Configuration

This guide discusses how to configure Wildbook to fit the study design of your project. Wildbook configuration is done at the file system level in .properties files and .json files. There is no UI to change these files, and we recommend a good Linux command line editor to change them. For changed values to be reflected in the UI, a Wildbook restart may be needed.

Changing Wildbook configuration is a technical task and can make Wildbook inoperable if misconfigured. We highly recommend tracking changes you make to Wildbook in source control. Source control prevents your changes from being lost when you upgrade Wildbook in the future.

Wildbook Configuration Structure#

By default Wildbook configuration options are stored as name-value pairs in .properties or .json files in:

<tomcat_install_dir>/webapps/wildbook/WEB-INF/classes/bundles

Wildbook has a look-here-first override directory for these files hardcoded to:

/data/wildbook_data_dir

Wildbook looks in the override directory first before looking in the installation directory, allowing you to easily maintain secure configurations (e.g., the database password in jdoconfig.properties) in a separate directory while still upgrading Wildbook entirely in the installation directory.

You can edit these files with a text editor, and after making changes, restart Tomcat to see the new configurations take effect.

Translation files#

Properties files placed in language-specific folders, such as "en" (English), "es" (Spanish), etc. represent translated files of the Wildbook user interface (UI).

Changing the database and persistence behavior#

Wildbook uses the DataNucleus Access Platform, which allows it to use a number of different databases and database types. Out of the box, Wildbook looks for PostgreSQL at localhost:5432/wildbook with username 'wildbook' and password 'wildbook'. You must have PostgreSQL installed and running, the 'wildbook' database created, and user 'wildbook' defined. Wildbook can automatically create the necessary tables and columns when it is first started. You do not need to create these yourself.

Database configuration is defined in the file jdoconfig.properties.

You can control Wildbook's database behavior by changing the values of these properties and by adding others, as defined in the DataNucleus documentation. This is recommended for software developers only. We do not recommend databases other than PostgreSQL and use PostGIS extensions in Wildbook.

Changing Wildbook's Appearance#

At its core, Wildbook uses Bootstrap for its mobile, responsive design. Most changes to display and design can be performed by changing:

  • header.jsp - This file defines the standard JavaScript and CSS imports used by Wildbook and implements any header graphics and navigation at the top of the pages.

  • footer.jsp - This file defines any HTML to display at the bottom of the page, such as copyright information.

  • commonConfiguration.properties - This file contains many configuration options for Wildbook, and it includes options for search engine optimization (SEO), page titles, and other basic HTML fields.

The HTML header for web pages in Wildbook is pulled from the file header.jsp. This file retrieves additional values from commonConfiguration.properties and lang/header.properties.

CSS changes#

Wildbook uses Less to create the manta.css file at build time. You'll need to modify the .less files in cust/mantamatcher/styles to make your custom changes.

Encounter Configuration#

Most Wildbook project configuration occurs in the commonConfiguration.properties file. This file is located in the following location in your Wildbook installation:

wildbook/WEB-INF/classes/bundles/commonConfiguration.properties

or in the override director (preferred) at:

/data/wildbook_data_dir/WEB-INF/classes/bundles/commonConfiguration.properties

You can alter important Wildbook Encounter parameters in this file, such as:

  • Number and type of measurements (e.g., weight, length, salinity, temperature, etc.)
  • Number of life stages (e.g., juvenile, sub-adult, adult)
  • Project species
  • Physical patterning codes
  • Physical tag parameters (SAT, PAT, tags, etc.)
  • tissue sample parameters, such as:
    • number of alleles
    • tissue sample storage method
    • haplotypes
    • biochemical measurements (13C, 15N, etc.)
  • and more...

When you modify configuration parameters in this file, you may have to restart the Tomcat web server for the changes to take effect.

Supporting multiple species#

By default, Wildbook assumes that information added to the database is for only one species of animal. You can configure Wildbook to support multiple species by setting the genusSpecies fields, and setting the showTaxonomy parameter to ''true'' in commonConfiguration.properties. The following example shows how to set these properties to differentiate between three species of large cats.

