Ecosystem Function Analysis (EFA) Baseline Monitoring Department of Environment and Natural Resources (DENR)

(1)

COOE Pty Ltd 29/06/2012

RESTORATION MONITORI NG PILOT FOR COORONG, LOWER LAKES RESTORAT ION PROJECT

Ecosystem Function Analysis (EFA) Baseline Monitoring Department of Environment and Natural Resources (DENR)

COOE Pty Ltd ABN 65 147 909751

(2)

(3)

RESTORATION MONITORI NG PILOT FOR COORONG, LOWER LAKES RESTORAT ION PROJECT

Ecosystem Function Analysis (EFA) Baseline Monitoring

Prepared for: Department of Environment and Natural Resources (DENR)

Prepared by: COOE Pty Ltd

Level 1, 19 North Terrace Hackney, SA PO Box 591 Littlehampton, SA 5250 +61 8 83985090, www.cooe.com.au

Document code: DENR.EFA.1 29062012 Version: Final

This document and the information contained within were produced by COOE solely for the use of the client identified on the cover sheet for the purpose for which it has been prepared. COOE undertakes no duty to or accepts any responsibility to any third party who may rely upon this document.

DOCUMENT CONTROL

Date Version Reason for change Author(s) Editor(s) 22 June 2012 1 Draft-Internal review Agnès Cantin Faith Cook

25 June 2012 1 Minor Edits Agnès Cantin Katherine Goss

29 June 2012 2 Final version Agnès Cantin

(4)

Executive Summary

The Department of Environment and Natural Resources (DENR) Coorong, Lower lakes and Murray Mouth (CLLMM) Program is delivering a five year restoration project in the CLLMM region. As an emergency response to prolonged drought, bioremediation and revegetation activities were initiated in 2009. In July 2011, the project transitioned to building ecological resilience into the natural systems of the Lower Lakes and surrounding lands.

A pilot study (this report) was commissioned to trial components of Ecosystem Function Analysis (EFA) at revegetation and reference sites. The goal of the pilot project is to track trends towards the long term objectives of the restoration program. The monitoring assessed the resilience of the ecosystems surveyed and identified future challenges of the program. To assess restoration activities COOE monitored two specific components at each site:

 Ecosystem Functional Analysis (EFA) - to track the ecological and morphological changes at each location in particular to demonstrate that biodiversity and habitat structure are increasing and erosion is decreasing,

 Physical and chemical stability - to track the physical and chemical stability of the rehabilitated sites, in particular to demonstrate that no significant acid generation and metal mobilisation in the environment has occurred.

A total of nine sites were surveyed which comprised of seven revegetation and two reference locations around Lakes Alexandrina and Albert. Locations varied in vegetation type, aspect, and slope. At five of the revegetation locations, only one habitat type was identified. At two locations, two habitat types were identified. Three transects were established per habitat type, at each location.

The two reference sites (Bonney Reserve and Mulungushi) were chosen to provide data on both the effectiveness of EFA and to provide target values for the rehabilitated sites. A total of 11 sites/33 transects were assessed. The following summarises the parameters that were assessed at each site and recommends future methods and frequencies:

Photo-point monitoring

At each location a permanent photo-monitoring point was established. This was located to provide a broad overview of the site and will, over time, provide a record of the progress of rehabilitation.

Vegetation monitoring

Vegetation monitoring utilising principles of Ecosystem Function Analysis (EFA) was conducted. A Point-Centered Quarter (PCQ) method was employed to measure key vegetation indicators. Vegetation cover was variable within each site yet all sites were dominated by ground and shrub cover with a large proportion less than 2 m in height. The dominance of ground cover is important for the flow of resources across the ground surface by providing resource mobilisation and transport areas.

The Bonney Reserve reference site recorded the most number of plants per hectare with a median of 11,0942 plants ha^-1 and also contained a variety of habitat structure from ground cover to mature canopy species.

(5)

Landscape organisation and function

Landscape characteristics were determined at each of the transect locations. Along each transect the landscape profile and vegetation patch characteristics were measured. Within each of the landscape characteristics, eleven indicators of soil surface condition were assessed.

The landscape function at the revegetated sites was comparative to the reference sites.

However, the sites surveyed were very different in terms of habitat characteristics and as such comparisons between sites will provide erroneous trends. For example, the samphire sites cannot be compared to the reference woodland sites.

We recommended that with on-going monitoring, landscape function be compared within a site rather than between sites or appropriate reference sites are found (potentially on protected areas of Hindmarsh Island).

Habitat complexity

The habitat complexity of a three meter strip, either side of each transect was assessed.

This assessment included recording canopy cover, shrub cover, ground vegetation cover, litter and free water availability.

A reference site, Bonney Reserve, recorded the highest score and was the only site to score on each structure index. Most revegetation sites scored very low, due to the lack of

structural features.

It is recommended that this method can be assessed on an annual basis to provide a broad overview of the landscape function at each site. Indicators can be adapted for the region and aim of the surveys.

Soil chemistry

Soil was collected from each vegetation transect to establish a chemical baseline. This baseline included soil fertility and metals parameters.

The soils were analysed by ALSE (NATA certified laboratory) for organic matter, electrical conductivity, pH, soil moisture, exchangeable cations (calcium, magnesium, potassium and sodium), minor anions (nitrate, nitrite, reactive phosphorus) and total metals (arsenic, cadmium, chromium, copper, lead, nickel and zinc).

Values obtained for heavy metals were all well below guideline values for soil ecological investigation levels.

The vegetation and soil differences between sites were reflected in the chemical results. In comparison to the woodland and grassy sites, samphire dominated habitats were more saline, had higher organic matter and higher exchangeable cations.

Elevated soil salinity and exchangeable sodium may interfere with plant growth. The results indicated that soils at these sites were strongly sodic However, the results obtained for these samphire sites is a reflection of the depositional landform characteristics at the sites.

(6)

Effectiveness of EFA for monitoring restoration activities

One of the desired outcomes of this monitoring trial was to determine the appropriateness of using EFA to assess restoration activities in the CLLMM region. Beyond trialling the method, the objective of this monitoring pilot was to gather baseline data, to track the trends towards the long-term objectives of the restoration program.

The use of EFA has been successful in characterising each site in regards to vegetation structure and landscape function. This report provides the first year’s baseline data from which future monitoring can be compared with.

