Abstract

The projected distributions of mangrove and saltmarsh in this dataset were developed using the method described in Hughes et al (2022) which can be downloaded from https://www.frontiersin.org/articles/10.3389/fenvs.2022.749950/full

The method involved the use of machine learning techniques to develop a statistical classification model for predicting the position of mangrove and saltmarsh in the present-day landscape in terms of topographic, hydrological and sedimentological predictor variables. That model was then used to identify and map positions in the future landscape displaying similar attributes. Three future landscapes were modelled based on the three SLR scenarios +0.5 m, +1.0 m and +1.5m. To bracket the potential outcomes three sets of projections were prepared. One where there was no restriction on future wetland locations in the landscape other than exclusion from urban areas, and two where future wetlands were restricted to specific natural and semi-natural land covers.

While the topographic, hydrological and sedimentological factors associated with the present-day distribution of mangrove and saltmarsh can provide a guide to where they might occur in future landscapes associated with SLR, numerous other factors not considered here will also have an influence (e.g. changes in catchment inflows, estuary entrance conditions and sediment supply; climate impacts on species present; existing tidal exclusion infrastructure; present and future management actions etc.).

Presentation form

Map digital

Edition

1

Dataset language

English

Dataset URI

https://datasets.seed.nsw.gov.au/dataset/266cdf74-2da9-48cb-8ba1-acdce66928b7

Purpose

This data set is one of the outputs from a state-wide, first-pass vulnerability assessment of mangrove and saltmarsh to SLR prepared under the Marine Estate Management Strategy Initiative 3. The purpose of the assessment was to identify estuaries where mangrove and saltmarsh have an apparently high or low vulnerability to SLR, so that further detailed research could be targeted to fill knowledge gaps. This data set can inform coastal planning at the estuary scale, but it does not replace the need for more detailed local scale assessment of management planning and actions for both existing and potential future wetland areas.

Status

Completed

Spatial representation type

grid

Spatial reference system

Spatial resolution

5 m

Additional information source

DATA STRUCTURE

The maps of projected mangrove and saltmarsh distribution in 110 estuaries are available in a GeoDatabase titled Wetland_Projections_SLR. The 110 estuaries investigated are listed in table 1.

The GeoDatabase contains four sets of maps, one for each sea level scenario: S0 - projections for present-day sea level; S1 - projections for +0.5m sea level rise; S2 - projections for +1.0m sea level rise; S3 - projections for +1.5m sea level rise.

Each set S0-S3 contains a spatial layer showing the projected distribution of mangrove and saltmarsh with no land use constraints other than exclusion from urban areas, and two projections constrained to specific land cover types. The naming convention is: SX; SX_nat_re; SX_nat_plus where SX designates the sea level scenario, i.e. present-day sea level (S0) or a projected future sea level (S1, S2, or S3).

SX is the projected habitat distribution for the specified sea level scenario with no land use constraints other than exclusion from urban areas.

SX_nat_re is the projected habitat distribution for the specified sea level scenario constrained to (i) natural land cover and (ii) low intensity land use based on NSW Land Use 2017 (https://datasets.seed.nsw.gov.au/dataset/nsw-landuse-2017-v1p2-f0ed).

SX_nat_plus is the projected habitat distribution for the specified sea level scenario constrained to (i) natural land cover, (ii) low intensity land use and (iii) agriculture and horticulture based on NSW Land Use 2017.

For example, the projected mangrove and saltmarsh distribution for the +0.5 m SLR scenario and no restrictions on future wetland location except exclusion from existing urban areas is S1. Similarly, the projected distribution under a +1.5 m SLR scenario with restrictions on location to natural land covers and low intensity land use is S3_nat_re.

In each layer mangrove is designated by the value 2 and saltmarsh by the value 3.

There are a small number of estuaries that include mangrove and/or saltmarsh in the present-day projections (S0), which have a NULL raster in the projections for higher sea level (S1-S3). This indicates that at the higher sea level scenario they are projected to have no mangrove or saltmarsh. This applies to the following estuaries at each sea level scenario: • S1: meringo_creek_S1 • S2: back_lagoon_S2, berrara_creek_S2, elliott_lake_S2, meringo_creek_S2, nadgee_river_S2, woolgoolga_lake_S2 • S3: back_lagoon_S3, berrara_creek_S3, congo_creek_S3, conjola_lake_S3, corunna_lake_S3, curalo_lagoon_S3, elliott_lake_S3, lake_brou_S3, lake_brunderee_S3, little_lake_wallaga_S3, meringo_creek_S3, nadgee_river_S3, station_creek_S3, woolgoolga_lake_S3

MAP ACCURACY

The map accuracy of the predicted present-day distribution of mangrove and saltmarsh can be assessed using the Matthews Correlation Coefficient (MCC), which is provided for each estuary in Table 1. The following qualitative descriptors apply: 0.0-0.19 zero to negligible correlation, 0.2-0.39 weak correlation, 0.4-0.59 moderate correlation, 0.6-0.79 strong correlation, 0.8-1.0 very strong correlation. An MCC value is provided for total wetland area, and mangrove and saltmarsh areas separately in each estuary.

High MCC values indicate high confidence in the predictive model for that estuary when applied to the present-day sea level and suggests a cautious level of confidence in the projections for future sea level scenarios, but there is no means to test the accuracy of future projections. It is recommended that results for estuaries where the MCC is between 0.4 and 0.6 are used with additional caution, and estuaries where MCC<0.4 the projections should not be used (see Hughes et al. 2022).

Originating controlled vocabulary

Dataset reference date

Responsible party

Metadata point of contact

Metadata date

2024-02-26T13:50:46.033150

Metadata language