0300-the-georeferencing-process.en.adoc

The Georeferencing Process

Locations that are not fully georeferenced in the field may eventually have to be georeferenced after the fact in order to be useful. One hopes in these situations that the collector of the original information followed best practices such as those described in [Elements for Describing a Location]. As will be seen in Calculating Uncertainties, below, many of the greatest sources of [uncertainty] arise from missing, ambiguous, or non-specific information, which could have been avoided, but that can no longer be overcome without knowledge from someone who was there at the time the [event] occurred.

Planning a Georeferencing Project

Before beginning a georeferencing project, whether for an individual researcher or a large institution, it is helpful to anticipate the kinds of challenges one might expect to encounter. It may appear to be a daunting task, but there are many ways the process can be simplified and made more practical. Having a suitable workflow (see Georeferencing Project Workflow) decided in advance can increase both the efficiency and the consistency of the quality of the resulting georeferences. The basic determiners for a project are what you have to start with and what outputs you want when you are finished. In an ideal world, the obvious practical questions such as the cost and how long it will take would not be important, but realistically, when balanced against the benefits of making the effort, these might be the major determining factors. Following is a representative list of questions that might affect planning of a georeferencing project:

• About the original data:

  • Where is the source data coming from (herbarium labels, ledgers, database, or a combination, etc.)

  • Are the source data already digitized?

  • How many distinct [locality] descriptions are there to georeference?

  • Are there terrestrial or marine locations to georeference? Or both?

  • Is the geographic scope local? Country-wide? Global?

• About the process:

  • What is the time frame for the project?

  • When in the broader workflow will georeferencing happen?

  • How much of the established best practices do I really need to follow?

  • What if we want to use our own methods?

  • What procedural documentation will we need to prepare?

  • Who will do what?

  • What expertise is needed?

  • What skills do those who will be involved possess?

  • Where can training be found?

  • What location resources (maps, gazetteers, tools) are available?

• About the final product:

  • What data quality target is there for the georeferences?

  • How will data validation take place?

  • How will the data be maintained?

  • How will the georeferenced data be used and by whom?

  • Will the georeferences be generalized on export (for sensitive species, for example)?

  • How can the georeferences be integrated back into the original data?

  • How can we incorporate suitable [data quality] feedback mechanisms?

The question, "When in the broader workflow will georeferencing happen?" is of particular significance. Is it best to georeference each record as you enter the data into the database? Or is it better to [georeference] in a batch after the data have been entered? There are arguments for each method, and again the circumstances of your institution should dictate the best method. If the data are stored taxonomically and not geographically (as is the case in the majority of instances) it is often best to georeference in a batch mode by sorting the locality data electronically, and in this way deal with many records on one map sheet or area at a time and not jump back and forth between map sheets. In other cases, it may be more important to minimize wear and tear on collections, and you may wish to database collections as they are received before distributing duplicates or sending on loan. There may be other good practical reasons to georeference as you go. One advantage of georeferencing as you go is that you may be able to do all the collections of one collector at a time, and virtually follow his/her path, thus reducing errors from not knowing which of several localities may be correct.

This document does not cover methods of general data entry. There are many ways that this may be conducted, including direct entry from the field notes, labels, or ledgers with the material brought to a data entry computer; or direct entry where tablets or laptops are brought to the material. There are also indirect methods, such as entry after using scanning or photographic (still or video) equipment to capture the original information so that data entry can be done after imaging. Capturing the images allows for digitization using handwriting and optical character recognition tools and crowdsourcing (e.g. [DigiVol], [Notes from Nature]). Some of these methods are just becoming practical, but you should make an active decision on the method that best suits the needs of your project. When digitization is in progress and each specimen is being handled, it is a good time to consider other actions, such as georeferencing (though we do not recommend this for the sake of efficiency), assigning persistent identifiers (PIDs) (see [Persistent Identifiers (PIDs)]), barcoding specimens and linking these to the database to save resources further down the line.

It is also important that the long-term maintenance of the data is considered early in the process. Project managers may wish to consider questions such as:

• How are we going to deal with corrections to the data?

• How do we handle feedback on data quality from data aggregators, data users, etc.?

• Do we have a process in place for documenting changes to the data?

• Have we budgeted sufficient resources for ongoing maintenance and data quality checking?

Georeferencing Project Workflow

A workflow covering all the georeferencing activities can be a valuable instrument, not only for improving the efficiency of the whole georeferencing process, but also for incorporating checks and balances, and improving the quality of the resulting product. The type of workflow may be determined by the nature of the data, the way the original data are stored or documented, the nature of the desired end product, and even by the general preferences of those involved. In the following subsections we propose a generic workflow that covers all of the major aspects of georeferencing projects. Note that some of the steps presented might not apply to every project, and one must take into account priorities as discussed in Planning a Georeferencing Project. This first section, Georeferencing Project Workflow outlines a recommended georeferencing project workflow in four phases. Subsequent sections deal with the details of some of the steps presented in the outline.

Based on an assessment of a variety of large-scale georeferencing projects that had efficiency as well as data quality in mind (e.g. Project Workflow Example – MaNIS/HerpNET/ORNIS), we recommend the following generic outline for a georeferencing project workflow using either the [point-radius] method or the [shape] method, or a mix of the two. This workflow can be used for projects that involve a single individual or a large collaboration, though some steps may apply more in one case than in another. Note that some of the actions included in different phases might happen simultaneously depending on the type and scale of the project.

Project Preparation Phase

• Commit to the use of a documented set of best practices such as those set forth in this document.

• Clearly define (and document) the goals of the project, including [data quality] requirements (see Planning a Georeferencing Project).

• Determine what data will be used as input for georeferencing.

• Select the tools to be used.

• Estimate the resources needed to complete the georeferencing preparation phase (see Resources Needed).

• Assign someone to manage the project.

• Acquire the resources needed to start the project.

Georeferencing Preparation Phase

• Assemble the data to be georeferenced.

• Prepare the data for georeferencing:

  • Make sure that original records are uniquely identified (ideally with PIDs, see [Persistent Identifiers (PIDs)]).

  • Make sure that original data are captured and safe from alteration during the georeferencing process.

  • Extract distinct combinations of all locality-related fields (including administrative geography, [elevation], etc.), generate unique identifiers (ideally GUIDs, see [Persistent Identifiers (PIDs)]) for each, and reference the corresponding locality identifier in each original record.

  • Use source-provided administrative geography fields to create and add standardized administrative geography values to the distinct locality records. This will help with the organization of georeferencing by region as well as facilitate lookups against geographic authorities. Optionally, extend this standardization to the contents of the specific locality fields as well. Though this approach has been taken in some large-scale georeferencing efforts such as those undertaken by CONABIO and SiB Colombia (Escobar et al. 2016), there is no clear evidence that the reduction in the number of distinct localities warrants the effort required to do this standardization. More research in this area is needed.

  • Label localities as marine, terrestrial, freshwater aquatic, or paleontological. The same locality description may refer to more than one category (e.g. locations on coasts) unless further constraining information is used (see Applying Spatial Constraints). If dealing with localities alone, you should account for all of the environmental possibilities.

