Ecostations Data Access Monitor
This proposal originated at a meeting with Neil Davies (Moorea IDEA, BIDS), Jorrit Poelen (GloBI, author of this text) and Charlotte Cabasse (BIDS) at BIDS on Aug 5, 2015.
(Island) Ecostations Data Access Monitor (EDAM) - Phase 1 - species lists, food webs and associated citations.
Develop methods to openly compare ecological data richness /completeness of (Island) Ecostations to promote open data and tool re-use between researchers, citizen scientists, educators, foundations and regulatory bodies.
Student project for BIDS Hacking Measurement Class Fall 2015: http://www.ischool.berkeley.edu/courses/i290-hm.
As large biodiversity collections and environmental data are accessible online, global research communities have an unprecedented access to (siloed) datasets. Now that methods are within reach that allow to combine and process biodiversity data at global scales, institutions can start to re-examine existing data to coordinate data collection efforts, evolve data sharing strategies and discover methods to efficiently sustain our (island) ecosystems. A first step toward integrating the data is to provide a side-by-side comparison of existing data associated with active island ecostation communities to stimulate knowledge sharing and collaboration. See Meyer et al. 2015 for a recent discussion about the importance of identifying "[...] Gaps in digital accessible information (DAI)[...]" or biodiversity data completeness.
Ecostation biodiversity data summaries are derived from openly available biodiversity data repositories (e.g. GBIF, iDigBio, GloBI). Initially only species lists and associated food webs are compiled for participating ecostations using automated data processing algorithms. For each ecostation, the completeness of the lists and webs are estimated. Also, the similarity of the lists and webs are calculated across the spatially separated island ecosystems to highlight ecological likeness. By providing EDAM, spatially and institutionally disjointed projects now have a data- driven method to see how much ecological data is available for specific spatio- taxonomic spaces. We hope that comparing available ecological data across ecostations will help stimulate collaboration between scientists, technologists, educators, local governments and research foundations to help better understand and sustain ecosystems around us.
In order to archieve EDAM Phase I, we need to:
- (scope) identify 3-5 island ecostations (e.g. Moorea/Oahu/Friday Harbor/Galápagos)
- (scope) define/identify spatial and taxonomic ranges for ecostations
- (data) generate/retrieve spatio-taxonomic species checklist at scale and on- demand (e.g.https://github.com/jhpoelen/effechecka, Map of Life mol.org)
- (data) construct local biotic interaction webs based on species interaction data and spatially explicit checklist: occurrence + interaction = local biotic interaction web estimate.
- (model, analysis) develop/adopt similarity measures for checklists and biotic interaction webs
- (model, analysis) develop/adopt completeness measures for checklists and biotic interaction webs
- (share) create a web-accessible visualization to make results (and associated data sources) easy to access.
Given the availability of relevant ecolological data and the limited scope of needed data processing capabilities, a first prototype of EDAM Phase I is estimated to take 2-3 months for a group of digitally literate graduate students with an allocation of ~10 hrs a week. The outcome of this project will help define future funding to further develop advanced phases of the EDAM project.
Jorrit Poelen, http://globalbioticinteractions.org, https://github.com/jhpoelen or jhpoelen at xs4all dot nl