Monitoring of the sites should be undertaken on a yearly basis until the data warrant a longer period of time. Recommendations are provided for reporting on results of future monitoring assessments and also work to maximise rehabilitation at the sites. Regular site visits throughout the year will assist in determine how the site as a whole in functioning.

(7)

Table of Contents

Executive Summary ... i

List of Tables ... v

List of Figures ... v

List of Appendices ... v

1 Introduction ... 1

2 Methods ... 2

2.1 Site characteristics and set-up ...2

2.2 Landscape photo-monitoring ...5

2.3 Vegetation survey ...5

2.4 Landscape functional characteristics ...6

2.4.1 Landscape organisation ... 6

2.4.2 Soil surface assessments ... 6

2.4.3 Habitat complexity... 7

2.4.4 Soil chemistry ... 7

3 Results and Discussion ... 8

3.1 Site characteristics ...8

3.2 Landscape photo-monitoring ...9

3.3 Vegetation survey ... 13

3.4 Landscape functional characteristics ... 16

3.4.1 Landscape function ... 16

3.4.2 Habitat complexity... 17

3.4.3 Soil chemistry ... 18

4 Comparison between revegetation and reference sites ... 21

5 Recommendations ... 22

6 Conclusion ... 24

7 References ... 25

Appendices ... 26

(8)

List of Tables

Table 1 The combination of soil condition classes to derive indices of stability, infiltration

and nutrient cycling ...6

Table 2 Summary statistics for habitat complexity scores taken from each transect at each site... 18

List of Figures

Figure 1 Overview of all locations surveyed in the CLLMM region. ...4

Figure 2 The PCQ method for measuring spatial distribution of vegetation ...5

Figure 3 Photographic record of each site taken from permanent locations. Photograph on the left is taken at the widest zoom and a close-up photograph of the site on the right (B). ...9

Figure 4 Vegetation horizontal cross canopy cover and height distribution in 0.5 m classes for each transect at each site... 14

Figure 5 Landscape function of each site based on stability, infiltration and nutrient cycling indices. The two reference sites are represented by (R). ... 17

Figure 6 Three contrasting ecosystem rehabilitation trajectories (Tongway & Hindley 2004) ... 23

Figure 7 Suggested reporting system for rehabilitation trend assessment ... 23

List of Appendices

Appendix A Details of site characteristics at each location ... 26

Appendix B GPS coordinates for the start (0 m) and end (50 m) of each transect at all sites (GDA 94) ... 37

Appendix C Transect and photopoint locations within each site ... 38

Appendix D Overview of each transect taken from the 50 m point. ... 47

Appendix E Details of the photo-monitoring points at each of the locations surveyed. ... 52

Appendix F Details of methods applied by ALS for each of the components of the soil analyses ... 54

Appendix G Summary of Point Centre Quarter data from each transect at the 11 sites surveyed ... 56

Appendix H Presence/absence list of flora species at each transect within each site ... 57

Appendix I Summary of the indices derived from the soil surface assessments for each transect at all sites surveyed. ... 66

Appendix J Summary statistics from all soil analyses at all sites ... 67

(9)

1 Introduction

The Department of Environment and Natural Resources (DENR) Coorong, Lower lakes and Murray Mouth (CLLMM) Program is delivering a five year restoration project in the CLLMM region. As an emergency response to prolonged drought, bioremediation and revegetation activities were initiated in 2009. In July 2011, the project transitioned to building ecological resilience into the natural systems of the Lower Lakes and surrounding lands.

To assess the restoration activities in the CLLMM region, a monitoring plan was written up, which recommended the use of EFA to monitor the resilience of the newly established habitat areas.

A pilot study (this report) was commissioned to trial components of Ecosystem Function Analysis (EFA) at revegetation and reference sites. The goal of the pilot project is to track trends towards the long term objectives of the restoration program. The monitoring assessed the resilience of the ecosystems surveyed and identified future challenges of the program.

EFA is a monitoring procedure that uses quickly determined field indicators to assess the functional status of an ecosystem. The conceptual framework is based on the economy of vital resources. It focuses on processes that regulate spatial movement or use of water, topsoil and organic matter in the landscape (Tongway & Hindley 2004).

The EFA field methodology uses simple, visual indicators, which are closely related to a range of physical, chemical and biological processes. These take a few seconds per indicator to assess in the field after training (Tongway & Hindley 2004). The focus of EFA procedures is on landscape processes, not on any specific form of soil, vegetation or biota. Therefore, EFA can be implemented across many landscape types, uses and managements (Tongway &

Ludwig 2006).

EFA has been applied and verified across landscapes varying from sandy deserts to tropical rainforest and in different geological settings (Tongway & Hindley 2003).

One of the desired outcomes of this monitoring trial was to determine the appropriateness of using EFA to assess restoration activities in the CLLMM region.

Beyond trialling the method, the objective of this monitoring pilot was to gather baseline data, to track the trends towards the long-term objectives of the restoration program.

Specifically, the following questions have been addressed:

 What are the infiltration, stability and nutrient cycling index scores one year after revegetation site establishment?

 What are the cover and structure levels of the tree, shrub, herb and grass structure layers one year after revegetation site establishment?

 What are the infiltration, stability and nutrient cycling index scores at two reference sites and how do they compare to the relevant revegetation site scores?

(10)

2 Methods

The surveys were conducted at nine locations in the CLLMM region, specifically around Lakes Alexandrina and Albert. The components of the monitoring program implemented one year after revegetation site establishment consist of:

 Photo monitoring points for recording change to the landscape and the surrounding natural environment;

 Vegetation transects for measuring the establishment and growth of vegetation; and

 Soil surveys for tracking changes to soil functional status, soil fertility and the build up of selected metals.

2.1 Site characteristics and set-up

Locations chosen for the surveys varied in regards to vegetation type, aspect, geology and soil type. In total eleven sites were chosen to provide a representation of the spatial extent of the revegetation activities as well as representations of the varying habitat types of the region. Detailed descriptions of each site are provided in Appendix A. Surveys were undertaken from the 29^th of May to the 2^nd of June 2012.

The climate of the region is temperate with an annual median precipitation of 33.6 mm, most of which falls during winter and spring seasons (May-September) (BOM 2012).

