remote sensing data and field datasets collected in national parks and state forests. The researchers found:

• overall levels of tree mortality, topkill and recruitment following the 2019-20 fires were similar to previous fire seasons, however the geographic extent of forest affected was far greater, leading to higher losses of trees at a regional scale
• high levels of tree mortality and topkill were generally restricted to areas burnt at high or extreme severity
• tree mortality and topkill were more likely for small and large trees, and trees with pre-existing basal injuries
• the compound impact of the preceding drought and 2019-20 fires may have led to demographic change to forests, such as a shift toward mid- and smaller-sized trees
• high seedling numbers and growth in the years following the fires was likely due to above-average rainfall.

Project F5: Hollow use review

The Program has engaged Associate Professor Ross Goldingay, Southern Cross University to review the scientific literature on managing, monitoring and researching hollow bearing trees in NSW forests. The review aimed to identify cost-effective approaches to monitor hollows and highlight research gaps. In addition, the review summarised the current knowledge on hollow use by key fauna species in harvested landscapes.

The review found that most studies have been short-term in nature. There is no clear consensus on the number of hollow trees that different hollow-dependent fauna required to maintain persistence. However, some individual arboreal mammals use a subset of their dens at a high frequency - for example, between 0.1 to 1 primary den trees per hectare based on home range sizes. When managing harvested forests for fauna persistence, it is important that target fauna can access hollows that are suitable in size and appropriately spatially distributed. Management should also account for hollow attrition.

The review outlined best methods to monitor hollow use by key species including the strengths and weaknesses of each. A significant research gap is to understand how den trees either persist or are lost in a harvested landscape, particularly following fire.

This work will inform species occupancy monitoring and modelling under the coastal IFOA monitoring program and broader Forest Monitoring and Improvement Program. The work will also inform hollow simulation modelling undertaken by the Australian National University as part of the program.

• Review - Management and monitoring of tree hollow resources (Southern Cross University, February 2021)

Project F6: Impacts of fire on hollow-bearing trees on state forests

The Program has engaged Dr Raphael Trouve, a research fellow at the University of Melbourne, to investigate the effects of fire events on hollow-bearing trees and tree hollow formation within the Coastal IFOA region. The work follows advice from Professor Phil Gibbons, Australian National University to improve ways to better predict the number of trees with hollows required under the Coastal IFOA.

Overall, this project aims to quantify the rate of mortality, collapse and formation of hollow-bearing trees on state forests following fires of different intensities in the Coastal IFOA region. The work will deliver a model that can be used within FCNSW’s FRAMES modelling system to predict the effects of a fire event based on other available covariates such as diameter at breast height and tree species group.

Project F7: Perpetuating tree hollows under the Coastal IFOA

Researchers at the Australian National University (ANU), led by Professor Phil Gibbons have delivered research and recommendations to improve hollow simulation modelling on state forests. This model predicts the number of trees with hollows perpetuated under the current Coastal IFOA conditions.

Working with modellers at the Forestry Corporation of NSW (FCNSW), the researchers used the Forest Resource and Management Evaluation System (FRAMES) to model the persistence and recruitment of hollow-bearing trees under a range of scenarios. Prof Gibbons methods paper outlines the suitability of FRAMES to model persistence of hollows.

Overall, the researchers found:

• based on data, less than 50 percent of hollow bearing trees are typically occupied by vertebrate hollow-dependent fauna; thus, using tree diameter and tree species alone to predict how many hollow-bearing trees occur will over-estimate the number of hollow-bearing trees suitable for vertebrate species
• the number of hollow-bearing trees and number of trees with hollows suitable for occupancy by vertebrate fauna were able to be predicted using a combination of FRAMES outputs and hollow occupancy models
• simulations for the Coffs Harbour Timber Zone showed the Coastal IFOA conditions perpetuate 8-10 trees per hectare with visible hollows and 2-3 trees per hectare with hollows suitable for occupancy by vertebrate fauna in the net-harvest area over a 200 year modelling period
• in areas permanently excluded from harvesting (approximately 51 percent of the study area), trees with visible hollows were predicted to increase from a mean of 8 per hectare to 26 per ha and trees with hollows suitable for occupancy by vertebrate fauna were predicted to increase from a mean of 3 per hectare to 10 per hectare.

With minor changes, Prof Gibbons suggests FRAMES can be used to simulate outcomes of the current Coastal IFOA and any proposed changes to the rule set on the tree hollow resource. In addition, FRAMES can be further improved over time to better simulate impacts.

The NSW Forest Monitoring Steering Committee has endorsed Prof Gibbons recommendations, including projects to collect further data to improve hollow-specific models for FRAMES. We produced a research note outlining the key findings from this project.

• Report - Perpetuating trees with hollows under the CIFOA (February 2024)
• Report - Predicting trees containing hollows suitable for vertebrate fauna in NSW forests (March 2023)
• Research note - New method to model hollow-bearing trees in coastal NSW forests (December 2024)

Explanatory note – greater gliders

Prof Gibbons’ research and report perpetuating trees with hollows under the Coastal IFOA was conducted prior to the introduction of the Site-specific biodiversity condition for Greater Gliders in the Coastal IFOA region on 16 February 2024.

Future modelling of hollow-bearing tree persistence within the Coastal IFOA region will incorporate the new requirements for tree retention within forestry operation areas specified in the Site-specific biodiversity condition for Greater Gliders in the Coastal IFOA region.