EcoAssets Building data linkages for
environmental reporting Shandiya Balasubramaniam (she/her)
@ShandiyaB
Data Analyst | Atlas of Living Australia
ESA-SCBO 2022

Image: Will Turner

I acknowledge the Traditional Owners of the lands on which I live and work, the Wurundjeri People of the Kulin Nation, and pay my respects to Elders past and present. I recognise the spiritual and cultural significance of land, water, and all that is in the environment to Traditional Owners, and their continuing connection to Country.

Image: Fiona Smallwood

EcoAssets

EcoAssets is a project that brings together 3 research infrastructures, the Atlas of Living Australia, the Terrestrial Ecosystem Research Network, and the Integrated Marine Observing System, to provide data in forms that support environmental reporting, such as the State of the Environment Report.

These 3 research infrastructures collect, manage, and organise environmental data across different categories:

Image: Germane Jaws

EcoAssets Atlas of Living Australia: national biodiversity database

The ALA is Australia’s national biodiversity database, bringing together data from natural history collections, government departments, researchers, community groups, and individuals

Image: Germane Jaws

EcoAssets Terrestrial Ecosystem Research Network: terrestrial ecosystem attributes

TERN measures key terrestrial ecosystem attributes at hundreds of representative locations and produces ecosystem datasets at a national scale

Image: Germane Jaws

EcoAssets Integrated Marine Observing System: monitoring coastal and open waters

and IMOS collects data on all aspects of the marine environment, with extensive monitoring activities in coastal and open waters

Image: Germane Jaws

EcoAssets Atlas of Living Australia • Terrestrial Ecosystem Research Network • Integrated Marine Observing System

Historically, data from these activities have certainly fed into environmental reporting. Notably, the 2016 State of the Environment report referenced particular datasets from these 3 infrastructures, but it was difficult to access the data directly, and there was no opportunity down the track for researchers to evaluate and review the conclusions that were reached using these data

Image: Germane Jaws

EcoAssets Building data linkages for environmental reporting

EcoAssets addresses this by delivering standardised, comparable data from these infrastructures, in a form that can be versioned to provide ongoing indicators over time.

Image: Germane Jaws

Focus Areas

The project has 3 broad areas of focus…

Image: Lottie Corin

Focus Areas Improve data pipelines

The first is to improve biodiversity data pipelines among the ALA, TERN, and IMOS. In the past, the process of updating and integrating datasets from TERN and IMOS into the Atlas has not always happened consistently. Improving these processes will ensure that our national views, through the Atlas, of species occurrence, distribution, and abundance are being regularly and consistently updated

Image: Lottie Corin

Focus Areas Improve data pipelines • Deliver aggregated datasets

The second is to deliver aggregated reference datasets in forms that are suitable for environmental assessment

Image: Lottie Corin

Focus Areas Improve data pipelines • Deliver aggregated datasets • Support ongoing versioning

And the third is to set up a workflow that supports ongoing delivery of these datasets, in such a way that they can be versioned over time, so this isn’t just a one-off event

Today, I’ll be focussing on this second area:

I’ll give an overview of how we started with over 100 million records in the Atlas, and ended up with condensed, summary datasets that could be used for reporting

Image: Lottie Corin

Imagine, if you will, that you’re looking at 110 million records in the Atlas. Some of them will be missing sufficient detail to make them useful, so we start by excluding those using the ALA General Data Quality Profile.

Description	Filter
Exclude all records where spatial validity is "false"	-spatiallyValid:"false"
Exclude all records with an assertion that the scientific name provided does not match any of the names lists used by the ALA. For a full explanation of the ALA name matching process see https://github.com/AtlasOfLivingAustralia/ala-name-matching	-assertions:TAXON_MATCH_NONE
Exclude all records with an assertion that the scientific name provided is not structured as a valid scientific name. Also catches rank values or values such as "UNKNOWN"	-assertions:INVALID_SCIENTIFIC_NAME
Exclude all records with an assertion that the name and classification supplied can't be used to choose between 2 homonyms	-assertions:TAXON_HOMONYM
Exclude all records with an assertion that kingdom provided doesn't match a known kingdom e.g. Animalia, Plantae	-assertions:UNKNOWN_KINGDOM
Exclude all records with an assertion that the scientific name provided in the record does not match the expected taxonomic scope of the resource e.g. Mammal records attributed to bird watch group	-assertions:TAXON_SCOPE_MISMATCH
Exclude all records with an assertion of the occurence is cultivated or escaped from captivity	-establishmentMeans:"MANAGED"
Exclude all records with an assertion of latitude value provided is zero	-decimalLatitude:0
Exclude all records with an assertion of longitude value provided is zero	-decimalLongitude:0
Exclude all records with an assertion of latitude and longitude have been transposed	-assertions:"PRESUMED_SWAPPED_COORDINATE"
Exclude all records with an assertion of coordinates are the exact centre of the state or territory	-assertions:"COORDINATES_CENTRE_OF_STATEPROVINCE"
Exclude all records with an assertion of coordinates are the exact centre of the country	-assertions:"COORDINATES_CENTRE_OF_COUNTRY"
Exclude all records with "Latitude is negated" warning	-assertions:"PRESUMED_NEGATED_LATITUDE"
Exclude all records with "Longitude is negated" warning	-assertions:"PRESUMED_NEGATED_LONGITUDE"
Exclude duplicates from different datasets that are not the representative record	-(duplicate_status:"ASSOCIATED" AND duplicateType:"DIFFERENT_DATASET")
Exclude all records where coordinate uncertainty (in metres) is greater than 10km	-coordinateUncertaintyInMeters:[10001 TO *]
Exclude all records with unresolved user assertions	-userAssertions:50001
Exclude all records with unconfirmed user assertions	-userAssertions:50005
Exclude all records where outlier layer count is 3 or more	-outlierLayerCount:[3 TO *]
Exclude all records where Record type is "Fossil specimen"	-basisOfRecord:"FOSSIL_SPECIMEN"
Exclude all records where Record type is "EnvironmentalDNA"	-(basisOfRecord:"MATERIAL_SAMPLE" AND contentTypes:"EnvironmentalDNA")
Exclude all records where Presence/Absence is "absent"	-occurrenceStatus:ABSENT
Exclude all records where year is prior to 1700	-year:[* TO 1700]

