Phenology data synthesis

Author

Ella Cathcart-van Weeren

NOTES TO FUTURE ELLA in the data cleaning phase before this analysis is done:

Here we are summarising and combining data from GBIF and then combining it with iNat data to increase sample size. Here is the data that was downloaded from GBIF, with additional observations from iNaturaliast (for the same species) to increase samples sizes of observations -This is an example workflow, summaries and figures for a subset of species (41).

Global Biodiversity Information Facility (GBIF)

Lets look at GBIF data by itself first. GBIF uses a category called Basis of Record to assign an observation type to each record. It describes the specific nature of the data record. The full definition of each of these classifications are defined in the Darwin Core Terms of reference and can be found here.

basisOfRecord Description
PreservedSpecimen A physical object representing one or more organisms, part of an organism, or artifact of an organism. Synonyms: voucher, collection, lot.
FossilSpecimen A physical object representing one or more fossil organisms, part of a fossil organism, or artifact of a fossil organism.
LivingSpecimen An organism removed from its natural occurrence and now living in captivity or cultivation.
HumanObservation A report by a known observer that an organism was present at the place and time.
MachineObservation A report by a monitoring device that an organism was present at the place and time.
StillImage A photograph, drawing, or painting.
MovingImage A sequence of still images taken at regular intervals and intended to be played back as a moving image; may include sound.
SoundRecording An audio recording.
OtherSpecimen Any type of specimen not covered by any of the categories above.

Reproductive phenology observations from different sources

Here is a summary of the total observations, and the number of different flowering, budding and fruiting observations from the different classifications of data. The ‘observation’ records are coming from climatewatch.

basisOfRecord total_n flower_obs bud_obs fruit_obs
HUMAN_OBSERVATION 40828 199 44 432
LIVING_SPECIMEN 53 0 0 0
MATERIAL_SAMPLE 3 0 0 0
OBSERVATION 2058 80 0 0
OCCURRENCE 2399 0 0 0
PRESERVED_SPECIMEN 10257 1970 1452 2493

Number of publishers contributing to each sources

Lastly here is the list of data contributes (called publishers on GBIF) based on the type of data they are contributing. Everything that

basisOfRecord publishers
HUMAN_OBSERVATION Notomys Seeds, NSW Department of planning, industry and environment, Pl@ntNet, Victorian Department of Environment, Land, Water and Planning, iNaturalist.org, naturgucker.de, NatureMapr, Observation.org, Questagame, Queensland Department of Environment and Science, Atlas of Living Australia, The University of Queensland, DEPARTMENT OF PLANNING AND ENVIRONMENT - Environment and Heritage, Local Government NSW, Griffith University - Griffith Centre for Coastal Management
LIVING_SPECIMEN Australian National Botanic Gardens, The Royal Botanic Gardens & Domain Trust
MATERIAL_SAMPLE Queensland Department of Environment and Science
OBSERVATION ClimateWatch, Atlas of Living Australia
OCCURRENCE Queensland Department of Environment and Science, Macquarie University
PRESERVED_SPECIMEN NSW Department of planning, industry and environment, United Herbaria of the University and ETH Zurich (Z+ZT), The Royal Botanic Gardens & Domain Trust, Centre for Australian National Biodiversity Research, Naturalis Biodiversity Center, Department of Environment and Natural Resources, Tasmanian Museum and Art Gallery, Royal Botanic Gardens Victoria, Western Australia, Department of Biodiversity, Conservation and Attractions, Queensland Department of Environment and Science, Missouri Botanical Garden, National Museum of Natural History, Smithsonian Institution, Auckland War Memorial Museum, University of Kansas Biodiversity Institute, Rancho Santa Ana Botanic Garden, University of Minnesota Bell Museum, Lund University Biological Museum, Meise Botanic Garden, MNHN - Museum national d’Histoire naturelle, Florida Museum of Natural History, Royal Botanic Garden Edinburgh, Museum of Evolution, Uppsala University, Royal Botanic Gardens, Kew, Museum of New Zealand Te Papa Tongarewa, South African National Biodiversity Institute, European Nucleotide Archive (EMBL-EBI), Landcare Research, The University of Vermont Pringle Herbarium, Utah State University, Australian Tropical Herbarium, Instituto de Botânica, São Paulo, University of California, Davis, Arizona State University Biocollections, University of Connecticut, Conservatoire et Jardin botaniques de la Ville de Genève - G, University of Alberta Museums, University of New England, Consortium of Pacific Northwest Herbaria, Colorado State University Herbarium (CSU), Auburn University Museum, Dipartimento di Biologia, Università di Pisa, The University of Melbourne, James Cook University, Australia’s Virtual Herbarium, Atlas of Living Australia, University of Wollongong, La Trobe University

How Citizen/community science observations are included

Before doing any further summaries, lets take a more in depth look at the contributors for the human observations. Interestingly we see that there are only two sources actually contributing and reproductive phenology data here: iNaturalist and the NSW Government.

publisher total_n flower_obs bud_obs fruit_obs
Atlas of Living Australia 362 0 0 0
DEPARTMENT OF PLANNING AND ENVIRONMENT - Environment and Heritage 196 0 0 0
Griffith University - Griffith Centre for Coastal Management 149 0 0 0
Local Government NSW 97 0 0 0
NSW Department of planning, industry and environment 33654 28 7 30
NatureMapr 113 0 0 0
Notomys Seeds 1 0 0 0
Observation.org 2 0 0 0
Pl@ntNet 94 0 0 0
Queensland Department of Environment and Science 225 0 0 0
Questagame 42 0 0 0
The University of Queensland 19 0 0 0
Victorian Department of Environment, Land, Water and Planning 160 0 0 0
iNaturalist.org 5704 171 37 402
naturgucker.de 10 0 0 0

Not all data that is stored in some databases feed into big aggregates like GBIF. Specifically if we look at the Human Observation classification, we see that only ‘verified iNaturalist records are included. Whilst this is not necessarily a bad thing, there are many more observations on iNaturalist that are still classified as ’Need ID’ but are good enough to use for phenology observations.

Can additional iNaturalist data increase sample size?

Lets compare the data that we got from iNaturalist with the data that we get in the GBIF download to see if they are the same or how they differ. Interestingly there a total of 382 duplicated observations. The vast majority of duplicated observations come from GBIF data 381.

Does how including iNatualist data directly from iNat increases sample sizes?

Yes we see that whilst most observations come from GBIF (specifically preserved specimens) we get nearly an additional 500 observations when we include observations directly from iNaturalist. Interestingly most of these are research grade which seems to indicate that GBIF is not pulling all observations from iNaturalist.