Results of the GISIN needs assessment survey
Introduction
The Global Invasive Species
Information Network has convened an Invasive Species System Task Group as part
of the Taxonomic Database Working Group. The task at hand is the development of
a system for the exchange of invasive species information over the Internet. A
critical step in the process is to determine what is required by the eventual
users of this protocol, including organizations that will be providing data
(providers), organizations that will be consuming data (consumers), the end-users,
and stakeholders. Below are the results of a needs assessment survey to obtain
this information, which ran from 15 December 2006 through 15 February 2007.
Summary of Results
There were 137 respondents from
41 countries, the vast majority (80%) identifying themselves as providers of
invasive species data and consumers of data through web services. Most data
providers (77%) potentially offer spatial/temporal information,
profiles/species pages (65%), and checklist information (59%).
Data consumers seek all
three kinds of information (spatial/temporal, profiles, and checklist). For
spatial/temporal information, respondents seek date of observation (91%), name
or type of location (83%), and spatial coordinates (75%). For species accounts
or profiles, they seek a wide variety of data types, most popular being (in
descending order) life history/ecology, habitat descriptions, impacts,
geographic range, pathway information, images, management information, descriptions,
references, and requirements or tolerances. For checklists, respondents seek
(in descending order) occurrence data, habitat affected, introduction and
dispersal information, invasiveness, local impacts, date of introduction, and
origin.
To filter or search for
spatial/temporal information, species profiles, and checklist information,
virtually all respondents (93-96%) use scientific name.
There were some areas of the
questionnaire where the majority of the respondents were not knowledgeable:
half do not know the level of web services their organization provides and/or
uses; 80% do not know what existing protocols are appropriate for invasive
species information management; and 75% do not know what schemas or grammars
would be acceptable to copy or extend to implement the proposed system.
The overwhelming operating
system preference (96%) is MS-Windows, followed by Linux (24%). Preferred
Internet frameworks were ASP (57%), JSP, 51%, and PHP (39%). Preferred
programming languages were Java (58%) and PHP (49%). The majority offer
Microsoft Internet Information Services (69%) for hosting web services,
followed by Apache HTTP Server (46%).
Concerning the respondents’
needs/wants related to the toolkit for implementing the proposed system, 80%
would like a software toolkit to install and configure on their own server.
For a summary of the
comments that were submitted by respondents, see the Appendix (below).
Acknowledgements
Thank you very much to
those who responded to the survey. Special Thanks goes to
those who made this survey possible, especially Michael Browne, Jim Graham,
Hannu Saarenmaa, Elizabeth Sellers, Annie Simpson, and last but not least
Jeremy Kranowitz of The Keystone Center Inc., who administered it.
Basic Terminology
Invasive species information
can be broken into at least three information categories: Profile, Checklist,
and Spatial/temporal.
Profile information is a species page or fact sheet information that describes the invasive
species itself, and can include life history, images, management, expertise,
and other secondary information.
Checklist information contains location-specific observations (e.g. alien
or native, invasive or not), its date and mode of introduction, whether it is
considered a problem, if has been controlled, etc.
Spatial/temporal information describes the position of an invasive species on the
earth at a particular point in time and can be represented by a point, polygon,
or reference to a geographic location such as a country, park, lake, etc.
NOTE: we are using the term
"invasive species" in a
very broad sense, i.e., including any
non-native species that are potentially harmful to the environment,
livelihoods, or human health. A section with additional technical
definitions can be found at the end of the survey results.
Additional Technical Definitions that
were provided to participants in this survey:
Web server:
a software package that allows a computer to provide web pages and web services
over the Internet. Examples: Microsoft Internet Information Server (IIS),
Apache, and Tomcat.
Internet framework: an extension to, or a feature of, a web server that allows programming
languages to be used to add functionality to a web solution. Frameworks are
often identified by the file extension of the scripting languages they support,
for example PHP (PHP: Hypertext Preprocessor), Cold Fusion (CFM), Active Server
Pages (ASP), Java Server Pages (JSP), and Common Gateway Interface (CGI).
Another Internet framework example is Apache Axis.
Markup language: a specific syntax and grammar that allows the visual representation of
relatively complex, text-based, data. Examples: HTML (hypertext markup
language) and XML (extensible markup language).