This table shows the 23 checks that the profile applies. We also excluded records prior to 1900, and those that were missing spatial coordinates or species IDs

This left us with occurrence records that were associated…

… with a known location …

… and a known species that could be placed within the broader taxonomic hierarchy

From here, we were able to extend this to include other fields of interest. For example, based on spatial coordinates, we could determine where a record sat within the protected area network, or which IMCRA or IBRA bioregion it was found in

Similarly, with an accurate species identification, we could classify records based on how they matched to threatened species or invasive species lists, which are really important for conservation and biosecurity projects.

From this point, we were able to facet and aggregate these data to produce views that start to represent environmental indicators

ibraRegion	yearStart	yearEnd	griisStatus	recordCount	speciesCount
Arnhem Coast	1900	1970	Introduced	206	19
Arnhem Coast	1900	1970	Invasive	11	3
Arnhem Coast	1900	1970	Native	7754	1791
Arnhem Coast	1971	1975	Introduced	49	19
Arnhem Coast	1971	1975	Invasive	1	1
Arnhem Coast	1971	1975	Native	5570	1658
Arnhem Coast	1976	1980	Introduced	11	8
Arnhem Coast	1976	1980	Invasive	2	2
Arnhem Coast	1976	1980	Native	5945	1163
Arnhem Coast	1981	1985	Introduced	15	10
Arnhem Coast	1981	1985	Invasive	1	1
Arnhem Coast	1981	1985	Native	3032	1116
Arnhem Coast	1986	1990	Introduced	72	26
Arnhem Coast	1986	1990	Invasive	12	3
Arnhem Coast	1986	1990	Native	14929	1736
Arnhem Coast	1991	1995	Introduced	199	41
Arnhem Coast	1991	1995	Invasive	20	6
Arnhem Coast	1991	1995	Native	12067	1572
Arnhem Coast	1996	2000	Introduced	602	59
Arnhem Coast	1996	2000	Invasive	152	6

For example, this is a summary of species occurrence from 1900 to 2020, for Australian terrestrial species organised by IBRA region and status as introduced, invasive, or native based on the Global Register of Introduced and Invasive species list for Australia. Counts are for the total number of records, and the number of distinct species in each place, during each period of time.

ibraRegion	yearStart	yearEnd	griisStatus	recordCount	speciesCount
Arnhem Coast	1900	1970	Introduced	206	19
Arnhem Coast	1900	1970	Invasive	11	3
Arnhem Coast	1900	1970	Native	7754	1791
Arnhem Coast	1971	1975	Introduced	49	19
Arnhem Coast	1971	1975	Invasive	1	1
Arnhem Coast	1971	1975	Native	5570	1658
Arnhem Coast	1976	1980	Introduced	11	8
Arnhem Coast	1976	1980	Invasive	2	2
Arnhem Coast	1976	1980	Native	5945	1163
Arnhem Coast	1981	1985	Introduced	15	10
Arnhem Coast	1981	1985	Invasive	1	1
Arnhem Coast	1981	1985	Native	3032	1116
Arnhem Coast	1986	1990	Introduced	72	26
Arnhem Coast	1986	1990	Invasive	12	3

One way of visualising this is with a heatmap, like the one on the right. Each row is an IBRA region, and each column is a period of time, these are 5 year time periods. Each cell is coloured according to the proportion of introduced or invasive species for that time and place, with darker colours representing higher proportions.