  • Create and uniquely identify distinct standardized localities and reference the standardized locality GUID in the non-standardized locality records.

  • Match standardized localities against existing localities that have already been georeferenced using satisfactory georeferencing methods and extract the existing georeferences (see Using Previously Georeferenced Records).

• Assess the characteristics of the data to be georeferenced (e.g. how many already have [coordinates] without georeferences? How many consist only of administrative geography? What is the geographic distribution of the localities?) with a view to determining the resources that will be needed to complete the project.

• Estimate the resources needed to complete the project using the information determined in the project preparation phase.

• Acquire the resources to complete the project.

• Train participating contributors and georeferencing operators (see [Data Entry] and [Training]).

• Establish a convention and tools to manage participation (assignments).

• Prepare data capture requirements and tools (see Data to Capture, User Interfaces, Using Standards and Guidelines, and [Mapping to Darwin Core]).

• Assign priorities to sets of standardized localities.

• Assign standardized locality sets to participants.

Georeferencing Phase

• Participants [georeference] assigned [locality] sets as outlined in Georeferencing Workflow – Localities.

• Participants utilize protocols and tools such as the {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^] and the Georeferencing Calculator (Wieczorek & Wieczorek 2020).

Project Follow-up Phase

• Verify georeferences to meet [data quality] requirements (e.g. map georeferenced records to ensure they fall in the correct hemisphere, country, etc.) (see [Data Checking and Cleaning]).

• Populate standardized [locality] records with data for the georeferences.

• Normal curatorial activity is not usually suspended during a georeferencing project, which opens the possibility that locality information could be changed for some records in the source database after being aggregated for georeferencing and before being re-incorporated in the source database. For database records that did not have changes to the [locality] information before re-incorporation, populate the original records from the standardized locality records with georeferences.

• Repatriate the original records with standardized georeferenced locality data appended to the corresponding institutions (this step is mostly relevant in collaborative projects).

• Support the incorporation of the standardized georeferenced locality data into the source data management systems (see [Accepting Feedback from Users]).

• Support the sharing of the standardized georeferenced original data (including additional generalizations and withholdings) in open data venues such as GBIF (see [Sharing Data]).

• Establish a long-term data maintenance policy that includes the management of feedback on data quality and the documentation of changes (see [Accepting Feedback from Users]).

Project Workflow Example – MaNIS/HerpNET/ORNIS

One of the major contributions of the Mammal Networked Information System (MaNIS) project (Stein & Wieczorek 2004) was the design and implementation of a set of georeferencing guidelines (Wieczorek 2001) and online resources for a collaborative georeferencing workflow. The same basic workflow was implemented with great success for the sister projects HerpNET and the Ornithological Information System (ORNIS). Between the three projects, more than 1.2 million localities were georeferenced for 4.5 million vertebrate occurrence records. The basic workflow was more or less as follows:

• Establish a [georeferencing method] and select tools to be used.

• Train participants (combination of help desk, forum, documents, and in the case of HerpNET, courses).

• Establish a convention and tools to manage georeferencing work packages for participants.

• Aggregate occurrences and extract distinct localities into a project [gazetteer].

• Engage participants to claim and complete (georeference) work packages.

  • Participant downloads work package.

  • Participant georeferences work package, consulting documentation and colleagues to resolve questions.

  • Finished work package is sent to the project coordinator.

• Project coordinator validates georeferences to meet [data quality] standards.

• Project coordinator populates communal [gazetteer] with validated georeferences.

• When georeferencing is completed for the entire project, project coordinator validates that localities for original occurrence records have not changed from the sources since they were added to the gazetteer and repatriates occurrence records with georeferences to participating data custodians.

• Everyone involved rejoices.

• Participants add georeference data to their data management systems as time and resources allow.

• Georeferenced occurrence records get shared via global biodiversity networks such as VertNet (Guralnick & Constable 2010) and GBIF.

Using Previously Georeferenced Records

It may be possible to use a look-up system that searches for similar localities that have already been georeferenced. For example, if you have a record with the [locality] "10 km NW of Campinas", you can search for all records with locality "Campinas" and see if any records that mean the same thing as "10 km NW of Campinas" have been georeferenced previously. Note that it is always worth verifying the georeference on a map — this can easily be done using software such as Google Maps, Google Earth, etc. Checking this way can reduce errors such as neglecting to add the minus (−) sign to a coordinate in the western or southern hemispheres.

An extension of this method could use the benefits of a distributed data system such as GBIF.org. A search could be conducted to see if the locality has already been georeferenced by another institution. At present, we quite often find that duplicates of occurrence records have been given significantly different georeferences by different institutions. Presumably this would not happen if best practices were followed, or if georeferencing is done by the original institution before distributing duplicates.

A preliminary study (Wieczorek pers. comm.) of roughly 33.1 million occurrences for 38.7 thousand plant taxa in GBIF from 15 April 2019 (GBIF 2019) showed that the records were associated with 7.2 million distinct locations, of which 25.7 per cent (30.9 per cent of occurrences) already had georeferences (i.e. term:dwc[decimalLatitude], term:dwc[decimalLongitude], term:dwc[geodeticDatum] and term:dwc[coordinateUncertaintyInMeters]). Of those without georeferences, exact matches (on geography plus locality fields, all turned into upper case) from other locations in GBIF could be found for 2.5 per cent of distinct locations (11.4 per cent of occurrences).

In the case where multiple possible georeferences are found using a lookup on previously existing georeferenced locations, the problem is knowing which of the several georeferences, if any, to choose.

If the georeference is not fully documented following best practices (including being reproducible), we recommend that existing georeferences not be used (or used only with extreme caution). Even if the georeference is documented, it should be checked visually on a map to be sure that it makes sense, just as for any new georeference.

Caution

The re-use of existing georeferences can propagate errors, if a mistake was made the first time. Existing georeferences should be verified just as for any newly generated georeference.

Resources Needed

Each institution will have needs for different resources in order to [georeference] their [location] data. The basics, however, include:

• Suitable computer hardware to support all of the below.

• A database and database software (spreadsheets may be apt for data capture, but they leave a lot to be desired compared to databases for data management, for which we do not recommend the use of spreadsheets). Note that there are a lot of database management systems already established and available for use with biodiversity data. See if any of these may do the job before developing your own as it may save a lot of extra work. Many also already include [data quality] aspects that could help improve the quality of your own data.

• Topographic or bathymetric maps (electronic, paper or both), geologic maps (for paleontological events) and/or speleological maps (for events in cave systems).

• Access to good gazetteers (many are available free via the Internet, either for downloading, or via online searching).

• Internet access (as there are many resources on the Internet that will help in georeferencing and locating places).

Data to Capture

One of the most important preparation steps for efficient georeferencing is to have an effective way to handle the data. This section will help you decide if your data capture framework will need modification or not, and to what extent.

Some georeferencing projects (e.g. MaPSTeDI (Murphy et al. 2004)) used a separate working database for data entry operators so that the main data were not modified and day-to-day use of the database was not hindered. This also meant that the working database could be designed optimally for data entry, rather than trying to accommodate other database management and searching requirements. The data from the working database can be checked for quality, and then integrated into the main database from time to time. Such a way of operating is institution dependent, and may be worth considering.