Table 1 lists the sites that were chosen for the surveys and the number of habitats identified within each. Two reference locations (Bonney Reserve and Mulungushi) were chosen to provide data on both the effectiveness of EFA and to provide target values for the rehabilitated sites. Bonney Reserve is located next to Camp Coorong and represents a natural landscape for the region. This site has not been cleared. The Mulungushi location on Hindmarsh Island has several sections with different revegetation timelines. The section of site surveyed at Mulungushi was revegetated in 2005 however, some in filling of plants still occurs within this section of the site. Figure 1 identifies where all locations surveyed are within the CLLMM region.

Table 1 Restoration monitoring site list with number of habitats surveyed Site Number of Habitat Types

Mundoo 1

Mulungushi (Reference) 1

Finniss 1

Point Sturt (Upper & Lower) 2

Boggy Lake 1

Fiebig Reserve (Upper & Lower) 2

Narrung 1

Camp Coorong 1

Bonney Reserve (Reference) 1

At each site, three 50 m transects were established and marked at each end with an iron picket and yellow cap. A copper tag was also placed on each picket identifying the year, transect number and either the zero or 50 m end. Where possible, the start of the transect was established on the upward slope edge of the local watershed and were spatially

distributed to obtain a representation of the heterogeneity of the site. Coordinates for each transect are provided in Appendix B. Detailed maps of the location of each transect and photo-point within each site is illustrated in Appendix C.

(11)

A photograph of each transect was taken at the 50 m point to provide an overview of the habitat surveyed and to provide a baseline to detect changes over time (Appendix D).

At Point Sturt and Fiebig Reserve, revegetation occurred within two different vegetation types. These were treated as different sites and surveyed separately.

Point Sturt Upper was located on the high side of the site just below the escarpment and Point Sturt Lower within a samphire dominant zone. Fiebig Reserve Upper was located within a grass/samphire zone and Fiebig Reserve Lower within a samphire/rush zone. In total eleven sites at nine locations (33 transects) were surveyed across the region.

(12)

Figure 1 Overview of all locations surveyed in the CLLMM region.

(13)

2.2 Landscape photo-monitoring

Eleven photo monitoring points were established in order to record large scale landscape changes resulting from the revegetation work. The location of these photo monitoring points were recorded by handheld GPS (Appendix E) and marked with an iron picket and a yellow plastic cap. At two locations, Point Sturt and Narrung, the layout of the site determined that photo-points were established to provide an overview at two aspects. One

0 m picket was established and at two points (45^o) from each other a sighter post was located. A copper tag was placed on the 0 m picket identifying it as a photo-monitoring point, the year and 0 m mark. Appendix E identifies the location of the 0 m picket and description of its approximate location at each site.

2.3 Vegetation survey

EFA provides insights into how the landscape is functioning, vegetation establishment and habitat development. In successful rehabilitation, steady improvements are expected in vegetative cover, vegetation development and stability features. EFA data should gradually trend upward and plateau as the ecosystem becomes stable and self-sustaining. Results over time will verify if the ecosystems have achieved these self-sustaining levels and can withstand climatic fluctuations.

A Point-Centered Quarter (PCQ) method was employed to measure key vegetation

indicators. Sampling points were established at 5 m intervals. Measurements included plant cover (width and breadth of plant canopy), density (plants per ha) and diversity.

At each sampling point the sampling area divided into quarters by mentally placing a line perpendicular to the transect line. The width, breadth and species, and distance to the perennial plant nearest to the tape, were measured to obtain the plant cover index value, density and diversity for each quarter Figure 2.

* showing measurements at a single point.

Figure 2 The PCQ method for measuring spatial distribution of vegetation

a b

d ^Transect c

(14)

2.4 Landscape functional characteristics 2.4.1 Landscape organisation

The landscape characteristics, profile and vegetation patch characteristics were determined at or along each transect. The different landscape characteristics along each transect were used to assess the soil surface. Landscape organisation that relate to vegetation cover is defined as the arrangement of zones that reflect run-on and runoff processes.

2.4.2 Soil surface assessments

For each of the landscape characteristics identified, eleven indicators of soil surface

condition were visually assessed on three replicate 1 m transects within each patch. The soil surface indicators examine the status of a specific surface process and are assessed as described in Tongway and Hindley (2000).

Three indices reflecting the emergent soil properties of stability, infiltration and nutrient cycling were derived by compiling subsets of these eleven indicators (Table 2). Their values are expressed as a proportion of a total maximum score, converted to a percentage. These indices express the habitat quality and have significance for monitoring in terms of:

1. Stability

The ability of the soil to withstand erosive forces, and to reform after disturbance;

2. Infiltration

How the soil partitions rainfall into soil-water (water available for plants to use), and runoff water which is lost from the local system, or may transport material away; and

3. Nutrient cycling

How efficiently organic matter is cycled back into the soil.

Table 2 The combination of soil condition classes to derive indices of stability, infiltration and nutrient cycling

Indicator Stability Infiltration Nutrient cycling

1. Rainsplash protection 

2. Basal cover of perennial grass  

3. Litter cover, origin & degree of

decomposition  

4. Biological soil crust cover  

5. Physical crust broken-ness  6. Erosion type & severity 

7. Deposited materials 

8. Surface roughness  

9. Surface resistance to disturbance  

10. Slake test  

11. Soil texture 

(15)

2.4.3 Habitat complexity

As vegetation develops in size and diversity, environmental niches and habitat structure develop and become more complex for fauna. An increase in habitat size and structure allows for shade, shelter and food resources for fauna.

The habitat complexity index presented in the EFA manual (Tongway & Hindley 2004) is a recently added component of EFA and has not been as rigorously tested as the rest of the methodology. It is assesses on the basis of five features:

1. Canopy cover 2. Shrub cover

3. Ground vegetation cover

4. The amount of litter, fallen logs and rocks; and 5. Free water availability.

A modified version of the method was applied during the surveys. At each transect from the 0 m point an overview of the habitat was assessed for approximately 3 m either side of the transect. A habitat complexity score of between 0-3 were applied for each feature at each transect. This provided a broad overview of the habitat complexity at the site.

2.4.4 Soil chemistry

Soil was collected from the vegetation transects to establish a chemical (soil fertility and metal content) baseline. Thirty-two soil samples were collected at the end (50m) of each transect as this, in most instances, represented the downward end of run-off at each site.