imcraRegion	yearStart	yearEnd	epbcStatus	recordCount	speciesCount
Abrolhos Islands	1900	1970	Not listed	1054	401
Abrolhos Islands	1971	1975	Not listed	291	149
Abrolhos Islands	1976	1980	Not listed	407	200
Abrolhos Islands	1981	1985	Endangered	1	1
Abrolhos Islands	1981	1985	Not listed	186	128
Abrolhos Islands	1981	1985	Vulnerable	1	1
Abrolhos Islands	1986	1990	Not listed	657	266
Abrolhos Islands	1991	1995	Not listed	1592	363
Abrolhos Islands	1996	2000	Conservation Dependent	1	1
Abrolhos Islands	1996	2000	Not listed	933	162
Abrolhos Islands	2001	2005	Conservation Dependent	2	1
Abrolhos Islands	2001	2005	Not listed	408	192
Abrolhos Islands	2006	2010	Not listed	2458	229
Abrolhos Islands	2011	2015	Endangered	1	1
Abrolhos Islands	2011	2015	Not listed	56	30
Abrolhos Islands	2016	2020	Endangered	10	1
Abrolhos Islands	2016	2020	Not listed	605	71
Anson Beagle	1900	1970	Endangered	1	1
Anson Beagle	1900	1970	Not listed	4110	1733
Anson Beagle	1900	1970	Vulnerable	5	3

We can do a similar thing with threatened species. This table shows the count of records and the number of distinct species for each IMCRA region, again, in those 5 year time periods, for each EPBC listing category.

imcraRegion	yearStart	yearEnd	epbcStatus	recordCount	speciesCount
Abrolhos Islands	1900	1970	Not listed	1054	401
Abrolhos Islands	1971	1975	Not listed	291	149
Abrolhos Islands	1976	1980	Not listed	407	200
Abrolhos Islands	1981	1985	Endangered	1	1
Abrolhos Islands	1981	1985	Not listed	186	128
Abrolhos Islands	1981	1985	Vulnerable	1	1
Abrolhos Islands	1986	1990	Not listed	657	266
Abrolhos Islands	1991	1995	Not listed	1592	363
Abrolhos Islands	1996	2000	Conservation Dependent	1	1
Abrolhos Islands	1996	2000	Not listed	933	162
Abrolhos Islands	2001	2005	Conservation Dependent	2	1
Abrolhos Islands	2001	2005	Not listed	408	192
Abrolhos Islands	2006	2010	Not listed	2458	229
Abrolhos Islands	2011	2015	Endangered	1	1

Putting that in a heatmap again looks like this, with IMCRA regions in rows and time periods in columns. The colours show the proportion of EPBC-listed species for each time and place, with the darker colours showing higher proportions

These data also let us interrogate how we collect evidence of an occurrence. The ALA includes information on the type of record, and for much of the 20th century, our evidence for an occurrence came from museum specimens. More recently, this has changed in favour of human observations, which includes data collected by ecologists, field monitoring programs, and increasingly citizen science projects

This image shows the proportion of records that are made up by human observations between 1900 and 1970, with IBRA regions shown in shades or yellow and orange, and IMCRA regions in shades of blue and purple. The darker shades represent higher proportions of human observations

And we can look at how this changes over time…

This could have implications for users, in terms of deciding what sorts of data to include (or not). For instance, more charismatic species might be over-represented in citizen science projects, giving the impression that there’s been a population boom. Or museums may have a bias towards collecting certain taxa, depending on the expertise among their staff

ecoassets.org.au

All these datasets are freely available right now, and I hope seeing some of these examples has given you all ideas for potential use cases. They’re available on the EcoAssets website, which can be accessed at ecoassets.org.au

Clicking on the DATA tab will navigate to a list of all the available assets, and clicking on any of these links will bring up a page with the option to download the data, and view the metadata. This includes a brief description of the dataset, and explanation of how we derived the aggregated counts, and information on how to cite these data if you do use them

Upcoming • Event-based datasets • State of Forests Report 2023

The datasets currently available on the website are based on occurrence records in the Atlas, and the next new and exciting thing we’re working on is generating event-based datasets for EcoAssets. IMOS and TERN both collect significant amounts of data at the level of site-based surveys or through species tracking activities, and ALA likewise has extensive datasets from providers like BirdLife, who carry out systematic surveys at regular intervals. We anticipate that these event-based datasets will complement the biodiversity datasets already available, and provide some measure of the extent of monitoring happening across the country.

We’re also working with a team at ABARES to support their data requirements for the 2023 State of the Forests Report, which involves facetting the data such that we can assess levels of biodiversity within the national forest estate

Image: David Clode

EcoAssets received investment (doi.org/10.47486/XN005) from the Australian Research Data Commons (ARDC). The ARDC is funded by the Australian Research Data National Collaborative Research Infrastructure Strategy (NCRIS).

Image: Joel Vodell