What are the fields you need in your database to best store georeferencing information? This may seem obvious, but it is surprising how often a database is created and finalized before it is determined exactly what the database is supposed to hold. Be sure not to lump together dissimilar data into one field. Always atomize the data into separate fields with very specific definitions and rules for their content.

It is also of benefit to name the fields unambiguously, as users tend to go by the field names rather than looking at the field definitions. Thus, 'latitude_in_degrees' is a better name than '[latitude]' for a field that is supposed to contain latitudes in [decimal degrees], while 'verbatim_latitude' is better name for a field that is supposed to contain the [latitude] in the format given in the source. The names and definitions of fields in [Darwin Core] (Wieczorek et al. 2012b) were created specifically with this principle of clarity in mind. In order to take advantage of a community standard set of definitions, it is not a bad idea to use the term names from Darwin Core as field names in the database if the semantics of the two are the same.

Note, however, that the georeferencing results might benefit from additional fields that are not described in Darwin Core (e.g. '[feature]_[radial]', 'radialUnits') in order to make it easier to reproduce the georeference and thus test its veracity. It is often tempting to include fields for the georeferenced [coordinates] and ignore any additional fields; however, you (or those who follow after you) are sure to regret this minimalist approach, because it severely limits the usability of the data. A Location occupies a physical [extent], not just a point. The associated information on methods used to determine the georeference, the extent, [radial], and [uncertainty] associated with the georeference are important pieces of information for the end user, as well as for managing and improving the quality of your information. The fields that are needed can be divided into two categories: the first consists of the fields associated with the textual description of the location, and the second consists of the fields associated with the spatially enabled interpretation as a georeference and the georeferencing process.

Note	When atomizing data on entry, always include a field or fields that record the original data in its verbatim form so that atomization and other transformations can later be revealed and checked.

Note	Automatic format transformations to [decimal degrees] may introduce [false precision]. See §1.6.

Note

Be careful with any automatic data formatting or transformation in your database, especially when incorporating original data. Sometimes databases are set to have a particular type or format for data in a given field (e.g. numbers, dates, etc.), which can change the original data and result in irrecoverable losses of information. In this sense, it is recommended that you set all verbatim data fields to be of type "text". Also be aware of the encoding of the data upon import and export, because if the encoding of the data does not match the encoding of the destination, the data can be corrupted.

Note	It is always advisable to test the structure of your database with a small sample of records before committing to using it for the whole project. In doing so, you may detect additional fields that are needed and/or fields that require definition review or that are not used at all.

A reference worth checking before developing your own database system is the Herbarium Information Standards and Protocols for Interchange of Data (Conn 1999, Neish et al. 2007), which, although set up for data interchange between herbaria, is applicable to most data from natural history collections.

Many institutions separate [locality] descriptions into their component parts; [feature] name, distance, [direction], etc., and store this information in separate fields in their databases. If this division of locality information is done, it is important not to replace the verbatim free-text locality field (the data as written on the label or in the field notebook), but instead add additional fields. This is because any transformation of data has the potential to lose information and to introduce errors, and the written format of the description may be the only original source available. The original information should never be overwritten or deleted.

Location-related fields to consider for georeferencing include all of the geography, locality, [elevation], [depth], and georeference terms in the Location class of Darwin Core (see term:dwc[location] and [Mapping to Darwin Core]) as well as the following fields that can have an influence on the georeference:

• As many levels of administrative subdivision as necessary (e.g. country, state, county, municipality, etc.), though if the geographic scope is multinational, it is better to name the administrative subdivisions more generically to avoid confusion (e.g. country, geog_admin_1, geog_admin_2, etc.)

• Feature name, feature-type, [offset] distance, offset direction, offset units

• Feature [shape], feature center, feature radial

• Township, range, section, subsection or similar for other [grid] systems

• Protected area

• Watershed

• Map quad

• [UTM] [easting], [northing], and zone

• For {marine}marine locations － nearest island, exclusive economic zone, etc.

• Elevation [accuracy], [vertical datum], and the method of determining elevation

• Depth [accuracy], vertical datum, and the method of determining depth

• Latitude degrees, latitude minutes, latitude seconds, latitude hemisphere, [longitude] degrees, longitude minutes, longitude seconds, longitude hemisphere

• Biome, to distinguish terrestrial, freshwater aquatic, and marine locations

• Event date (best to follow and enforce a standard format, such as ISO 8601 (ISO 2019). Note that if your project is dealing with [location] information only (dissociated from occurrence or event records), this may not be possible or advisable.

• Fields in the Darwin Core term:dwc[GeologicalContext] class for paleontological occurrences

Note

When adding extra fields to your database, always consider that the more fields you add, the higher the chances that data entry operators could make a mistake. Therefore, although having more fields has many advantages when it comes to checking the results, try to avoid over parsing information if not really necessary.

Applying Data Constraints

One of the key ways of making sure that data are standardized and accurate is to ensure, to the extent possible, that data are put in the correct field and that only data of an appropriate type can be put into each field by design. This is done by applying constraints on the data fields – for example, only allowing values between +90 and −90 in the field for [decimal latitude]. Many of the errors found when checking databases could have been easily avoided if the database had been set up correctly in the first place. The use of pick lists is essential where the field should contain only values from a restricted list of terms.

More complex constraints may also be possible. With {ecological}ecological or survey data for example, one could set [boundary] limits between the starting [locality] and ending locality of a [transect]. For example, if your methodology always uses 1 km or shorter transects, then the database could include a boundary limit that flagged whenever an attempt was made to place these two points more than 1 km apart.

For more information on constraints, see various sections under Uncertainty Due to the Extent of the Feature.

User Interfaces

Good user-friendly interfaces are essential to make georeferencing efficient and rapid, and to cut down on operator errors. The design should take into consideration the specific details of the georeferencing workflow, and optimize simultaneously for both overall efficiency, and consistency of the data entry process. This will improve accuracy and cut down on errors. The layout should be friendly, easy to use, and easy on the eyes. Where possible (and the software allows it) a number of different views of the data should be presented. These views can place emphasis on different aspects of the data and help data entry operator proficiency by allowing different ways of entering the data and by presenting a changing view for the operator.

In the same way, macros and scripts can help with automated and semi-automated procedures, reducing the need for tedious (and time-consuming) repetition. For example, if the data are being entered from a number of collections by one collector, taken at the same time from the same [location], the information that is repeated from record to record should be able to be entered using just one or two keystrokes.

If maps are being used to assist in determining georeferences, a view that sorts the data geographically may also make the process more efficient by allowing the data operator to see all the records that may fall on one map sheet. Finally, it is also important to decide which fields the data entry operators should see when they are georeferencing. Fields such as date of collection, collector, specimen ID, taxonomy, habitat, and formation (for paleontological records) are very helpful for georeferencers to see along with the more obvious locality data.