#show taxonomy showTaxonomy = true

#for multi-species libraries, fill out the genus and species for each supported animal type, starting with genusSpecies0 genusSpecies0 = Panthera leo genusSpecies1= Panthera tigris genusSpecies2 = Panthera pardus

Additional species can be set by incrementing the properties, such as genusSpecies3, genusSpecies4, etc.

Configuring encounter states for workflow#

Encounters can be assigned to particular states, often reflecting critical parts of a project workflow. For example, states can be used to define what data has been 'approved' for use in your study or is 'unidentifiable' due to poor quality. By default, new data in Wildbook is put into an 'unapproved' state that represents its status as unreviewed and of uncertain quality.

The available states are defined in the following section of commonConfiguration.properties.

#encounterState options, the precursors to future workflow encounterState0=unapproved encounterState1=approved encounterState2=unidentifiable

Additional states can be set by incrementing the properties, such as encounterState3, encounterState4, etc.

Configuring location IDs (study sites)#

Defining the boundaries of study areas is an important aspect of mark-recapture study design. Wildbook allows you to configure study areas as an attribute that can be assigned to each Encounter, categorizing data by where it occurred. Encounter.locationID is one of three ways of defining location in Wildbook. The three ways are:

  • Encounter.locationID - assigns an Encounter to a human-defined study area
  • Encounter.decimalLatitude/decimalLongitude - assigns an Encounter to precisley measured GPS coodinates
  • Encounter.verbatimLocality - a general description of location provided by the data submitter from which locationID may be determined.

Encounter.locationID may be the most valuable form of location identifier in Wildbook. GPS location may not be known, and verbatim descriptions of location may vary significantly by submitter. LocationID is powerful because it specifically defines whether a data point should be included or excluded from a study area during mark-recapture analysis and can be used to define which animals to match against in the Wildbook Image Analysis pipeline.

To configure the list of study sites in your Wildbook, change the values in locationID.json. Click here for more information about location ID and its configuration.

Configuring encounter measurements#

You can configure reported encounters to have multiple recorded measurement observations (e.g., length, width, height). The measurement types are defined in commonConfiguration.properties and in lang/commonConfigurationLabels.properties, starting with the number 0 and incrementing by one for each new measurement type.

For each measurement, you must define a measurement name, the corresponding units, and the general sampling protocols applied to measurements in your study. You can turn and off measurements altogether with the showMeasurements entry, which can be set to true or false.

#show measurements showMeasurements = true

#Measurements measurement0=weight measurement1=length measurement2=height

measurementUnits0=kilograms measurementUnits1=meters measurementUnits2=meters

#Sampling Protocol for all Measurement types samplingProtocol0=estimate samplingProtocol1=measure

Configuring and localizing measurement labels#

The names of measurements, their units, and sampling protocols can be localized into multiple languages in Wildbook in the language-specific copies of commonConfigurationLabels.properties. Notice the .label extension added to the base name of the measurement-related property from commonConfiguration.properties (see above).

#Labels for Measurements weight.label=Weight length.label=Length height.label=Height 13C.label=13C 15N.label=15N 34S.label=34S celsius.label = Celsius salinity.label = Salinity WaterTemperature.label = Water Temperature kilograms.label=Kilograms meters.label=Meters ppm.label=ppm

samplingProtocol0.label=Estimated Value samplingProtocol1.label=Directly Measured

After making changes, restart Tomcat to see them take effect.

Configuring life stages#

Observations of animals are often categorized by their life stage, such as adult, sub-adult, juvenile, hatchling, etc. You can configure the life stages for your project with the following settings in commonConfiguration.properties.

#show lifeStage showLifestage = true

#defined life stages lifeStage0=juvenile lifeStage1=sub-adult lifeStage2=adult

Additional stages can be set by incrementing the properties, such as lifeStage3, lifeStage4, etc.