Programming language: a defined syntax and grammar for communicating a
sequence of operations for a computer to complete. Programming languages
include scripting languages (e.g. PHP, JavaScript, Visual Basic for
Applications, Python) and compiled languages (e.g. C++, C#, Visual Basic). Java
is a special case and is partially compiled.
Protocol:
a defined sequence of messages structured with a specific syntax and grammar
for communicating a certain type of information between two computers.
Protocols are typically used in networking to pass packets of information between
computers. Examples: HTTP (hypertext transfer protocol), FTP (file transfer
protocol), SMTP (simple mail transfer protocol).
Web solution: a web-based, complete answer to a problem. Example: GISIN is proposing
a web solution for the exchange of invasive species data.
Compiled programming language: a programming language that is converted from a
human readable form to a machine readable form before it is executed (run).
Most operating systems and commercial applications are written in compiled
languages because they typically execute faster and the source code is not
available to end users. Examples: C++, C#, Visual Basic.
Scripting programming language: a programming language that is converted from human
readable form to machine readable form on a line by line basis. Scripts are
easier to create but take longer to execute than compiled programming
languages. Examples: PHP, JavaScript, Visual Basic for Applications, Python.
Basic Questionnaire Results (Additional comments submitted by
participants in this survey are summarized in the Appendix below.)
How will you potentially interact with invasive
species data? (Check all that apply) (n=136)
As a provider of invasive species data
to others through web services 81%
As a consumer of invasive species data
through web services 80%
As an end-user of invasive species data on the web 65%
As a stakeholder who may be providing support for the protocol 37%
If you are a data provider, what type of data can you
potentially offer? (check all that apply) (n=128)
Spatial/temporal information
(where/when the species is found) 77%
Profiles/species pages (life history,
images management, expertise) 65%
Checklist information (alien or native,
invasive or not, date of introduction, etc.)
59%
None (I am not an invasive species data provider) 14%
If you are a data consumer, what is the general type
of data you wish to have? (check all that apply) (n=132)
Spatial/temporal information (where/when the species
is found) 87%
Profiles/species pages (life history, images,
management, expertise) 85%
Checklist information (alien or native, invasive or
not, date of introduction, etc.) 84%
None (I am not an invasive
species data consumer) 5%
What is the level of web services your organization
provides and/or uses? (n=127)
I don't know
45%
Provide biological standards-based services to
others 23%
Use industry standards-based services to obtain
information (RSS feeds, WMS, etc.) 21%
Use biological standards-based services to obtain
information (DiGIR, BioCase, etc.) 18%
Provide industry
standards-based services to others 13%
None 12%
Which types of data are you interested in for
spatial/temporal information? (check all that apply) (n=131)
Date of observation
91%
Locations by type and name (i.e. Pacific Ocean, City
of
Spatial coordinates (i.e. lat/long or UTM) 75%
Polygons 54%
Which types of data are you interested in for
profiles/species pages? (check all that apply) (n=133)
Life history/ecology 97%
Habitat descriptions 95%
Impacts 95%
Geographical range 93%
Introduction & dispersal pathway
info 92%
Images
91%
Management information 90%
Descriptions 89%
References 89%
Requirements or tolerances 87%
Common names or synonyms 81%
Identification guides (e.g., keys)
77
Similar species (i.e. look alikes)
75%
Expertise 71%
Uses 66%
Which types of data are you interested
in for checklist information about invasive species? (check all that apply)
(n=134)
Occurrence (absent, eradicated, reported,
established, uncertain, etc.) 96%
Habitat affected 96%
Introduction and dispersal cause, route
and vector 94%
Invasiveness (invasive, not invasive,
not specified, uncertain) 93%
Local impacts 93%
Date of introduction/first observation,
etc. 91%
Origin (alien, native, not specified,
uncertain, cryptogenic) 90%
Introduction mode (intentional / unintentional, legal / illegal) 87%
Local prevention or management activities 87%
Abundance (spatial extent & density) 86%
Population trend 82%
Which of the following would you use to
filter spatial/temporal information? (check all that apply) (n=134)
By Genus & species 93%
By habitat 79%
Area type & name (e.g.