Using Standards and Guidelines

Standard methodologies, in-house standards, and guidelines can help lead to consistency throughout the database and cut down on errors. A set of standards and guidelines should be established before any georeferencing begins (see [Documentation]). They should remain flexible enough to cater for new data and changes in processes over time, though careful thought beforehand can minimize the need for methodological changes, which might lead to inconsistencies where earlier efforts are lacking compared to those produced under newer protocols. Standards and guidelines in the following areas can improve the quality of the data and the efficiency of data entry:

• Units of measure. Use a single unit of measure in interpreted fields. For example, do not allow a mixture of feet and meters in [elevation] and [depth] fields. Irrespective of this, the original units and measurements should be retained in a verbatim field.

• Methods and formats for determining and recording [uncertainty] and [extent].

• Required fields (fields that must have meaningful, non-empty values).

• Format for recording [coordinates] (e.g. degrees/minutes/seconds, degrees/decimal minutes, or [decimal degrees] for [latitude] and [longitude]).

• Original source(s) of place names and features.

• Dealing with typographical errors and other errors in the existing database.

• Number of decimal places to keep in the various fields with decimal numbers.

• How to deal with "empty" values as opposed to the numerical value zero (Note: configure databases to not supply 0 for an empty value).

• How to deal with mandatory fields that cannot be filled in immediately (e.g. because a reference has to be found). There may be a need for a default value that flags that the information is still required.

• Methods for data validation that will be carried out before a record can be considered complete and verified.

Determining and documenting your institution’s own georeferencing best practice in manuals, for example that suit the circumstances of that institute (including language, local software and resources, etc.) can help maintain consistency as well as assist in training and [data quality] recording. As an example, see Escobar et al. 2015, where an internal document for the Alexander von Humboldt Institute in Colombia has been developed and put into practice. See also [Documentation].

Data Entry Operators

One of the greatest sources of georeferencing [error] is the data entry process. It is important that this process is made user-friendly and set up so that many errors cannot occur (e.g. through the use of pick lists, field constraints, etc.). The selection and training of data entry operators (see under [Training]) can make a big difference to the final quality of the georeferenced data. As mentioned earlier, the provision of good guidelines and standards can help in the training process and allow for data entry operators to reinforce their training over time.

Georeferencing Workflow – Localities

At the heart of any georeferencing project is the hands-on georeferencing of individual [locality] descriptions. The value of getting this part right can’t be overstated.

Regardless of what other steps might have preceded this in a project workflow, for individual localities we recommend the following georeferencing workflow — refined from Wieczorek et al. 2004.

• Choose the [georeferencing method] (e.g. [point-radius], [bounding-box], [shape]) to use. You may do this for all localities or on a case by case basis (see Georeferencing Methods).

• Parse the locality into locality clauses (see Parsing the Locality Description).

• Identify the [feature](s) and determine the [locality type] of the most specific [locality clause] (see Classifying the Locality Description).

• Find the feature(s) in a spatial data source (e.g. map, [gazetteer], GIS layer, application programming interface (API)) that can give you an idea of where the feature is with [coordinates], a bounding box, a point-radius, or a shape).

• Determine the boundaries of the feature(s) (see Setting the Boundaries of the Feature) including all constraints (see Applying Spatial Constraints).

• Follow the protocol in the {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^] to georeference the locality based on the locality type of the most specific clause and the shape or point-radius of the constrained feature from the previous step.

• Document the sources and methods sufficiently to make the resulting georeference reproducible (see [Objectives]).

Though the list of steps above applies to a single locality record, the most efficient way to implement these steps might be to do each step for all of the localities in the set, and use the results of that step to organize the next step. For example, by identifying the features from all of the most specific clauses, one could filter localities by feature and with the accumulated body of information about the feature from all the localities at hand, georeference all of the localities containing the same feature together. One could also do statistics on the number of records affected by determining the boundaries of each feature and use that to prioritize which localities get georeferenced, if resources do not otherwise cover georeferencing everything. This kind of feature extraction could be done in the aggregate georeferencing preparation stage (see Georeferencing Project Workflow).

Parsing the Locality Description

Locality descriptions are often given in free text and encompass a wide range of content in a vast array of formats. An important part of the georeferencing process is to have a consistent way to interpret the text into spatial forms that can be operated on analytically. To do this, look for the parts of the description that can be interpreted independently, called locality clauses, each of which can be categorized into a [locality type] (see Classifying the Locality Description) that uses a specific set of rules to georeference (Wieczorek et al. 2004).

Classifying the Locality Description

There is a lot of variation in the way clauses are written and the types of features they reference, but there are actually very few basic locality types, though these may have many variations depending on the feature type referenced. The {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020^]) was written specifically to explain how to [georeference] all of the most common variations of locality types and [feature] types (Wieczorek et al. 2004):

• [coordinates] only (e.g. 27°34'23.4" N, 121°56'42.3" W)

• geographic feature only (e.g. "Bakersfield")

• distance only (e.g. "5 mi from Bakersfield")

• [heading] only (e.g. "North of Bakersfield")

• distance along a [path] (e.g. "13 miles east (by road) from Bakersfield")

• distance along orthogonal directions (e.g. "2 miles east and 3 miles north of Bakersfield")

• distance at a heading (e.g. "10 miles east (by air) from Bakersfield")

• distances from two distinct paths (e.g. "1.5 miles east of Louisiana State Highway 1026 and 2 miles south of U.S. Highway 190")

• dubious (e.g. "presumably central Chile")

• cannot be located (e.g. "locality not recorded")

• demonstrably inconsistent (e.g. "Sonoma County side of the Gualala River, Mendocino County")

• captive or cultivated (e.g. "San Diego Wild Animal Park")

A full locality description may contain multiple clauses. The goal of a [georeference] is to describe the [location] where all of the clauses are true simultaneously. In GIS terms, this would be the intersection of the shapes for all the clauses in the locality description. As humans, we would choose the clause that is most specific and [georeference] based on that, using the information from the other clauses to filter from among multiple possibilities. For example, a locality written as

bridge over the St. Croix River, 4 km N of Somerset

should be georeferenced with a locality type "geographic feature only" with subtype {gqg}#feature-with-obvious-spatial-extent[Feature – with Obvious Spatial Extent^] as in {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^] based on the bridge as the [feature]. Of course, the second clause helps us to determine which bridge (something we wouldn’t be able to do without that second clause), but beyond that the second clause contributes nothing to the boundaries of the feature, nor to the [uncertainty] in the final georeference.

If the more specific part of the locality cannot be unambiguously identified, then the next less specific part of the locality ("4 km N of Somerset" in the example above) should be georeferenced. In a case such as this, annotate in the georeference remarks with something like "unable to find the bridge georeferenced '4 km N of Somerset'".

Some locality descriptions give information about the nature of the [offset] (‘by road’, ‘by river’, ‘by air’, ‘up the valley’, etc.). Having this information simplifies the choice of offset-based locality type as [Offset at a Heading] or [Offset along a Path].