Configuring physical tag metadata#

If you're using physical tags (SAT, PAT, marker, etc.) during your research, you can record metadata about those tags and associate it with an Encounter recorded at tag deployment. The following attributes (examples below) are available to configure:

  • showMetalTags - whether physical marker tags are used in your study should be shown on the Encounter page in Wildbook. Values are true/false.
  • metalTagLocationX - The list of placement locations for the tag on the body of the animal
  • showAcousticTag - whether acoustic tags are used in your study should be shown on the Encounter page in Wildbook. Values are true/false.
  • showSatelliteTag - whether satelite tags are used in your study should be shown on the Encounter page in Wildbook. Values are true/false.
  • satelliteTageNameX - The list of satellite tag manufacturers/providers to help identify the type of tag.
#tag parameters showMetalTags=true metalTagLocation0=left metalTagLocation1=right

showAcousticTag=true

showSatelliteTag=true satelliteTagName0=Wild Life Computers satelliteTagName1=SirTack

Configuring visual patterning codes#

Often times, the visual coloration of species can be divided into categories to allow for easier filtering for individual identification. For example, humpback whales can be individually identified by their flukes and have a graded fluke coloration system, ranging from 1-5.

Wildbook allows each Encounter to be assigned a patterning code to match the observed visual features of the animal in the Encounter.

You can configure pre-defined patterning codes for your project in commonConfiguration.properties. The property showPatterningCode must be set to true, and then a sequentially numbered set of patterningCodeX names and values must be defined.

Here is a configuration example of giant manta patterning codes from the Wildbook MantaMatcher.org.

#defined patterningCodes for distinct visual marking types for individuals identified with photo-identification showPatterningCode = true

patterningCode0 = normal pigmentation patterningCode1 = black pigmentation - melanistic patterningCode2 = white pigmentation - leucistic

Configuring elevation and depth#

You can configure whether elevation or depth are recorded in your study as Encounter properties in the commonConfiguration.properties file. Set maximumElevationInMeters to true to make elevation a measurement of your study. Set maximumDepthInMeters to true to make depth an measurement of your study.

#show elevation/depth

maximumElevationInMeters = false maximumDepthInMeters = true

Restart Tomcat after making changes.

Configuring tissue samples and analyses#

Wildbook allows you to add one ore more records for biological samples collected during an Encounter with an individual animal. Biological samples have the following attributes defined in commonConfiguration.properties:

  • tissueTypeX - a sequential definition of the types of tissues that can be collected in your study (e.g., fecal, blood, skin, biopsy, etc.)
  • biologicalMeasurementTypeX - a sequential definition of the types of chemical measurements that can be analyzed and determined from tissue samples in your study (e.g., fatty acid Carbon and Nitrogen measurements)
    • biologicalMeasurementUnitsX - a sequential definition of the units of measurement for chemical measurements on tissue samples
    • biologicalMeasurementSamplingProtocolX - a sequential definition of the protocols used to determine the measurements
  • numLoci - the number of loci to provide allele values for if genotyping is performed as an analysis on the tissue sample
  • numPloids - leave at 2 per loci
  • alleleRelaxMaxValue - the value difference allowed between allele values to consider them a "match" when looking for other genotypes that might match or be similar.

#tissue sample types tissueType0 = Tissue sample tissueType1 = Fecal sample tissueType2 = Mucus sample tissueType3 = Blood sample tissueType4 = Parasite sample

#biological measurement types biologicalMeasurementType0 = 13C biologicalMeasurementType1 = 15N biologicalMeasurementType2 = 34S

#corresponding biological measurement units biologicalMeasurementUnits0 = ppm biologicalMeasurementUnits1 = ppm biologicalMeasurementUnits2 = ppm

#corresponding biological measurement sampling protocols biologicalMeasurementSamplingProtocols0 = Lipids extracted biologicalMeasurementSamplingProtocols1 = No lipids extracted, corrected biologicalMeasurementSamplingProtocols2 = No lipids extracted, uncorrected

#genetic parameters numLoci = 14 numPloids = 2 alleleRelaxMaxValue = 5

Restart Tomcat after making changes.

Custom content display#

You can import custom functionality into each Encounter's display in encounter.jsp. For example, spot pattern matching-related functions are imported from file spotMatchingAlgorithm.jsp (see example below).

Import definitions are found in file encounter.properties file in the localized directory (e.g., WEB-INF/classes/bundles/en/encounter.properties).

#define the module JSP files to import #files must be placed in the encounters directory of the Wildbook webapp (i.e., in the same directory as encounter.jsp) jspImport0=spotMatchingAlgorithm.jsp #jspImport1=myFile.jsp

Restart Tomcat after changing values.