After scrapping the surface vegetation and soil, four separate random samples were mixed together from the top 5cm and composited to one sample. Samples were collected and stored in pre-labelled glass jars and store in a cold ice cooler.

The soils were analysed by ALSE (NATA certified laboratory) for organic matter, electrical conductivity, pH, soil moisture, exchangeable cations (calcium, magnesium, potassium and sodium), minor anions (nitrate, nitrite, reactive phosphorus) and total metals (arsenic, cadmium, chromium, copper, lead, nickel and zinc).

Details of the laboratory methods for each analysis is presented in Appendix F.

Using the results of the exchangeable cations the exchangeable sodium percentage (ESP) was calculated to determine soil dispersion properties or the “sodicity” of the soils. The ESP is calculated as follows:

ESP = Exchangeable {(Na)/(Ca + Mg + K + Na)} x 100

The following classifications are used to characterise the ESP percentage: non-sodic <6%;

sodic 6 - 0%; moderately sodic 10 - 15%; strongly sodic 15 - 25%; and very strongly sodic

>25%.

(16)

3 Results and Discussion

3.1 Site characteristics

Sites varied in regards to their vegetation type, slope, soil type and aspect. A detailed assessment of each site is provided in Appendix A.

At four sites, vegetation comprised primarily of saltmarsh species. These included Finniss, Point Sturt Lower, Boggy Lake Reserve, and Fiebig reserve. The sites were situated close to the lakes and were low lying. The soil type at these sites was heavy dark clay. These sites contained few weeds but in some sections, dense kikuyu was observed. On the edges of the samphire zones, shrubby sections were present.

Mundoo and Narrung were low rising sites with fringing samphire zones. Soil type is characterised dark heavy clay within the samphire zones and the grassy zones were sand- loam soils.

The remainder of the sites ranged from dense multi-layered vegetation at the two reference sites and open grass and shrub land at the other sites. The slope at these sites ranged from mildly undulating (e.g. Mulungushi and Camp Coorong) to steep (e.g. Point Sturt Upper and parts of Finniss). The steeper and more undulating sites varied in soil type but generally contained coarser sediment and sand.

Most of the sites have had grazing pressure from stock removed. Occasionally, stock enter Boggy Reserve and may affect revegetated and established plants. Rabbits activity were noted at Boggy reserve, which may be a factor in the revegetation survivorship. Rabbits may be present at other sites but were not observed during the survey.

(17)

3.2 Landscape photo-monitoring

A photographic record of each site is provided in Figure 3. Photographs were taken to allow for a broad overview of the areas surveyed at each site.

Mundoo 1 Mundoo 1B

Mulungushi 1 Mulungushi 1B

Figure 3 Photographic record of each site taken from permanent locations. Photograph on the left is taken at the widest zoom and a close-up photograph of the site on the right (B).

(18)

Finniss 1 Finniss 1B

Point Sturt 1 Point Sturt 1B

Point Sturt 2 Point Sturt 2B

Figure 3 Photographic record of each site taken from permanent locations. Photograph on the left is taken at the widest zoom and a close-up photograph of the site on the right (B).

(continued)

(19)

Boggy Lake Reserve 1 Boggy Lake Reserve 1B

Fiebig Reserve 1 Fiebig Reserve 1B

Narrung 1 Narrung 1B

(continued)

(20)

Narrung 2 Narrung 2B

Camp Coorong 1 Bonney Reserve 1

Bonney Reserve 1B

(continued)

(21)

3.3 Vegetation survey

Figure 4 summarises the vegetative cover in square metres per cover per hectare (m²ha^-1) resolved into 0.5 m height slices for each transect. Vegetation cover was variable within each site yet all sites were dominated by ground and shrub cover with a large proportion less than 2 m in height. The dominance of ground cover is important for the flow of resources across the ground surface by providing areas of resource mobilisation and transport. However, canopy cover is an important component of the habitat providing shelter for fauna as well as protection from erosive forces (wind and rainfall).

Canopy cover was minimal at the sites measured and is mainly a reflection of the site use history. Canopy cover of approximately over 5 m in height was mostly recorded from one transect at the reference site, Mulungushi (Figure 4). Although not recorded from the PCQ measurements canopy cover was dominant around transect 2 at Bonney Reserve. These were not recorded as the nearest plant measured was mostly low shrubs.

Bonney Reserve, a reference site, recorded the most median number of plants per hectare (11,0942 ha^-1 ±SD 60,640). This site contained a complex habitat structure, from ground cover to mature canopy layers. Bonney Reserve also recorded the highest species diversity of 33 species.

The lowest density of plants was recorded from Point Sturt Upper (658 plants ha^-1±SD 237).

Many of the planted species were dead at this section of Point Sturt. Very few mature species were present, numerically demonstrated by the higher distance between plants measured.

Appendix G summarises the data recorded during the PCQ surveys. The lowest species diversity was recorded at Fiebig Reserve Lower and was as a result from the dominance of samphire at this site. The full list of the species identified and their presence within each site is in Appendix H.

Point Sturt Lower and Boggy Lake Reserve were dominated by saltmarsh species. Given that these habitats rarely contain tall plants, it is not expected that the canopy cover

measurements will increase over time. Canopy cover will increase on the fringes of these zones where the habitat and soil type is more suitable for trees to establish.

Very few of the surveyed plants appeared grazed. Of all the plants measured only 1.7%

were noted as having grazing pressure. Camp Coorong contained the highest number of grazing affected plants with Allocasuarina verticillata heavily affected, of which some were infested with mealy bugs. Grazing pressure on Melaleuca halmaturorum was noted from Fiebig Reserve Upper and Mundoo. Very few individual plants were noted within Bonney Reserve and this included Boronia sp.

Ground cover is important for the functionality of a habitat and flow of resources. It will be important that future monitoring look for trends in vegetation cover including an increase in ground and canopy cover at the sites.