Example 1. Classifying the locality description

country	AR
stateProvince	Neuquén
county	Los Lagos
locality	12.3 km N of (by road) Nahuel Huapi, elev: 760m

In this example, there are four fields contributing five separate clauses. The three administrative geography terms each have one clause of the type "Geographic feature only" with subtype "Feature – with obvious spatial extent" (see {gqg}#feature-with-obvious-spatial-extent[Feature – with Obvious Spatial Extent^] in {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^]), while the locality field contains a clause ("12.3 km N of (by road) Nahuel Huapi") of the type "Distance along path" (see {gqg}#offset-distance-along-a-path[Offset – Distance along a Path^] in {gqg}[Georeferencing Quick Reference Guide^]) and a clause ("elev: 760m") of the type "Geographic feature only" with subtype "Feature – Path" (see {gqg}#feature-path[Feature – Path^] in {gqg}[Georeferencing Quick Reference Guide^]). The most specific of all five clauses is "12.3 km N of (by road) Nahuel Huapi".

It is sometimes possible to infer the nature of the offset path from additional supporting evidence in the locality description. For example, the locality

58 km NW of Haines Junction, Kluane Lake

suggests a measurement by road since the final coordinates by that path are nearer to the lake than going 58 km NW in a straight line. At other times, you may have to consult detailed supplementary sources, such as field notes, collectors’ itineraries (see Using Collector Itineraries), diaries, or sequential collections made on the same day, to determine this information.

If any of the clauses in the locality description is classified as one of the three locality types, ‘dubious’, ‘cannot be located’, or ‘demonstrably inaccurate’, then the locality should not be georeferenced. Instead, an annotation should be made to the locality record giving the reason why it is not being georeferenced. See also {gqg}#difficult-localities[Difficult Localities^] in {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^].

Setting the Boundaries of the Feature

Regardless of the method to be used ([shape], [bounding-box], or [point-radius]), the georeferencing protocols for nearly every [locality type] begin with the identification of the features of reference in the [locality] description and the determination of the geographic boundaries of their extents. This is usually the most critical and time-consuming part of the protocols. It is best to use a visual reference to determine boundaries. If a feature name search on a visual source does not reveal the feature of interest, it is a good idea to use [coordinates] from a [gazetteer] to find the feature on a map, and then use the map to find the boundaries:

• Point-radius method: store the [corrected center] of the constrained boundaries from the previous step as decimal [latitude] and decimal [longitude] and store the [geographic radial] as a distance in the units given in the most specific [locality clause]. If there are no distance units in that clause, use meters (see Point-radius Method).

• Bounding Box method: store the furthest north, south, east, and west coordinates on the constrained boundaries of the feature (see Bounding Box Method).

• Shape method: store the resulting constrained boundaries as a shape (see Shape Method).

Use information from other clauses, such as administrative geography, information from other location fields such as [elevation], and environmental information (e.g. terrestrial, freshwater aquatic, marine, taxon-specific) to constrain the extent as appropriate (see Applying Spatial Constraints and Applying Data Constraints).

Applying Spatial Constraints

There are many ways that a [location] can be constrained beyond what the geography and [locality] descriptions alone suggest. Doing so relies on applying additional location information, such as [elevation] or [depth], lithostratigraphic information for fossils, or information outside the location information, such as environmental constraints for a particular species. There are important implications about workflow and effort that need to be considered when applying additional constraints. For example, if taxon constraints are going to be applied, the georeferencing cannot be done strictly on location information, which means it has to be done on occurrence records, or on an index combining location and taxon. This would be much slower than georeferencing based on location alone. A good compromise would be to georeference in multiple stages, with the first stage based on location information, and a subsequent stage including the rest of the occurrence information, and perhaps a final stage of review by collectors to be able to set term:dwc[dwc:georeferenceVerificationStatus] to "verified by collector" – the best status a georeference can possibly have.

Taxon Constraints

It is common to encounter [locality] descriptions for which the boundaries and [uncertainty] could be reduced if the taxon and its environmental or geographic constraints are known.

One case in which a taxon constraint might be applied is where a locality description would be georeferenced in a distinct manner if it was known to be terrestrial, aquatic, or marine. Here even the life stage of a taxon could be taken into account.

{marine}OBIS (the Ocean Biodiversity Information System) uses the World Register of Marine Species (WoRMS 2019) to determine if a species can be classified as either marine or terrestrial. Note, however, that there are many species listed in the WoRMS database that occur on coastal shores or in estuaries (i.e. species that could be regarded as both marine and terrestrial at some stage during their life cycle), so caution needs to be taken when using this method in georeferencing.

At the generic level there are similar biome-matching services available through the Interim Register of Marine and Nonmarine Genera (IRMNG) (Rees 2019), and the associated LifeWatch taxon matching services.

Another case where taxon might be taken into account is where a distribution range or environmental domain suggests a restriction in the boundaries of a location. However, this kind of constraint on a georeference is not recommended, because an organism whose location falls outside of an established range map may indicate a genuine outlier, or a taxon misidentification. Given that, such information can help distinguish between two possible locations of the same [feature] name where one possible location fits within the environmental domain for the taxon, and the other outside the range. This auxiliary information is also particularly useful after georeferencing, to reveal records of possible range extensions, exotic invasions, or cryptic taxa.

Using Date Constraints

The date is an important characteristic of an [event] and must be recorded. Towns, roads, counties, and even countries can change names and boundaries over time, and can even cease to exist as extant features. Rivers and coastlines can change position, billabongs and ox-bow lakes can come and go, and areas of once pristine environment may become farmland or urban areas.

Example 2. Date constraints

“Collecting localities along the Alaska Highway are frequently given in terms of milepost markers; however, the Alaska Highway is approximately 40 km shorter than it was in 1942 and road improvements continue to re-route and shorten it every year. Accurate location of a milepost, therefore, would require cross-referencing to the collecting date. To further complicate matters, Alaska uses historical mileposts (calibrated to 1942 distance), the Yukon uses historical mileposts converted to kilometers, and British Columbia uses actual mileage (expressed in kilometers).” From Wheeler et al. 2001

To the extent possible, the aim is to have a [georeference] and its uncertainties based on the conditions at the time an [event] occurred at a [locality]. There are two major implications associated with this. One is that current maps and gazetteers may not reflect the conditions at the time of the event, and the other is that old maps and gazetteers may not represent well the conditions of later events.

We recommend that this sort of constraint be used in a followup workflow step to deal with localities at the event level rather than try to construct a gazetteer that includes collecting dates.

Using Elevation or Depth Constraints

Elevation can often be used as a constraint to distinguish between two similarly named localities or to refine the uncertainty in a georeference. If both maximum and minimum elevations are given, then the contours of these limits may be used to constrain the [extent] of a locality and therefore its uncertainty. If a single value is given for elevation, then the precision of that value can be used to estimate the minimum and maximum elevations as described in §3.4.6 Uncertainty Related to Offset Precision. The {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^] describes how to georeference using elevation constraints in section {gqg}#feature-path[2.1.3.3. Feature – Path^]. The same considerations can also be applied to occurrence depths in cases of benthic organisms, or when the depth of the waterbody floor is available in non-benthic occurrence records, or to exclude geographic regions where waterbody depth is shallower than occurrence depth given.

Using Collector Itineraries

Collector’s itineraries and expedition tracks can be a useful adjunct in discovering locations that are otherwise difficult to find, especially where there may be more than one possible [location] based on a [feature] name. This may be done through using field notebooks, published reports and maps, searching for the localities of specimens with adjacent collecting numbers, etc. With historic collecting events (i.e. before the days of modern transport), you may also be able to restrict the area to look in by limiting the distance a collector may have been able to travel within one day. Note that the collector name and date are essential pieces of information in tracking itineraries, and therefore can not be done on localities alone. We thus recommend that this sort of constraint be used in a followup workflow step to deal with unresolved localities rather than try to construct a [gazetteer] that includes collecting dates, collector names, and collector numbers.

