A framework for evaluating changes in extent and condition of woody native vegetation communities in the Murray Darling Basin. An approach for developing scenarios for future landscape transformation to enhance native vegetation communities

1. Managing vegetation for multiple
benefit outcomes –
Diagnosis and Prognosis
Richard Thackway
Presentation to Murray Darling Basin Authority
22 September 2016

2. MDBA has identified two development areas
for native vegetation:
1. Analysis framework for evaluation of extent and condition
of woody vegetation communities
2. Developing scenarios for future landscape transformation

3. Definitions
• Change in a plant community (type) due to effects of land
management practices:
– Structure
– Composition
– Regenerative capacity
• Resilience = the capacity of an plant community to recover to
a reference state following a change/s in land management
• Transformation = changes to vegetation condition over time
• Condition, resilience and transformation are assessed relative
to fully natural a reference state
Vegetation condition

4. VAST = Vegetation Assets States and Transitions
NVIS = National Vegetation Information System
VIVIVIIIIII0
Native vegetation
cover
Non-native vegetation
cover
Increasing modification caused by use and management
Transitions = trend
Vegetation
thresholds
Reference
for each veg
type (NVIS)
A framework for assessing modification of
native vegetation extent and condition
Condition states
Residual or
unmodified
Naturally
bare
Modified Transformed Replaced -
Adventive
Replaced -
managed
Replaced -
removed
Thackway & Lesslie (2008)
Diagnostic attributes of VAST (classes):
• Vegetation structure
• Species composition
• Regenerative capacity
VAST-2 criteria
and indicators
Change & Trend
NVIS

5. 1925
Occupation
Relaxation
Anthropogenic
change
Net benefit
time
1900 20251950
Reference
changeinvegetation
indicatororindex
1850 1875 1975 2000
Model of ecosystem change
(causes & effects)
Baseline

6. Accounting for changes in
native veg type, extent and condition
Land managers use LMP to influence ecological function at sites
and across landscapes by changing:
• Vegetation structure
• Species composition and
• Regenerative capacity
LMP deliberately &/or unintentionally do this by:
• Modifying
• Removing and replacing
• Enhancing
• Restoring
• Maintaining
• Improving
*
* Natural disturbances
Function
Structure &
Composition
LMP = land management practices

7. Goal is to develop whole of landscape monitoring
of criteria and indicators relative to reference states
0
20
40
60
80
100
1985 1990 1995 2000 2005 2010
Year
FPC
Source: Tim Danaher and Phil Tickle
Site and landscape scales

8. 1925
Occupation
Relaxation
Anthropogenic
change
Net benefit
time
1900 20251950
Reference
changeinvegetation
indicatororindex
1850 1875 1975 2000
VAST
classes
Evaluation of ecosystem change that
rely on remote sensing metrics however …
Hyperspatial,
hypertemoral &
hyperspectral
remote sensing
Aerial photos
and digital
imagery
Aerial photos
Quantitative
Baseline
Time before remote
sensing

9. Caution - remote sensing change is relative to a baseline
- it is only half the answer – NEED REFERENCE STATE
Zero/constant baseline (e.g. environmental planting = reveg)
Response
variable/s
Time
Start of
activity/
intervention
Time
Varying baseline (e.g. environmental watering)
Response
variable/s
Single intervention &
climatic variability
Baseline
Response to activity/ intervention
Indicator 13: Overstorey height
Indicator 4: ground water

10. Photo by Peter Coyne
1740
1906
Phillip Island, South Pacific – extent and condition
Photo State Library NSW: JW Beattie
1860 already denuded

11. year
score%
Pine – Hardwood Subtropical Rainforest, Phillip Island, Sth Pac
Pigs
released
Uninhabited
island
Pigs died
out
Goats, rabbit
and fowl
released
Goats died
out
Rabbits
eradicated
Rabbit
control
commenced
Commenced
passive & active
restoration.
Minimal ecological
monitoring
Phase 1 Phase 2 Phase 3 Phase 4

12. Evaluation framework
• Has been extensively used to evaluate condition outcomes
– Diagnosing and reporting condition (i.e. status and trend) of
vegetation types
– Evaluating examples of restoration and regeneration

13. Transformationscore
Years
1800
2016
Reference
Developing scenarios for future landscape
transformation
Modified
Transformed
Replaced/
managed
Residual
Replaced/
adventive
VAST Classes
1850 19501900 2000 2050 2100
Replaced/
removed
Baseline
Classes can be modelled as extent and condition
Extentnative