(22)

Mundoo Mulungushi-Reference

Finniss Point Sturt Upper

Point Sturt Lower Boggy Lake

Figure 4 Vegetation horizontal cross canopy cover and height distribution in 0.5 m classes for each transect at each site

0 1000 2000 3000 4000

0.5 1 1.5 2 0.51 1.5 2 0.5 1 1.5 2

123

Vegetation Horizontal Cross Section Surface Area (m²ha^-1)

Transect\Height Class (m)

0 500 1000 1500 2000 2500 3000 3500 4000

0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5

123

Vegetation Horizontal Cross Section Surface Area (m²ha^-1)

Transect\Height Class (m)

0 1000 2000 3000 4000

0.5 1 1.5 2 0.5 1 1.52 0.5 1 1.5 2

123

0 1000 2000 3000 4000

0.5 1.5 2.5 3.5 1 2 3 0.5 1.5 2.5 3.5

123

0 1000 2000 3000 4000

0.5 1 1.5 2 0.5 1 1.5 2 0.5 1 1.52

123

0 1000 2000 3000 4000

0.5 1 1.5 2 0.5 1 1.5 2 0.5 1 1.5 2

123

(23)

Fiebig Reserve Upper Fiebig Reserve Lower

Narrung Camp Coorong

Bonney Reserve

Figure 4 Vegetation horizontal cross canopy cover and height distribution in 0.5 m classes for each transect at each site (continued)

0 1000 2000 3000 4000

0.5 1.5 2.5 3.5 1 2 3 0.5 1.5 2.5 3.5

123

0 1000 2000 3000 4000

0.5 1.5 2.5 1 2 0.5 1.5 2.5

123

0 1000 2000 3000 4000

0.5 1 1.5 0.5 1 1.5 0.5 1 1.5

123

0 1000 2000 3000 4000

0.5 1 1.5 0.5 1 1.5 0.5 1 1.5

123

0 2000 4000 6000 8000 10000 12000 14000 16000

0.5 1.5 2.5 3.5 0.5 1.5 2.5 3.5 0.5 1.5 2.5 3.5

123

Vegetation Horizontal Cross Section Surface Area (m²ha^-1)

(24)

3.4 Landscape functional characteristics 3.4.1 Landscape function

Soil surface characteristics involved in the assessment of infiltration index include; perennial grass basal cover and shrub foliage cover, litter cover, soil surface nature, surface resistance to disturbance, slake test, and soil texture.

The infiltration index indicates how quickly water will permeate into the soil profile, not how much water the soil will store. A sandy soil often has a high infiltration rate, but very low soil moisture storage.

To evaluate soil moisture retention, we require the stability index to describe the soil profile characteristics that relate to water storage capacity, which depends on depth of profile, soil texture of whole profile, and gravel content (Ata Rezaei et al. 2006).

The value of stability index is obtained from several observations of the soil surface, but a high stability index does not necessarily always mean that the site has high production potential. If a high stability index value coincides with high nutrient cycling index and landscape organisation index then the high stability index can be associated with extensive vegetation cover, reflecting high soil productivity (Ata Rezaei et al. 2006). Although it is a useful index to assess soil stability, it is not simply related to soil productivity and plant production.

Figure 5 summarises each index for each rehabilitated site surveyed and indices are compared with the two reference sites, Mulungushi and Bonney Reserve. The foremost trend to note is that, on the whole, the revegetation sites are already functioning on a similar level to the reference sites. With on-going monitoring the primary concern will be that the landscape function within a site remains at or increases from the baseline values and remain comparable to reference sites. The score for each index at each transect is provided in Appendix I.

The stability index is relatively high across all sites with values ranging between 42.27 % at Mulungushi and 56.44 % at Mundoo. The higher value at a rehabilitated site compared to a reference was from a higher contribution from perennial cover and hence rainsplash

protection. Mulungushi contained more loose sandy soils and very low ground cover of perennials. With increasing vegetation, plant litter should also increase therefore influencing the stability index.

The infiltration index was more variable between the sites surveyed. Values ranged between 23.77 % at Boggy Lake and 43.63 % at Mundoo. Again the higher cover of perennial

grasses at Mundoo contributed to the higher value. Boggy Lake, a samphire dominated site, had almost no perennial or litter cover. Soil texture, which influences the infiltration index, was at several sites (e.g. Mulungushi, Bonney reserve and Point Sturt Upper) comprised of sandy soil, which was not coherent. Overall, perennial grasses were low in cover at many sites and it can be recommended that revegetation also includes low lying grasses.

The nutrient cycling index is influenced by three indicators: cover of perennial grass, biological crust cover and surface roughness. Values recorded ranged between 16.77 % at Boggy Lake and 29.47 % at Bonney Reserve. The three indicators scored highly at all transects at Bonney reserve. Again due to the dominance of samphire at Boggy Lake and lack of habitat diversity nutrient cycling scored very low.

(25)

Figure 5 Landscape function of each site based on stability, infiltration and nutrient cycling indices. The two reference sites are represented by (R).

3.4.2 Habitat complexity

Habitat complexity along each transect was measured through visual observation of the overall habitat. Table 3 summarises the mean score for each structure observed and the mean total score for each site. Bonney Reserve recorded the highest score and was the only site to score on each structure index. Most other sites scored very low and contained few structural features.

It rained during the survey week. This influenced soil moisture at many of the sites.

While most indicators are measured during the soil surface assessments, there are often patches of the site that can be missed. Therefore, we assessed the overall habitat complexity of each site and/or transect, to pick up any indicators that were not scored during the soil surface assessments.

As the revegetated plants grow, each habitat complexity score should increase. After further monitoring, it will be possible to alter some of the habitat complexity indicators to be more specific to the region.

Indicators in future surveys can include the following:

 Groundcover--vines, creepers, cryptogams

 Under-storey-grasses, herbs, 0-0.1m

 Mid-storey-small shrubs, 0.5-1.5m

 Upper-storey-tall shrubs, 1.5-3.0m

 Over-storey- >3.0m

0 10 20 30 40 50 60 70

Mu n d o o Fi n n is s P t Stu rt Up p er P t Stu rt Lo w er B oggy La ke Fi eb ig Upp er Fi eb ig Lo w er N arrun g C am p C ooro n g Mu lun gu sh i ( R ) B onn ey R eserv e (R )

Lan ds cap e Fu nct ion %

Stability Infiltration Nutrient Cycling

(26)

 Ants/other fauna (score on species and abundance)

 Scats

 Water availability

Table 3 Summary statistics for habitat complexity scores taken from each transect at each site.