Using Ship Logs

{marine}Digitized ships logs contain a wealth of data (Dempsey 2014) and are valuable data resources. A freely downloadable database of surface marine observational records from ships, buoys, and other platform types is available as the International Comprehensive Ocean-Atmosphere Data Set (NOAA 2018). Be aware that the [accuracy] of records obtained from this dataset vary, depending on the original source, and are not always documented.

Using Geological Context

Maps or GIS layers of geological contexts, such as formations, can be used to narrow the [location] in the case of a paleontological specimen that includes such information in the shared content of the record. For example, if a fossil is taken from the surface in the Fox Hills formation (which is Cretaceous in age), that can distinguish the location from nearby different formations on the surface, like a habitat could do in an ecological context.

Georeferencing Methods

The distinction between georeferencing methods is in the basic approach taken to capture spatially enabled [location] data. Within each method there should be protocols for how to produce georeferences based on the input [locality] description and supporting information. The goal of any georeferencing method and its specific, documented protocols should be to create a spatial representation of the entire location, including all uncertainties involved, with sufficient accompanying information and documentation to make the georeference reproducible.

Point Method

Based on the aspirations for georeferencing methods described in the previous paragraph, the point method, consisting of only [coordinates], or coordinates in a [coordinate reference system], is insufficient to be useful except to center a point on a map (and even that potentially incorrectly without the coordinate reference system). The point method does not give any indication of scale, though the mistake is often made to try to represent scale and/or uncertainties in the [precision] of the coordinates. For these reasons, the point method is NOT recommended as the end product of a georeferencing workflow.

Point-radius Method

The result of the [point-radius] method (Wieczorek et al. 2004) is a geographic coordinate (the "[corrected center]"), its [geodetic datum], and a [maximum uncertainty distance] as a radius. The length of the radius must be large enough so that a circle centered on the corrected center and based on that radius encompasses all of the uncertainties in the interpretation of the [location]. The point-radius is a very simple representation of the location that contains all of the places that the [locality] description might refer to, but may also circumscribe areas that do not match the locality description. That’s OK. The point-radius circle can also be intersected with other spatially enabled information to constrain the effective area within the circle, such as [elevation], to derive a [shape] representation of the [locality]. For example, calculate the intersection of a point-radius circle with the shape of the matching elevation contours in a [geographic information system] to get a shape that better matches the described locality. Similarly, one could calculate the intersection of an exposed geological formation with a point-radius [georeference] to refine the latter into a shape. The detailed recommended protocols for georeferencing using the point-radius method are given in the {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^].

Bounding Box Method

The result of the [bounding-box] method (Wieczorek et al. 2004) is a set of two [coordinates], one for each of two corners diagonally opposed on the bounding box along with their [coordinate reference system]. The corners define the minimum and maximum values of the coordinates, within which the whole of the [location] and its uncertainties is contained. Like the [point-radius] method, the bounding box method results in a very simple representation of the [location] that contains all of the places that the [locality] description might refer to, but may also contain areas that do not match the locality description.

Unlike the [point-radius] method, this method has no scalar [maximum uncertainty distance] to be able to easily understand or filter on the size of the enclosed region, though one can be calculated using half the distance between the two corners as given by Vincenty’s formulae (Vincenty 1975, Vincenty 1976). Thus, a bounding box [georeference] can be turned into a point-radius georeference by using the distance just described as the [geographic radial], and from that finding the [corrected center], which will not be equal to the [geographic center] of the bounding box, except where the bounding box spans equal distances north and south of the equator or is based on a metric [grid].

A point-radius georeference can be turned into a bounding box georeference by using the geographic radial from the corrected center of the point-radius to determine the coordinates of the east-west and north-south extremes of the bounding box.

Note

Though transformations can be made back and forth between point-radius and bounding box representations of a location, it is not recommended, because the transformed georeference will necessarily be bigger than the original, and therefore contain more area that does not pertain to the actual location. Better to georeference directly using the method of choice.

Like the point-radius circle, the bounding box can also be intersected with other spatially enabled information to constrain the effective area within.

Shape Method

The [shape] method (also called the polygon method by some (Yost 2015)) of determining [uncertainty] is a conceptually simple method that delineates a [locality] using geometries with one or more polygons, buffered points, or buffered polylines. A combination of these shapes can represent a town, park, river, junction, or any other [feature] or combination of features found on a map. While simple to describe, the task of generating these shapes must account for all the uncertainties, and that can be difficult. Except for the simplest locality types, creating shapes is impractical without the aid of digital maps, GIS software (for buffering, clipping, etc.), and expertise, all of which can be relatively expensive. Also, except for a [bounding-box], which is an extremely simple example, storing a shape in a database can be considerably more complicated than storing a single pair of [coordinates] with a scalar uncertainty distance as in the [point-radius] method. [Darwin Core] (Wieczorek et al. 2012b) offers the field term:dwc[dwc:footprintWKT], in which a [geometry] can be stored in the Well-Known Text format (ISO 2016) accompanied by the [coordinate reference system] in the field term:dwc[dwc:footprintSRS]. Particular challenges to making this method practical for georeferencing natural history collections data include assembling freely accessible digital cartographic resources and developing tools for automation of the georeferencing process (Yost n.d.). This is because, not only does the geometry of the feature usually need to be created (unless it is an administrative [boundary] or other shape available in a spatial data layer), but also all the points in the feature geometry have to be used in combination with the uncertainties to arrive at a final shape that includes the location with its uncertainties and nothing more. Note that GEOLocate (Rios 2019) does produce an "error polygon" (Biedron & Famoso 2016) in addition to a point-radius, but how this is done is not documented in detail.

Of all the methods discussed in this document, the shape method has the potential to generate the most specific digital spatial descriptions of localities, leaving out areas that are not viable as part of the location. A point-radius can be easily derived from a final shape by using the [corrected center] for the coordinates and the [geographic radial] of the georeference (not just the feature) for the [maximum uncertainty distance]. See Example of using spatial fit on the results of both a point-radius method and a refined shape method of describing uncertainty. Assuming the blue-shaded area is the "true" locality as we know the species is terrestrial, and the red circle is the point-radius method of representing the uncertainty, the ratio of the area of the red circle (5.726 sq km) divided by the area of the blue shaded area (~4.1 sq km) gives a spatial fit for the point-radius of 1.39. for one example of where a point-radius may be refined by using the shape method. See also [Polygons].

Probabilistic Method

Other shape-based methods have been proposed that use probabilistic approaches (Guo et al. 2008, Liu et al. 2009). Since these methods are even more difficult than the [shape] method, and there are currently no tools available to take advantage of these methods, we do not discuss them further in this document.