By pathway 61%
By impact 58%
Spatial bounding box (i.e. MinLatitude,
MaxLatitude, etc.). 55%
Date range when observed (MinDate,
MaxDate) 55%
Date range when introduced 54%
By introduction mode (intentional /
unintentional, legal / illegal) 54%
By common name 54%
By Genus 44%
By Family 43%
By Order 18%
By Kingdom 16%
By Class 13%
Which of the following would you use
for filtering profile/species page information? (check all that apply) (n=134)
By Genus & species 96%
By habitat type 76%
By introduction and dispersal pathway
type 71%
By common name 60%
By native range 53%
By requirements or tolerances 52%
By Genus 46%
By Family 46%
By similar species 41%
By management type (e.g. chemical control) 41%
By uses 29%
By expertise 24%
By Order 21%
By Kingdom 18%
By Class 17%
Which of the following would you use
for filtering checklist information about invasive species? (check all that
apply) (n=134)
By Genus & species 93%
Occurrence (eradicated, intercepted,
expanding, controlled, cultivated) 80%
Origin (alien, native, etc.) 74%
Invasiveness 72%
Introduction and dispersal cause, route
and vector 70%
Impact type 68%
Abundance (spatial extent and density)
63%
Introduction mode (intentional / unintentional, legal / illegal) 62%
Prevention or management activity
61%
Date range of introduction/first observation, etc. 60%
By common name 55%
Population trend 52%
By Genus 45%
By Family 42%
By Order 20%
By Kingdom 17%
By Class 16%
If we were to base the protocol on an
existing protocol, which of the following would be acceptable? (check all that
apply) (n=134)
I don't know 80%
Simple HTTP Request/Response (XML for
the response) 19%
DiGIR (Distributed Generic Information Retrieval) 8%
TAPIR (TDWG Access Protocol for Information Retrieval) 7%
RSS Feeds 6%
SOAP (Simple Object Access Protocol)
5%
BioCase 4%
OAI-PMH (Open Archives Initiative - Protocol for Metadata
Harvesting) 2%
SPARQL/RDF (SPARQL Protocol & RDF Query Language/Resource
Description Framework) 2%
Which of the following types of
schemas/grammars would be acceptable for us to copy or extend, to implement
this protocol? (check all that apply) (n=132)
I don't know 75%
Web Map Service (WMS--dynamic
maps) 14%
IAS-PS (Invasive Alien Species Profile
Schema) 14%
DarwinCore 11%
ABCD 5%
WFS (Web Feature Service--returns vector data) 5%
WCS (Web Coverage Service--returns raster data) 4%
Which of the following operating
systems are acceptable for a toolkit to support? (check all your organization
can support) (n=84)
MS-Windows 96%
Linux
24%
Unix 12%
Mac 10%
Which of the following Internet
frameworks do you feel are acceptable for a toolkit? (check all that apply)
(n=49)
Microsoft
Active Server Pages (ASP) 57%
Java
Server Pages (JSP) 51%
PHP
39%
Common Gateway Interface (CGI)
27%
Apache Axis 12%
Cold Fusion (CFM) 12%
Which of the following programming
languages do you feel are acceptable for a toolkit? (check all that apply)
(n=43)
Java
58%
PHP
49%
Jscript 35%
VBScript 35%
Python 23%
Perl 19%
C++ 16%
Cold Fusion (CFM) 12%
C# 9%
Which web servers do you have available
for hosting web services? (n=39)
Microsoft
Internet Information Services (IIS) 69%
Apache HTTP Server 46%
Apache Tomcat 28%
Sun Java System Web Server 10%
Zeus Web Server 0%
Which of the following toolkit options
would you find desirable? (check all that apply) (n=51)
Software toolkit to install and
configure on one of your servers 80%
Documentation for you to write your own
service (given that the protocol is simple)
61%
A turn-key solution with a separate computer that you connect to your
network and configure 25%
How many hours of technical support
could you allot to install this toolkit? (n=61)
One
31%
Whatever it takes 26%
Four 16%
Eight 13%
Other 7%
I don't know 7%
How many hours of technical support
could you allot for toolkit upgrades per year? (n=61)
Whatever it takes 26%
One
26%
Four
23%
Eight 11%
I don't know 8%
Other 5%
How many hours of development could you
allot to refine the system we are developing? (n=61)
Whatever it takes 28%
One
25%
Four 15%
Other 13%
Eight 11%
I don't know 8%
APPENDIX: Summary of
submitted comments
Of the 136 respondents, 75 provided at least one
detailed comment. A brief summary of some of the comments is provided here:
Please list how will you potentially interact with
invasive species data.