Marked Individuals#

The following aspects of Marked Individual data records can be configured in Wildbook.

====Social relationships==== Marked Individuals in Wildbook can have Relationship associations between them. These are meant for more long-term types of relationships, such as "mother-calf" or social group "member". Relationships and their roles can be set in the UI on the Marked Individual page individuals.jsp.

To configure the types of social relationships in your Wildbook and the roles that Marked Individuals can play in them, change the following fields in commonConfiguration.properties:

  • relationshipTypeX - a sequentially numbered set of relationship types to be recorded in your study.
  • relationshipRoleX - the role that an individual can play in a Relationship.

Here are some example configurations:

#social relationships-related data relationshipType0 = social grouping relationshipType1 = familial

relationshipRole0 = member relationshipRole1 = mother relationshipRole2 = calf

Restart Tomcat after making changes.

Email Notifications#

The following email addresses are sent messages when important events occur in Wildbook. These email addresses are specific to your project.

  • sendEmailNotifications - Defines whether Wildbook should send any emails at all. Set to true, this parameter instructs Wildbook to send email updates for several events, such as:

    • new Encounter submission reports

    • new marked individual identifications to adopters ad submitters

      MarkedIndividual resights to adopted and submitters

  • autoEmailAddress - The email address from which Wildbook messages will be sent. If you have a secured mailhost, it's important to configure the mailhost to allow emails to be sent from this address.

  • newSubmissionEmail - The email address to send notices of new Encounter reports submitted to Wildbook.

#email addresses and parameters

sendEmailNotifications=true autoEmailAddress=webmaster@someplace.org newSubmissionEmail=submissions@someplace.org mailHost=localhost

Internationalization (I18N)#

Wildbook is a web-based application with an internationalization (I18N) foundation in its code base. Wildbook is not yet fully translated. Most of the JSP files are localized but none of the servlets are yet. Work is underway to complete localization.

Wildbook loads translated strings from the standard Java properties files, which reside in Wildbook's WEB-INF/classes/bundles directory or can be overridden (on a file-by-file basis) in the equivalent path in the data directory.

Setting the supported languages#

You can set the default language for Wildbook in commonConfiguration.properties with the defaultLanguage attribute, which takes a two-letter code for input. The full list of supported language, which must include the default in position 0, is configured with sequential language//X// name-value pairs. Corresponding localized Strings could be found in the appropriate language directory of Wildbook, such as WEB-INF/classes/bundles/es for 'es' (Spanish).

In the example below, English is set as the default language "en", and Wildbook is configured to provide the user with English and Spanish options for display.

defaultLanguage = en

language0 = en language1 = es #language2 = fr #language3 = de

The name displayed for each language can also be configured in commonConfiguration.properties.

en = English es = espa\u00f1ol fr = français

With languages configured and Tomcat restarted, Wildbook displays configured languages as flag options.

It looks for a correspinding flag icon in the images directory. Examples:

  • flag_es.gif
  • flag_en.gif
  • flag_fr.gif

If your language flag is not present in Wildbook, you may need to add it.

Translating text#

Most pieces of text in Wildbook pages (HTML and JSP files) are contained in translatable properties files. For example, the page for an Encounter (encounter.jsp) pulls its strings from WEB-INF/classes/bundles/en/encounter.properties (English text). Here is an example of the contents of encounter.properties.

encounter = Encounter

unidentifiable_title = Unidentifiable Encounter Number unapproved_title = UNAPPROVED Encounter Number title = Encounter identified_as = Identified as: workflowState = Workflow state: setWorkflowState = Set Workflow State matched_by = Matched by status = Status ...

===== Legacy Spot Matching ===== Wildbook was originally developed for use with whale sharks (Rhincodon typus) and other spotted species. The internal spot pattern recognition system is highly accurate and had scaled well to over 25000 patterns on whaleshark.org. While our newer pattern recognition efforts in Wildbook Image Analysis will eventually replace them, the legacy spot pattern recognition system is still highly accurate and usable in Wildbook.

By default, the spot pattern recognition system is turned off. It can be turned back on in commonConfiguration.properties by setting the useSpotPatternRecognition value to true.

#pattern recognition

useSpotPatternRecognition=true