14. NSW, SB Bioregion, Cumberland SF, Transect 2 (ex-comp 3a, 7a, 7b, 7c)
Reference pre-European: Sydney Blue Gum High Forest
Commenced
managing
area for
recreation.
Weed control.
Arboretum
abandoned
Cleared &
sown to
improved
pasture for
grazing &
orchard
Commenced
grazing
native
pastures
Indigenous
people
manage the
area
Grazed area
gazetted as
State Forest,
commenced
planting
arboretum
Area
logged for
building
houses
and fences
Commenced
managing area
as a future
production
forest. Weed
control
Explorers
traverse
the area
and site
selected
Ceased
grazing.
Area
purchased
as a future
working
forest
Modified
Transformed
Replaced/
managed or
removed
Residual
Replaced
/adventive
VAST

15. Before developing any scenarios it is critical to assess where, when
and how landscapes were/are transformed relative to a reference
LU = Land Use, LMP = Land Management Practices
VAST Diagnostic attributes
Time

16. Bioregion: Mulga Lands
Reference state: Extent of pre-clearing Regional Ecosystem (RE) 6.3.13
Atriplex spp., Sclerolaena spp., species of Asteraceae and/or short grasses open-herbland on alluvial
plains
Source: Daniel Ferguson, Queensland Herbarium

17. Need to assess which of the 10 key criteria are affected
and by how much i.e. Condition
Soil
Vegetation
Regenerative capacity/ function
Vegetation structure &
Species composition
1. Soil hydrological status
2. Soil physical status
3. Soil chemical status
4. Soil biological status
5. Fire regime
6. Reproductive potential
7. Overstorey structure
8. Understorey structure
9. Overstorey composition
10. Understorey composition

18. NSW, SB Bioregion, Cumberland SF, Transect 2 (ex-comp 3a, 7a, 7b, 7c)
Function – Regenerative capacity
Criteria #1 Criteria #2
Criteria #3 Criteria #4
Disturbance regime

19. NSW, SB Bioregion, Cumberland SF, Transect 2 (ex-comp 3a, 7a, 7b, 7c)
Function – Regenerative capacity
Criteria #5 Criteria #6

20. NSW, SB Bioregion, Cumberland SF, ex-comp 3a, 7a, 7b, 7c
Vegetation structure
Indicators:
#13: Height
#14: Foliage cover
#15: Age structure
Indicators:
#16: Height
#17: Foliage cover
#18: Age structure
Criteria #7
Criteria #8

21. NSW, SB Bioregion, Cumberland SF, Transect 2 (ex-comp 3a, 7a, 7b, 7c)
Species composition
Criteria #9
Criteria #10

22. Conclusions
• Vegetation types are kept in various extents and condition states
depending on social and economic ideals, values, goals and drivers
• Conversions to non-vegetated, non-native and highly modified native
commenced at European settlement; the reference state
• Conventional mapping and monitoring of vegetation type, extent and
condition using remote sensing is only started around 70 years ago
• Transformations are changes in condition and extent relative to a
reference state: function, structure & composition
• Net gains in condition and extent(spatially and temporally) can be
accounted for as changes in function, structure and composition
• To achieve long term lasting changes in extent and condition we must
engage land managers using language of public and private benefits
• Scenarios that aim to transform extent and condition must be
informed locally by evidence of previous land management practices

23. Thank you

24. Supplementary Information

25. Dynamics
Rainfall is assumed the main driver of natural system dynamics
It is a key dataset used in developing chronologies of effects of land
management on vegetation condition:
• Period 1900 – current1
• Average seasonal rainfall (summer, autumn, …) is highly correlated to
LMP2 and their effects on function, structure and composition
• Rainfall anomaly is calculated above and below the mean
• Two year running trend line is fitted to give insights into El Niño and La
Niña events3
1. Bureau of Meteorology: modelled monthly rainfall 5km resolution national dataset
2. LMP – Land management practices
3. El Niño events are associated with prolonged periods of below average rainfall and at times, devastating droughts. A rapid
onset of a La Niña event, following a severe El Niño event, have been associated with major soil erosion events and major
cycles of regeneration and germination

26. Differentiating natural disturbance events and
land management practices
Photos by Garry Dowling and Richard Thackway
2006 20132009
Low intensity grazing -
cell grazing sheep
Severe dust storm Low intensity grazing -
cell grazing sheep

27. Reference state: Pre-clearing Regional ecosystem 6.3.13
Atriplex spp., Sclerolaena spp., species of Asteraceae and/or short grasses open-
herbland on alluvial plains
Illustrates a
baseline but
not the
Reference state
Photo:MelindaLaidlaw
High intensity cattle grazing -
Continuous or set stocking
Source: https://publications.qld.gov.au/storage/f/2014-09-18T23%3A49%3A36.560Z/re-seq-landzones.pdf

28. A Framework for Program Evaluation
Source: http://sphweb.bumc.bu.edu/otlt/mph-modules/ProgramEvaluation/ProgramEvaluation2.html
Need continuous
social learning linked
to adaptive
management

29. Yapp, Walker and Thackway (2010)
VAST 1 Residual/unmodified*
VAST II1 Transformed
* Reference state
Multiple
benefits
(ecosystem
services) and
condition
states