·
Provide free
access to all IAS databases important to rice and rice based farming systems
·
Working out
where a species is likely to become a problem, based on where it is a problem
now.
·
To try and help
us work out where to put resources in terms of which species we should be
prioritising for future or potential management.
·
Provider of aquatic
invasive species data via publishing them in on-line journal Aquatic Invasions http://www.aquaticinvasions.ru/
·
Teaching
·
Use for pest
risk analysis
·
Provider of
invasive species data through non-web applications.
·
Involved in
research into IAS (plants) and want to make my research more readily available.
·
Hands-on efforts
to control invasive species.
·
Provide a way
for field staff to collect, submit, and download invasive species data.
·
Sharing the data
and the source of data with others.
·
To educate the
general public and other staff members about invasive species threatening our
area.
·
Help disseminate
info via my organization.
·
Encouraging
others to collect invasive species data.
·
As a basis for
further research.
·
Providing
financial support to eradicate, control and prevent invasive species.
·
Site management
and inventory level.
·
Discussing
monitoring protocol and risk assessment development.
·
Exploring how we
can share data on invasive plant species.
·
Broker of information
for other end users.
·
As a
collaborator.
What is the general type of data you wish to have?
·
Metadata
information may be important as well as primary contact information (data
manager)
·
References to
the published literature on each species
·
Many profiles/species
pages exist already. Spatial data is hardest to get and the most difficult to
upkeep.
·
Quarantine
actions/regulations in other countries for a pest species or group
·
Particulars on
trapping (sentinal) types, placement, lures, effective detection
·
Links to any
existing risk assessments that have been done (either on related pathways or
for intentional introductions)
·
Information on
other programs tackling invasive species to allow for an exchange in techniques
and experiences.
·
Control/management
history of invasive species sites, areas surveyed (or not surveyed)
·
GIS datasets
·
information how
to manage/control/eradicate IAS
·
Information on
damage, preferably quantitative.
·
Plus policy (use
of bounties, native animal pests, role of commercial harvest) and social
information - e.g. Codes of Practice for management (Animal Welfare
considerations)
·
impact of
species if known
·
Potential world
and current distribution
·
Habitat profiles
for prediction modeling
·
How likely is it
that a species currently in low densities / not widely distributed, may become
a significant problem
·
For species that
are assessed as being potential threats, we would be interested in main invasion
pathways in other areas
What is the level of web services your organization
provides and/or uses?
·
Currently in the
development phase of the project and have not moved to web distribution.
·
Provide
information on threats, biology, pathways and control for target species.
·
Obtains and
provides policy-relevant information to promote screening of live animal
imports
·
We are in the
process of creating web services.
·
Internet
Information Server, Oracle webservices, Microsoft Access
·
We would
probably not maintain our own service but collaborate with service providers
·
Use cold fusion
for web site. Contributors can add own data and references monitored through a
web manager.
Which types of data are you interested in for
spatial/temporal information?
·
Country
presence/absence only with dates
·
Availability of
herbarium records or information about source data (eg. is the record verified,
and if so how)
·
Best is a grid
map with a database behind
·
Much of the
historical information is not georeferenced, so names of places (toponyms)
would be useful
·
Historical
information from herbarium specimens
·
Location by
type, for habitat and climate requirements analysis
·
Species name,
extent of infestation
·
There is an
URGENT need to know about IAS in all kinds of Protected Areas
·
Species native
range and origin
·
Extent of threat
·
Google kml files
for easy graphical displays
Which types of data are you interested in for
profiles/species pages?
·
Differentiation
of area of origin, area of secondary distribution; estimates of cost-benefit.
·
Gene sequences
for cryptic IAS.
·
Cost of
eradication/control.
·
Altitudinal
range.
·
Economic losses
due to the IS, costs of prevention/control, and socio-economic value of the IS
as a useful species. Also invasiveness, impacts, reproduction, dispersal,
establishment and prevention/control potential.
·
Current research
being done.
·
Host data (for
pathogens).