Calculating Uncertainties

Regardless of the method, uncertainties in georeferenced data are essential to document, so that the data’s fitness for use and thus their overall [data quality] can be understood. There are sources of uncertainty in each [locality] interpretation as well as in the data sources used to georeference, and any physical measurement that might need to be made (such as on maps, digital or physical). Each of the sources of uncertainty have to be taken into account to capture the overall uncertainty in a resulting georeference.

Whenever subjectivity is involved, it is preferable to overestimate each contribution to uncertainty. The following seven sources of uncertainty are the most commonly encountered. These are explained below and can be accounted for by using the Georeferencing Calculator (Wieczorek & Wieczorek 2020).

• Uncertainty due to the [extent] of the [feature] in the [locality] description.

• Uncertainty in coordinate source.

• Uncertainty in map measurements.

• Uncertainty related to [coordinate precision].

• Uncertainty from unknown [coordinate reference system] or [datum].

• Uncertainty related to [heading].

• Uncertainty related to [offset] [precision].

Uncertainty Due to the Extent of the Feature

The first step in determining the [coordinates] for a [locality] description is to identify the most specific [feature] within the locality description. Coordinates may be retrieved from gazetteers, geographic name databases, maps, or from other locality descriptions that have coordinates or shapes. We use the term ‘[feature]’ to refer to not only traditional named places, but also to places that may not have proper names, such as road junctions, stream confluences, highway mile pegs, and cells in grid systems (e.g. Quarter Degree Square Cells, see [Quarter Degree Squares]). The source and [precision] of the coordinates should be recorded so that the validity of the georeferenced locality can be checked. The original [coordinate system] and the [geodetic datum] should also be recorded. This information helps to determine sources and the [maximum uncertainty distance], especially with respect to the original [coordinate precision].

How do we take into account the uncertainty due to the shape of the feature? The method that results in the least uncertainty is to find the [smallest enclosing circle] (Matoušek et al. 1996) that contains all of the points on the [geographic boundary] of the feature. If the center of the circle does not fall on or within the [boundary] of the feature, choose the point nearest to the center that is on the boundary. This is known as the [corrected center]. The distance from the corrected center to the farthest point on the geographic boundary of the feature is called the [geographic radial]. The geographic radial is the uncertainty due to the [extent] of the feature (see [img-polygon-center]).

Every feature occupies a finite space, or ‘extent’. The extents of features are an important source of uncertainty. Points of reference for features may change over time – post offices and courthouses are relocated, towns change in size, the courses of rivers change, etc. Moreover, there is no guarantee that the person who recorded the locality information paid attention to any specific convention when reporting a locality as an [offset] from a feature. For example,

4 km E of Bariloche, Argentina

may have been measured from the post office, the civic plaza, or from the bus station on the eastern side of the heavily populated part of town, or anywhere else in Bariloche, which is actually quite large. When calculating an offset, we generally have no way of knowing where the person who recorded the locality started to measure the distance. The determination of the boundaries of a feature are discussed in Setting the Boundaries of the Feature.

It is also worth noting that the extent of a feature may have changed over time, so the date of the recording may also be important when calculating an extent and thus the geographic radial. In many cases (especially for populated places), the current extent of a feature will be greater than its historical extent and the uncertainty may be somewhat overestimated if current maps are used.

If the locality described is an irregular shape (e.g. a winding road or river), there are two ways of calculating the "center" coordinates and determining the [radial]. The first is to measure along the vector (line) and determine the midpoint as the [location] of the feature. This is not always easy, so the second method is to determine the [geographic center] (i.e. the midpoint of the extremes of [latitude] and [longitude]) of the feature. This method describes a point where the uncertainty due to the extent of the feature is minimized (what we are calling the [corrected center]). The radial is then determined as the distance from the determined position to the furthest point at the extremes of the vector. If the geographic center of the shape is used and it does not lie within the locality described (e.g. the geographic center of a segment of a river does not actually lie on the river), then the point nearest the geographic center that lies within the shape (corrected center) is the preferred reference for the feature and represents the point from which the geographic radial should be calculated (see [img-polygon-center]).

When documenting the georeferencing process, it is recommended that the feature, its extent, radial, and the source of the information (including its date) all be recorded. For details on georeferencing, see {gqg}#geographic-feature-only[Geographic Feature Only^] in {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^].

Geographic coordinates can be expressed in a number of different coordinate formats. Decimal degrees provide the most convenient coordinates to use for georeferencing for no more profound reason than a locality can be described with only four attributes – [decimal latitude], [decimal longitude], [datum], and uncertainty (Wieczorek 2001).

Uncertainty in Coordinate Source

There are many ways of finding [coordinates] for a [location], including using a [gazetteer], a GPS, aerial photogrammetry, digital maps, or paper maps of many different types, and scales.

Uncertainty in Paper Map Measurements

One of the most common methods of finding [coordinates] for a [location] is to estimate the location from a paper map. Using paper maps can be problematic and subject to varying degrees of inaccuracy. Unfortunately, the [accuracy] of many maps, particularly old ones, is undocumented. Accuracy standards generally explain the physical [error] tolerance on a printed map, so that the net uncertainty is dependent on the map scale (see Horizontal accuracy based on 0.5 mm of accuracy per unit of map scale, except for the 1:250,000 map series where the figure supplied with the data has been used.).

Map reading requires a certain level of skill in order to determine coordinates accurately, and different types of maps require different skills. Challenges arise due to the [coordinate system] of the map ([latitude] and [longitude], [UTM], etc.), the scale of the paper map, the line widths used to draw the features on the maps, the frequency of [grid] lines, etc.

The accuracy of a map depends on the accuracy of the original data used to compile the map, how accurately these source data have been transferred onto the map, and the resolution at which the map is printed or displayed. For example, USGS maps of 1:24,000 and 1:100,000 are different products. The accuracy is explicitly dependent on scale but is due to the different methods of preparation. When using a map, the user must take into account the limitations encountered by the map maker such as acuity of vision, lithographic processes, plotting methodologies, and symbolization of features (e.g. line widths) (Hardy & Field 2012).

With paper topographic maps, drawing constraints may restrict the accuracy with which lines are placed on the map. A 0.5 mm wide line depicting a road on a 1:250,000 map represents 125 meters on the ground. To depict a railway running beside the road, a separation of 1-2 mm (250-500 meters) is needed, and then the line for the railway (another 0.5 mm or 125 meters) makes a total of 500-750 m as a minimum representation. If one uses such features to determine an occurrence locality, for example, then minimum uncertainty would be in the order of 1 km. If thicker lines were used, then appropriate adjustments would need to be made (Chapman et al. 2005).

The National Standard for Spatial Data Accuracy (NSSDA) (FGDC 1998) established a standard methodology for calculating the horizontal and vertical accuracy of printed maps, which state that 95% of all points must fall within a specified tolerance (1/30" for map scales larger than 1:20,000, and 1/50" for map scales smaller than or equal to 1:20,000).

Horizontal accuracy based on 0.5 mm of accuracy per unit of map scale, except for the 1:250,000 map series where the figure supplied with the data has been used. shows the inherent accuracy of a number of maps at different scales. The table gives uncertainties for a line 0.5 mm wide at a number of different map scales. A value of 1 mm of error can be used on maps for which the standards are not published. This corresponds to about three times the detectable graphical error and should serve well as an uncertainty estimate for most maps.