Mundoo Mulungushi Finniss Point Sturt

Upper Point Sturt Lower Structure Mean SD Mean SD Mean SD Mean SD Mean SD

Tree Canopy (%) 0 0 1.67 0.58 0 0 0 0 0 0

Shrub Canopy (%) 0 0 1.33 0.58 1 0 2.33 1.15 0 0

Ground herbage 1 0 0 0 1 0 1 0 1.33 1.15

Logs, rocks, debris

etc (%) 0 0 0 0 0 0 1 0 0.33 0.58

Soil Moisture 1 0 1 0 1 0 1 0 2.33 1.15

Mean total

(max. 12) 2 0 4 1 3 0 5.33 1.15 4 1

Boggy Lake

Fiebig Reserve

Upper

Fiebig Reserve

Lower

Narrung Camp

Coorong Bonney Reserve Structure Mean SD Mea

n SD Mea

n SD

Tree Canopy (%) 0 0 0 0 0 0 0 0 0 0 1.33 0.58

Shrub Canopy (%) 0 0 0 0 0 0 1 0 0 0 1.67 0.58

Ground herbage 2 0 2 0 2 0 1 0 0 0 1.67 0.58

Logs, rocks, debris

etc (%) 0 0 0 0 0 0 0 0 0 0 1.33 0.58

Soil Moisture 3 0 1 0 1 0 1 0 1 0 1 0

Mean total

(max. 12) 5 0 3 0 3 0 3 0 1 0 7 0

3.4.3 Soil chemistry

This section contains some discussion about the results of soil chemical testing. Summary statistics of the results from the soil analyses are in Appendix J.

pH

The pH of soil indicates the strength of acidity or alkalinity in the soil solution that affects plant growth, soil constituents, and soil micro-organisms. Soil is neutral when pH is 7, it is acid when pH is less than 7 and alkaline when it is greater than 7.

Median soil pH across the sites varied from 6.8 to 8.5. Overall soil pH was moderately to strongly alkaline (>7) at most sites except for Finniss and Bonney reserve where pH levels were slightly acidic at just below 7.

Electrical Conductivity

High salt levels can adversely affect plant growth, soil structure, water quality and

infrastructure. Soil salinity was variable across all sites and measured between a median of 35 and 5630 µS/cm.

(27)

Point Sturt Lower and Boggy Lake were moderately saline and as such dominated by salt tolerant plants, such as samphires. These sites were characterised by heavy loam soils. All other sites were classified as non-saline.

Moisture Content

The amount of water that can be stored in soil and evaporated or used by plants is an important indicator for the production and health of vegetation. Soil moisture is dependent on soil type with coarse, sandy soils holding less water than heavy silty clay soils.

Soil moisture varied widely between the sites sampled with a median between 5 and 32.2

%. Lower soil moisture values were recorded at Camp Coorong, Bonney reserve, and Mulungushi which were sites characterised by sandy soils. Whilst sites with more silt and loam such as Boggy Lake, Point Sturt and Lake Albert recorded higher soil moisture values.

Total Organic Content

Organic matter contributes to soil fertility by increasing available nitrogen and minerals. In addition to providing nutrients and habitat to organisms living in the soil, organic matter also binds soil particles into aggregates and improves the water holding capacity of soil.

The median organic content varied between 0.47 and 8.02 %. Higher organic matter values were recorded for sites with loamy soils (e.g. Bonney reserve, Point Sturt Upper and Fiebig Reserve Lower) whilst sites with sandy soils (e.g. Finniss, Mulungushi and Camp Coorong) recorded lower organic matter (median <2 %).

Total Heavy Metals

Analyses of total metals included arsenic, cadmium, chromium, copper, lead, nickel and zinc.

All soils naturally contain trace levels of metals. The presence of metals in soils is not necessarily indicative of contamination but can be related to the geology of the parent material from which the soil was formed.

Levels recorded from the samples were compared against NEPC guidelines for soil ecological investigation levels (NEPC 1999). All seven metals were well below guideline values at all sites.

Nutrients

Plant nutrients in soil come originally from the parent material from which the soil was formed. Nutrients analysed included nitrite, nitrate and phosphorus. Of all the essential nutrients, nitrogen is required by plants in the largest quantity and is most frequently the limiting factor in plant productivity. After nitrogen, phosphorus is the most important nutrient element for plant growth. Whilst most of the phosphorus in soils is mineralised, reactive Phosphorus, which was tested, is that which is available to plants.

Total oxidised nitrogen (nitrite + nitrate) varied between and within sites. Overall, median values varied between 0.1 and 14.2 mg/kg. Higher values were recorded from damp, loamy soils. Median reactive phosphorus ranged between 0.5 and 40 mg/kg. Lower values were recorded from both reference sites that were dense with native vegetation whilst sites dominated with samphire obtained higher phosphorus levels. Higher levels are due to two

(28)

reasons – the breakdown of the blue-green algal (cyanobacterial) mat and the fact that it is a depositional environment.

Exchangeable cations

This is a measure of the ability of the soil to hold and exchange nutrients and essential elements with plants, particularly the nutrients calcium, magnesium and potassium. Good fertile soils with high clay content and moderate to high organic matter levels usually have a cation exchange capacity of 10 or higher. The major cations are calcium, magnesium, potassium, sodium. These are held in the soil by organic matter and clay.

All exchangeable cations recorded from Camp Coorong were in low concentrations. At Point Sturt and Fiebig Reserve values exchangeable cation concentrations were high. A high percentage of exchangeable sodium can cause soil structural dysfunction through clay dispersal and very low rates of hydraulic conductivity.

At the Boggy Lake and Point Sturt Lower, soil salinity and exchangeable sodium levels where high enough to interfere with plant growth. This is reflected in the dominance of samphire at both sites.

The exchangeable sodium percentage (ESP) was calculated to determine the sodicity of the soils. A sodic soil, by definition, contains a high level of sodium relative to the other

exchangeable cations (i.e. calcium, magnesium and potassium). In sodic soils, much of the chlorine has been washed away which cause clay particles to lose their tendency to stick together when wet. As a result, sodic soils may affect plant growth and such soils tend to develop poor structure and drainage over time.

Soils measured varied from non-sodic (Camp Coorong, Bonney Reserve, Point Sturt Upper, Mulungushi and Narrung) to very strongly sodic (Finniss, Point Sturt Lower and Boggy Lake).

Mundoo was classified as sodic, Fiebig Reserve Upper was strongly sodic and Fiebig reserve Lower moderately sodic. The very strongly sodic sites were samphire dominated sites and may be a natural reflection of the nature of the soils in such habitats.

