The National Institute of Invasive Species Science A Global Organism Detection & Monitoring System

Greg Newman - Research Associate/Graduate Student Natural Resource Ecology Laboratory Colorado State University National Institute of Invasive Species Science

Integrating Invasive Plant Species Data in the Midwest: Solutions for Data Collection and Management

North Central Integrated Pest Management Center Invasive Plant Association of Wisconsin Madison, WI

January 25th, 2008 Outline • Why GODM? • What is GODM? • Main Features • Core Database Design • Potential Applications • Data Exchange Challenges • What is the future of GODM?

January 25th, 2008 Why GODM? • Integrated disparate data • Multi-scale perspective • Centralized system of common data subsets • Occurrence data and associated attributes • One of many “middle-men” • Data exchanges with many other databases • Awareness of what might be invading next

January 25th, 2008 What is GODM? • A research endeavor • A system to store invasive species data • Centralized storage of single contributions, periodic large data ingests, & regularly scheduled data harvesting events • Common subset of attributes (site, visit, organism) • Controlled vocabulary • An online mapping application • A decision support system

January 25th, 2008 System Components ---

Browser Database

C++ Web Server PHP Pages Libraries

Raster Layers

January 25th, 2008 System Architecture

January 25th, 2008 Core Database Design

Organizations

Organism To Area Projects Taxon Units

Area Visit Organism Data (name,code) (name,code) (TSN)

Attributes Spatial Data (x,y) Auxillary Data

Treatments

January 25th, 2008 Main Features

• Manage accounts (My profile, create projects, manage project members)

• Gather data (EcoNab PALM application, filed protocols & standards)

• Browse data (By location, species, organization, project, map)

• Contribute data (Pre-defined, file configurations, custom codes)

• Analyze data (Extract raster pixel values from point locations)

• Download data (Downloadable files as CSV, TXT, SHP, filters)

January 25th, 2008 January 25th, 2008 Existing Situation • GIS software • GRASS, AutoCAD, ArcINFO, ArcView, ArcMap, ArcSDE • Internet browsers • , , , , Blazer • Databases & online map applications • DBASE, Access, MSSQL Server, MySQL, Postgres, Oracle • Numerous ArcIMS sites • The National Map, NBII sites • Regionally focused systems (IPANE) • Web-service enabled sites • GBIF, DiGir, VegBank, USGS NAS system • Modeling capabilities • C-Squares, KGS Mapper, Google Earth, Life Mapper

January 25th, 2008 Traditional workflows

Measure plants Enter data Create analysis Analyze data in the field and manually into spreadsheet in record data spreadsheet or required format manually database

• Limited to data in this data set only • Pstored locally on • Data quality issues • Data transformation single PC in files • Tedious • Projection issues • Results not publically available • Expensive • Requires many nor reproduceable different software packages to format • Requires knowledge data for analyses of programming

Publish data • Data / conclusions based on isolated data set

January 25th, 2008 • Limited scales Current paradigm

• GIS software • By-and-large desktop use • Limited to isolated, organization-based datasets • Internet browsers • Typically used to search for quick answers, pubs, data • Limited use as online GIS mapping application • Databases & online map applications • Static, not real-time • Requires server maintenance and updates • Limited features / application functionality • Slow performance

January 25th, 2008 Multi-scale usage

Population Plot data Subplot data Auxilary data Community Collection of plots in aspen Landscape All plots as point locations for species

January 25th, 2008 Example Map – Tamarix locations

January 25th, 2008 Example Map – Points, lines, polygons

January 25th, 2008 Future Directions

GIS use By-and-large internet applications & web services Uses data from multiple sources Internet use Advanced web applications Online editing capabilties, dialogs, etc. Online mapping applications Real-time data updating while you watch (AJAX) Web services enable better user customization True web based ‘applications’ High performance, fast

January 25th, 2008 Paradigm shift re: online GIS and workflows Newly developed tools must be…

• Built to be used by ecologists, not programmers

• Created with end-users in mind

• Developed to be reusable modules that are compatible with other tools

• Capable of providing output(s) in many formats quickly and easily

• ‘Analysis-ready’ and capable of repeated results

• Easy to use and simple

January 25th, 2008 Future workflows

Measure plants Beam data Data now Analyze your in the field and directly to immediately data sythesized record data centralized data accessed online with others or directly into warehouse in real-time. by itself handheld device (topical-based with GPS, like GODM for Data may be Create camera, and invasives) via tagged as predictive web access cell phone sensitive for models that can internet call to privacy issues if in turn be used web service needed. as predictive surfaces for Users may other species create analysis subset filtered Contribute as needed algorithms for use by others

Publish data • Data / conclusions based on merged data sets or isolated data sets • Multiple spatial / temporal scales

January 25th, 2008 • Repeatable analyses & results can be compared with results from new techniques Acknowlegments

APHIS

January 25th, 2008 3rd Party Modules

C++ Libraries PHP Other TIFF Library Fpdf Windows 2003 Server PNG Library PHP SQL Server JPEG Library Extensions… Internet Information Server AIGridIO from ESRI Shapefile from ESRI Proj4 from DOI GeoTIFF ECW from ERMapper

January 25th, 2008 Research versus Production

Attribute Research Production Quality Accurate Robust

Number of Users Few Lots

Technology Latest and Greatest What works

Learning Curve Typically Long Short

Support ? Must

Documentation Minimal and techy is ok Must be complete and easy to understand

Bottom Line If it’s cool they will come If it doesn’t work, they will go elsewhere

January 25th, 2008 January 25th, 2008 January 25th, 2008 January 25th, 2008 Future Vision

INPUTS OUTPUTS

County-level National-scale maps data on vascular of non-native CLEARING plants (BONAP) species distributions HOUSE

National data on Information Predictive models birds, mammals, and Management and of habitats vulnerable diseases (USGS) modeling (USGS, to invasion NASA, CSU, UCD) Watershed-level •Data gathering Predictive models data on fishes • Species taxonomy of the spread of (USGS) • Data formatting Invasive species • Synthesis • Predictive modeling Point data on public National, regional, • Analysis and lands (USFWS, and local priorities display tools NPS, USGS) for control efforts • Data accessibility via the web Reports on the Vegetation and soils status and trends plot data (USFS, of non-native USGS, BLM) species in the U.S.

July 21, 2006 January 25th, 2008 ---

August 9, 2006 January 25th, 2008 Past and present workflows

Field sampling Historically field notebooks, now some use of handhelds Data entry Historically manual entry into database on single PC Now upload from field hand-held unit directly into DB

August 9, 2006 January 25th, 2008 Colorado’s Worst Invasive Species Top Ten Plants

1.Canada thistle 2.Diffuse knapweed 3.Field bindweed 4.Hoary cress 5.Jointed goatgrass 6.Leafy spurge 7.Musk thistle 8.Russian knapweed 9.Spotted knapweed 10.Yellow toadflax

January 25th, 2008 Extra stuff

January 25th, 2008