·
Associated
legislation.
·
Some indication
of who regulates it or who considers it invasive.
·
Economic cost of
prevention and control.
·
Legislation (by
geopolitical area)
·
Quarantine,
detection, and prevention.
·
Public education
efforts (and effectiveness thereof).
·
Pathway
information should include volume of trade and export/import countries.
·
Case studies of
successful and failed management (that can be used to learn from).
·
Approved
biological control agents available for release
·
Cost of control
per hectare (or other unit) in areas where it is being managed.
·
Any costs
associated with, for example, loss of production or loss of native species or
human health impacts for any area affected by it where this assessment has been
carried out by a local agency. (We use Weighted Average Gross Margin as a
proxy, but every country is likely to have different ways of assessing things,
so maybe this field could be most useful as a descriptive field using examples
from anywhere an assessment like this has been undertaken.)
·
Potential
spread: This also might only be able to be done descriptively.
·
RSS feeds etc.
with links to current news, research papers, genbank sequences, etc. for a
particular taxon
·
Web-based links
Which types of data are you interested in for
checklist information about invasive species?
·
Distribution
rate
·
Potential animal
and human health impacts
·
Specific details
of control attempts and results of such attempts. Also by known distribution.
·
Common names
·
Ensure that
standardised definitions are used so that the information is consistent and transferable.
Also, give a confidence rating on the quality/accuracy of the data
·
Regulatory
status
·
Whether they
were "observed" to be invasive, or "judged" to be invasive
(e.g. after Pest Risk Assessment).
·
Whether there is
a legal or other relatively official status given to them such as
"quarantine pest", "unwanted", "environmental weed"
etc.
·
Biological
control agents evident.
·
For plants,
growth form is an important character that is not addressed in taxonomy
·
Associated taxa
(pollinators, parasites, pathogens, predators, etc.)
What would you use to filter spatial/temporal
information?
·
By known
distribution, by invasiveness, by impacts, by reproductive, dispersal, establishment
and prevention/control potential.
·
Eco-functional
group (e.g. phytoplankton, macrozoobenthos, fish, etc.)
·
By impact; both
the mechanics of impact (e.g. predation, "strangling" by vines, loss
of production, outcompetes native plants, human health impacts, clogs waterways
etc.) and types of native biodiversity impacted. Also, by impact type--whether
biodiversity is impacted (i.e. are they a CONSERVATION concern as opposed to a
health or economic concern)
What would you use for filtering checklist
information about invasive species?
·
Control and
prevention potential
What existing protocol would you recommend?
·
Really doesn't
matter as long as it's reliable (both the information and the interface).
·
Global Invasive
Species Database.
·
The easier the
better.
·
I have no notion
of informatics but could ask our informatician to fill this questionnaire.
Which schemas/ grammars would be acceptable for us to
copy or extend, to implement this protocol?
·
To extend
Plinian Core for the species page information.
·
Interactive
online maps are not very useful as a reference, downloadable datasets are
needed. Citation instructions should be given on the website as well, with
complete metadata.
·
One customised
for invasive species woud probably be the best option.
·
Any schema is
acceptable provided it is logical, user-friendly/intuitive design, and unclunky
·
See above --
just add a header line to each table and provide metadata linking any
relational tables. Which DarwinCore!!!
·
As many as
possible, provided it is clear how terminology / concepts are used, so
"apples" and "apples" are put together as much as possible.
Other than Windows, Linux, Unix, and Mac, which
operating systems are acceptable for a toolkit to support?
·
MS Access, MS
Excel, shapefile imports.
·
Write it in Java
so the O/S is more or less irrelevant.
·
Not sure. Our
site is ColdFusion and mySQL database.
·
None. You should support browsers that work on all
the above.
·
Should consider
Windows Mobile for portable devices.
Which Internet frameworks do you feel are acceptable
for a toolkit?
·
Requirements
should be met by any scripting language.
·
asp.net
·
Don't know, but
make it user-friendly.
·
It doesn't
matter; it's up to the provider.
·
Use HTTP
requests. The user should be unaware of
what you're using on the back-end.
·
As long as users
everywhere will be able to use it with their basic internet systems, it doesn’t
matter.
Which programming languages do you feel are
acceptable for a toolkit?
·
Javascript
·