Title

Lord Howe Island Vegetation Map 2016 VIS_ID 4450

Data Quality Statement

Name: Data Quality Statement

Protocol: WWW:DOWNLOAD-1.0-http--download

Description:

DQS for Lord Howe Island vegetation map, 2016

Function: download

Lord Howe Island Vegetation map data package

Name: Lord Howe Island Vegetation map data package

Protocol: WWW:DOWNLOAD-1.0-http--download

Description:

Download ZIP Package Data

Function: download

Code

32be1f18-346b-42ab-b026-1d6821394806

Name

ANZLIC Metadata Profile: An Australian/New Zealand Profile of AS/NZS ISO 19115:2005, Geographic information - Metadata

Version

1.1

Type

vector

Authority code

GDA94 Geographic (Lat\Long)

Code identifying the spatial reference system

4283

Topic category

Title

ANZLIC Search Words

Reference date

2008-05-16

NSW Place Name

Lord Howe Island

Coordinate reference system

Authority code

urn:ogc:def:cs:EPSG::

Code identifying the coordinate reference system

5711

End position

N/A

Contact info

Organisation name

Lord Howe Island Board

Responsible party role

pointOfContact

Lineage

Airborne Digital Sensor (ADS40 by Leica Geosystems) data at 10cm resolution became available from NSW Land and Property Information (LPI) in January 2012. Aerial Photography Interpretation (API) was undertaken for this project using ArcGIS 9.0 and ArcGIS 10.1 with an orthorectified stereo image (or without a stereo view) in normal colour and an enhanced image which was stretched in the 600 to 700nm range to improve resolution of patterns in the vegetation. Using the digital version of the Pickard (1983) vegetation map as a template, the ADS40 imagery was used to refine the spatial accuracy of vegetation extent of Pickard’s linework. This included trimming areas of the Pickard linework where it extended beyond extant vegetation or landmass, and adding areas such as isolated trees and smaller remnants of vegetation which were not originally mapped by Pickard. This produced a refined draft map incorporating Pickard’s polygons and attributions. A stratified random sampling design was applied based upon the assumption that vegetation types broadly represent a surrogate for underlying environmental variables. The number of sites to be sampled within each vegetation type was weighted by area. A minimum of one site was allocated for vegetation types that occupied small areas, and up to seven full floristic sites and up to 10 rapid floristic sites were allocated to more extensive vegetation types. Not all sites that were selected were surveyed, and the actual number of sites completed in some vegetation types varied from the number allocated, due to access and time constraints. Sampling of floristic sites was undertaken during July 2013. Floristic data were gathered from 20m X 20m quadrats positioned as close as possible to the pre-selected site location. Biophysical information including slope, aspect, geology, lithology and evidence of disturbance were recorded. Vegetation structural information (height range, dominant species, foliage cover) was recorded for each discernible vegetation stratum, and all vascular plant species present within the quadrat were recorded and assigned a modified Braun-Blanquet (1932, cited in Conrad and Fuller 1983) cover-abundance score between 1 (<5%) and 6 (76-100%). Rapid floristic sites were similar to full floristic sites in positioning of the site and collection of locality information. However, the only floristic data recorded were up to six dominant species in the upper tree stratum, and up to three dominant species in each lower stratum, along with an estimate of the percentage foliage cover of each stratum. Data from 86 full floristic and 105 rapid floristic (canopy taxa only) sites were investigated using a hierarchical agglomerative clustering strategy available in PATN (Belbin 1990, 1995) to determine the main floristic groups for the study area. Separate full-floristic and canopy-only (combining canopy data from all sites) analyses were undertaken. Exotic taxa were included in the full floristic analysis. The Bray-Curtis (Bray & Curtis 1957) association measure was used to determine site similarity. A hierarchical classification of sites was derived from a clustering strategy using a Beta value of -1 in a flexible unweighted pair group arithmetic averaging (UPGMA). A nearest-neighbour analysis was then used to identify potentially misclassified sites, and a fidelity analysis applied to the resultant floristic groups, arising from the full floristic analysis, to identify diagnostic (indicator) species for each group (Bedward 1999). A complete vegetation community list for the LHIG was compiled by merging or splitting full floristic and canopy-only groups, and adding those vegetation types or mapping units that were not sampled during the current survey but were recognised and described in previous surveys, viz. Pickard (1983), Hutton (2001) and DECC (2007). Detailed profiles of each LHIG vegetation community sampled and recognised in this study, and for which floristic and physiognomic data are available, were prepared.
222 plant taxa were recorded in floristic sites, including three threatened species and 47 exotic species. Twenty-four plant communities were derived at a dissimilarity measure of 0.65. A twenty-fifth group was excluded as it represented recent post-disturbance regeneration (see Figure 2). Two major groupings, split on the basis of relative species richness, were apparent in the full floristic dendrogram. The first grouping comprised communities with low species richness, including coastal grasslands and sedgelands, swamp forests and saltmarsh, and fernlands and shrublands of exposed, rocky areas. The second grouping contained species-rich oceanic rainforests of the lowlands and southern and northern hills, and montane cloud forests of the southern mountains. Analysis of canopy-only data from all 191 floristic sites resulted in the recognition of three additional communities – Group 3 Grey Mangrove low estuarine forest, Group 4 River Mangrove low estuarine forest, and Group 19 Pandanus - Metrosideros riparian forest.
A saltmarsh (Community 7 Saltwater Couch saltmarsh of poorly drained, brackish flats) and two rainforest communities (Community 19 Maulwood-Kentia Palm- Cottonwood-Greybark lowland forest; Community 22 Hill Rose – Forky Tree forest of rocky creeks and slopes) that were not recognized in previous studies were identified and mapped in the current study.
This survey resulted in the recognition of 39 mapping units for the LHIG. Including: • 33 described vegetation communities from the full floristic analysis, canopy-only analysis, and additional Pickard (1983), Hutton (2001) and DECC (2007)communities • 3 map units identifying non-specific native vegetation (Environmental Plantings, Native Regeneration and Native Remnant) • 2 map units identifying non-native vegetation (Plantations and Exotic)
• 7 map units representing physiographic features (Estuary, Landslip, Rock, Beach, Water , Cliff, Dam) Waterfall communities are described as occurring on vertical cliffs on the north and west faces of Mount Gower, down to around 500m altitude (DECC 2007). but these communities are not included in this map and report, as they are of extremely restricted occurrence, generally inaccessible, and cannot be detected using Aerial Photography Interpretation (API).

Limitations on public access

Contact position

Data Broker

Organisation name

Lord Howe Island Board

Responsible party role

pointOfContact

Contact position

Data Broker

Organisation name

Lord Howe Island Board

Responsible party role

distributor