The relationship between electrical conductivity and ESP (EC:ESP) can determine the possible effects of salinity and available sodium on plant growth. A relationship of the EC:ESP ratio was obtained Point Sturt Lower and Boggy Lake indicating both saline and sodic soils. This is a reflection of the natural state of these ecosystems. Plant selection for such sites need to be able to withstand the conditions.

(29)

4 Comparison between revegetation and reference sites

Revegetation sites varied from samphire dominated sites to open grassy habitats.

Vegetation structure comprised mostly of grasses and low shrubs with minimal to no canopy present. Litter cover was generally very low and cover of perennial grasses was nearly absent at some sites. Soil type at the revegetation sites varied widely depending of the vegetation present. Soils varied from heavy clay soils at the samphire sites to loose sandy soils at sites such as Narrung. Soil salinity, organic matter, and nutrients were higher at some of the revegetation sites. The following provides a direct comparison between the reference sites and adjacent revegetation sites.

Mundoo and Mulungushi

Mundoo was surveyed one year after restoration works whilst Mulungushi represents a revegetated site after five years. The comparison between these two sites indicates what to expect from ongoing monitoring over a four-year period.

Mundoo was primarily an open grassland habitat with a high ground cover of weeds. Litter cover and cover of perennial grasses was low. The surface area of vegetation cover was one of the lowest in comparison to other revegetation sites and clearly lower than that of

Mulungushi. However, plants of less than 1 m dominated vegetation cover. As such, habitat complexity was very low. Only ground herbage and soil moisture contributed to habitat complexity scores.

There were clear differences in landscape and habitat function at the reference site,

Mulungushi. The area of Mulungushi surveyed represented an open woodland with grasses, shrubs and canopy cover. Litter cover and cover of perennial grasses was much higher in comparison to Mundoo. The surface area of vegetation cover comprised of various canopy heights. The retention of a diversity of canopy heights is an important factor to track over time. Habitat complexity scored well for tree and shrub canopy yet ground herbage was low.

Non-natives mostly dominated ground cover.

The landscape function indices were higher at Mundoo compared with Mulungushi. This was primarily a factor of the difference in vegetation and soil types between the sites. Sandier soils were recorded at Mulungushi. As the vegetation matures and hence influences the landscape function indices at Mundoo it may be possible that indices will be comparable with those of Mulungushi. Prior to restoration works, Mulungushi was an open paddock

dominated by non-native ground cover, similar to Mundoo. Through ongoing monitoring and comparisons of these sites, it will possible to detect how habitat complexity influences landscape function.

Camp Coorong and Bonney Reserve

Camp Coorong was surveyed one year after restoration works whilst Bonney Reserve represents remnant vegetation for the region. Bonney Reserve provides a long-term target for the restoration program within the region.

Camp Coorong scored very low on habitat complexity with only soil moisture a factor. The tree planting methods employed has heavily modified the landscape. The surface area of

(30)

vegetation cover was from plants of less than 1 m in height. Much of the sections surveyed showed signs of erosion with minimal litter and perennial cover to protect the soils.

Bonney Reserve contained high-density vegetation and a complex habitat structure. Very few non-native species were observed. Litter cover and cover of perennial grasses scored much higher compared to Camp Coorong. The soil type at Bonney Reserve comprised mostly of sandy soils with low organic matter and low levels of nutrients. The higher score from habitat complexity provides a long-term target for Camp Coorong.

There were clear measured differences between the two reference sites and their adjacent revegetation sites. These differences illustrate that EFA is effective at detecting differences between sites and measuring landscape function and vegetation structure. The complex habitat structure and soil surface qualities were reflected in the results of the reference sites. Similarly, the nature of the ecosystems of the restoration sites was observed through the results. The indices from reference sites provide good long-term targets for restoration activities.

The selection of the reference sites was appropriate to compare with results from Narrung, Camp Coorong, Point Sturt Upper, and sections of Mundoo. The vegetation type, habitat structure and soil type at the reference sites reflected the landscape function targeted for the revegetation sites. For the low-lying samphire dominated sites, appropriate reference sites will reflect the vegetation and habitat structure characterising these sites. At present the reference sites are only appropriate for a selection of the revegetation sites surveyed.

5 Recommendations

The objective of this pilot was to primarily establish a baseline survey of the restoration activities, while determining the functional status of the ecosystem at each location. Whilst comparisons from the data can be made between the sites, habitat resilience will only be determined with on-going monitoring.

Monitoring of the sites should be undertaken on a yearly basis until the data warrant a longer period, to say every 3–5 years. If data are collected regularly, a time series record of ecosystem change or development is provided. By comparing data with reference sites, it is possible to see if the disturbed site is developing adequately.

As already observed from the results, revegetation sites are functioning on a similar level to the reference sites. However, long-term monitoring will have to assess that within each site this landscape function remains at baseline levels or even increases. Appropriate reference sites will need to be chosen for the low-lying samphire sites.

Figure 6 illustrates three potential scenarios from which to assess results obtained from long-term monitoring. When indices are plotted over time, it is possible to analyse the future likelihood of the rehabilitation. Curve A represents an appropriate trajectory shape, implying that the rehabilitation is on-track and no problems have been identified. It is characterised by a steep initial response followed by a steady increase over time. Curve B shows that potential problems have been identified that need further analysis. Curve C shows problems are identified that need urgent attention. All indices at a given site should exceed the critical threshold value if ecosystem rehabilitation is to be judged successful (Tongway and Hindley, 2004). There is no minimum time limit attached to EFA monitoring of rehabilitation. The sigmoidal curve of the results over time is of more significance than the actual individual yearly data.

(31)

Figure 6 Three contrasting ecosystem rehabilitation trajectories (Tongway & Hindley 2004)

For future reporting, the conceptual application of rehabilitation trajectories can be

simplified to Figure 7. With on-going monitoring a trend assessment of rehabilitation can be developed where management actions can be determined through the following system:

There is an urgent problem needing attention.

There is a possible problem. Use a more rigorous method to ascertain its seriousness.

There is no discernible problem with the trend.

Figure 7 Suggested reporting system for rehabilitation trend assessment

Further recommendations for reporting on results of future monitoring assessments include:

 how biological processes are becoming more prominent and ultimately dominant;

 how erosion, sedimentation and litter removal are declining, ultimately to non- discernible levels;

 how soil aggregates no longer slake; and

 how the structure, composition and function of the vegetation is developing.

Further work can be done to maximise rehabilitation and therefore ensuring that basic ecological function is maintained. Management actions could include:

 Retaining woody debris and increasing perennial vegetation cover to retain resources and protect the soil;

 Reducing any remaining grazing pressure from stock to allow native plants to set seed and grow beyond browsing height; and

 Provide any additional protection where successful germination and establishment occurs.

In future, assessments can be done to determine the habitat available for fauna and the status of fauna in the ecosystem. As vegetation becomes larger and more diverse, the site as a whole often develops to be more suitable for fauna (Tongway and Hindley, 2004). This can be done through the habitat complexity method which can be further developed for the region.

(32)

There is a potential issue of the scale of the site and location of transects. As the location of the transects are fixed over time, there may be issues over the whole site that are not being captured during the EFA surveys. As such, regular site inspections are recommended. These can incorporate the habitat complexity method on regular intervals to provide an overview of the whole site’s function.

The use of LIDAR (Light Detection and Ranging) can be a innovative and useful tool in rehabilitation projects. This method can also overcome the issue of scale by providing a whole site assessment. However, it would not be a replacement for the use of EFA, which provides fine-scale landscape function assessment. LIDAR provides high accuracy maps of the surface of the sites to provide a three-dimensional assessment of erosion. It provides data on basic vegetation height, vegetation strata, biomass and vegetation cover. It is recommended that this can be done every five years, as this will be more efficient at detecting trends.

6 Conclusion

The maintenance of diverse and healthy native vegetation communities requires both active and passive management based on informed decisions relating to habitat quality. Applying EFA techniques to the revegetation sites provides an opportunity to achieve best-practice rehabilitation assessment and monitoring. It is a scientifically valid method of quantifying rehabilitation success. The results presented provide a solid baseline to compare with for future monitoring.

Important information gathered during the surveys includes an overview of the landscape function, vegetation structure and soil health at each site. It is evident that there are distinct differences between the sites mainly due to the natural characteristics present. For example, samphire dominated sites will have distinctive differences in landscape function to woodland or grassy sites. Comparisons between such sites may not be feasible and therefore

comparisons can only be made of the ecosystem function within a site over time. It is recommended that more appropriate reference sites be selected for some of the samphire revegetation sites.

The use of EFA at the restoration sites has proved effective at providing a baseline of the current landscape structure and function. For the purposes of measuring restoration activities, EFA has been successful at assessing the vegetation structure and status of the soil surface at each site. With ongoing monitoring EFA will test the changes within a site and assess how indices trend with maturing vegetation. The method will be able to detect over time if the landscape is subject to stress and disturbance. The goal is to follow the trends of the EFA indicators to detect a time when the landscape has become self-sustaining as an ecosystem. The effectiveness of EFA to assess and monitor revegetation activities of the CLLMM region can only be determined with on-going monitoring.

(33)

7 References

Ata Rezaei, S., Arzani, H., Tongway, D. 2006. Assessing rangeland capability in Iran using landscape function indices based on soil surface attributes. Journal of Arid Environments 65:

460-473.

BOM (Bureau of Meteorology). 2012. Climate Data Online. Commonwealth of Australia. URL:

http://www.bom.gov.au/climate/data/. Accessed: 18 June 2012.

NEPC (National Environmental Protection Council) .1999. National Environment Protection (Assessment of Site Contamination) Measure 1999.

Tongway, D.J. and Hindley, N.L. 2000. Ecosystem Function Analysis of Rangeland

Monitoring Data: Rangelands Audit Project 1.1, National Land and Water Resources Audit, Canberra 35 p.

Tongway, D. and Hindley, N. 2003. Indicators of Ecosystem Rehabilitation Success. Stage 2 Verification of EFA indicators. http://www.cse.csiro.au/research/program3/efa/

Tongway, D. and Hindley, N. 2004. Landscape function analysis: a system for monitoring rangeland function. African Journal of Range & Forage Science 21(2): 109-113.

Tongway, D and Ludwig, J. 2007. Resource retention and ecological function as restoration targets in semi-arid Australia. In: Restoring natural capital: Science, Business and Practise (Eds J. A. Aronson, S. J. Milton, and J. Blignaut). Washington, D.C.: Society for Ecological Restoration International and Island Press. (1323).

(34)

Appendices

Appendix A Details of site characteristics at each location

Site No. 1 Date 29/05/12

Site Name Mundoo Observers AC, SS, RW, JB

Transect 1 Transect 2 Transect 3

Position (GPS)

(0m) 0307449/6064504 0307438/6064578 0307495/6064630 Transect Compass

Bearing 54ô 48ô 200ô

Position in Landscape

SW corner 50m from

gate NW of entrance gate Close to samphire zone Lithology Fine sediment. Light

brown colour Fine sediment. Light

brown colour Fine sediment. Light brown colour Soils

(Texture/Fraction) Medium/Sand Medium/Sand Medium/Sand

Slope Flat Flat Flat

Aspect NE NE SSW

Vegetation Type Open grassland Open grassland Open grassland

Landuse Revegetation Revegetation Revegetation

State of Soil

Surface Intact/Sandy Intact/Sandy Intact/Sandy

Comments

(35)

Appendix A Details of site characteristics at each location (continued)

Site No. 2 Date 01/06/12

Site Name Mulungushi Observers AC, JB

Transect 1 Transect 2 Transect 3

Position (GPS)

(0m) 0304107/6068021 0304108/6068095 0304125/6068127 Transect Compass

Bearing 36ô 45ô 290ô

Position in

Landscape Near stand of Callitris, heading towards main

house. Open scrubland. Within Allocasuarina

patch. Close to main road.

Lithology Light yellow/brown sand Light yellow/brown sand Light yellow/brown sand

Soils

(Texture/Fraction) Light-medium/Sand Light-medium/Sand Light-medium/Sand

Slope Undulating Undulating Undulating

Aspect NNE NE NW

Vegetation Type Open woodland Open woodland Open woodland Landuse Revegetation (2007) Revegetation (2007) Revegetation (2007) State of Soil

Surface Intact/Sandy Sandy/some evidence

of borrows Sandy/some evidence of borrows

Comments