The table uses data from several sources. The TOPO250K Map series is the finest resolution mapping that covers the whole of the Australian continent. It is based on 1:250,000 topographic data, for which Geoscience Australia 2007, Section 2 defines the accuracy as "not more than 10% of well-defined features are in error by more than 140 meters (for 1:250,000 scale maps); more than 56 meters (for 1:100,000 maps)". The USGS Map Horizontal Uncertainty is calculated from US Bureau of Budget (1947) (reported in United States National Map Accuracy Standards (USGS 1999)) which states that "As applied to the USGS 7.5-minute quadrangle topographic map, the horizontal accuracy standard requires that the positions of 90 percent of all points tested must be accurate within 1/50th of an inch (0.05 centimeters) on the map. At 1:24,000 scale, 1/50th of an inch is 40 feet (12.2 meters)." These values need to be taken into account when determining the uncertainty of your georeference.

Table 1. Horizontal accuracy based on 0.5 mm of accuracy per unit of map scale, except for the 1:250,000 map series where the figure supplied with the data has been used.

Scale of Map	Map Horizontal Accuracy (Geoscience Australia)	Map Horizontal Accuracy (USGS)	NSSDA Horizontal Accuracy (FGDC 1998)
1:1000	0.5 m	2.8 ft (0.85 m)	3.2 ft (1 m)
1:10,000	5 m	28 ft (8.5 m)	32 ft (10 m)
1:25,000	12.5 m	70 ft (21 m)	47.5 ft (14.5 m)
1:50,000	25 m	139 ft (42 m)	95 ft (29 m)
1:75,000			142.5 ft (43.5 m)
1:100,000	50 m	278 ft (85 m)	190 ft (58 m)
1:250,000	160-300 m	695 ft (210 m)	475 ft (145 m)
1:500,000			950 ft (290 m)
1:1 million	500 m	2,777 ft (845 m)	1,900 ft (580 m)

If you are using phenomena that do not have distinct boundaries in nature to determine a locality (such as soils, vegetation, geology, timberlines, etc.) then err vastly on the side of conservatism when determining an uncertainty value as such boundaries are seldom accurate, often determined at a scale of 1:1 million or worse and would have a minimum uncertainty of between 1 and 5 km. Also be aware that coastlines vary greatly at different scales (see Chapman et al. 2005) and rivers are often straightened on smaller scale maps, and can thus include uncertainties far greater than are generally recorded on maps whose accuracies are determined from "well-defined" points such as buildings, road intersections, etc. In addition, coastlines and river paths can change greatly over time (World Ocean Review 2010) and thus the date of the map needs to be taken into account when determining uncertainty.

In addition to the inherent inaccuracies of printed maps, one must consider inaccuracies that can arise from using maps to measure distances. These potential inaccuracies are a direct consequence of the projection of the map and one’s ability to distinguish between two adjacent points, which may be affected by your measuring device and even your eyesight. A straight line distance measurement only works on a map in an equal distance projection, where distance follows the same scale regardless of the orientation. Unless the conditions for measuring are particularly poor, it is reasonable to use 1 mm as a value for measurement error on physical maps. Depending on the scale of the map, this translates into a distance on the ground.

Uncertainty in Digital Map Measurements

Digital versions of traditional paper maps that have been scanned or digitized by hand using a digitizing tablet to trace lines, have an extra layer of [uncertainty] (Dempsey 2017). Depending on how the map was digitized, the [error] may be small or large when compared to the scale of the original map. In parts of the world where digitized maps are not readily available, they can be scanned and rectified using satellite data (Raes et al. 2009). Scanned maps often (and should always) include information on the [accuracy] added by the digitizing process (see ASPRS 1990). Be careful when using digital maps, and record any information on the scanning accuracy if that information is available. Always err on the cautious side when recording the uncertainty of your [georeference] when using maps of this type (ASPRS 2014).

Note	A digital map is never more accurate than the original from which it was derived, nor is it more accurate when you zoom in on it. The accuracy is strictly a function of the scale and digitizing errors of the original map, plus the additional error added by the digitization process.

Caution

Care must be used when using a digital map that records the scale in the form of text (e.g. 1:100,000) rather than by using a scale bar, as the resolution of the computer screen, and the level of zooming will change the apparent scale of the map being viewed. (It does not change the scale at which the map was prepared). This also applies to maps printed from a digital map. When preparing digital maps, always include scale as a scale bar and do not just record scale in textual form (e.g. 1:20,000).

Measurement error is not unique to physical maps, it also enters into measurements on digital media. In general, the resolution of the media affects one’s ability to distinguish between two points, and this in turn can be affected by the extent to which the media is zoomed. Note that zooming does not improve the accuracy of the original source from which the media was derived. That accuracy remains an independent factor, as described in the earlier paragraphs in this section. Naturally, the greater the zoom, the easier it is to pinpoint a [location]. This effect of zoom on digital media also has an effect on one’s ability to measure along a [path] in that medium. The greater the zoom, the easier it is to follow the path faithfully and thus determine a distance along that path with the least error. The greater the curviness of the path, the greater the potential effect on accuracy. Note also, that the scale of the map may reduce the curviness of a path (road, river, etc.) and that small-scale maps tend to smooth out the paths of rivers, roads, coastlines, and other curved linear features (Chapman et al. 2005).

Using OpenStreetMap, Google Maps and Google Earth

With the ever increasing availability of high-quality satellite imagery and shapes for geographic features, online digital map resources are increasingly being used to find features and their boundaries, and to [georeference]. Some sites have tools that are particularly suited for drawing and measuring on maps. In Google Maps, for example, the measuring tool can be initiated by clicking at your starting point or origin, then using right-click to select Measure distance from a pop-up menu. You can then click on your end point and a line segment with distance indicators will join the two chosen locations. You can click repeatedly to trace a [path], such as along a road or river. You can also close the shape to make a polygon by clicking on the starting point again. Once you have your line or polygon, you can modify the node positions (for example after zooming in further), and add intermediate nodes. It can also be used to determine distance from a point, such as "5 km N of [feature]". By closing the polygon, you can get an area as well as total distance. Determine [uncertainty] as you would for any other map, but be aware of the effects of the level at which you may be zoomed in. One’s capacity to point accurately is higher at higher zoom levels. One can test the effect empirically by trying repeatedly to put a marker on the center of a feature that can be seen at low zoom levels, then checking how far off they are on average at higher zoom levels.

The positional error on Google Maps and Google Earth is poorly documented and varies both geographically and with imagery resolution. We recommend the conservative combination of root mean square error from Google Earth and Landsat imagery of 89.7m estimate derived by Potere 2008 for Google Earth or Google Map readings in or before 2008. After that, we recommend the 8m (95 per cent confidence interval) estimated by Paredes-Hernández et al. 2013. Limited data based on the [accuracy] of street junctions on OpenStreetMap (Helbich et al. 2012) suggests that this source has accuracy of the same order of magnitude as the Google products. Note that measurements in Google Earth and Google Maps are direct lines and don’t account for changes in elevation.

Elevation coverage from Google Maps is inconsistent, it can be obtained by reading the contour lines in mountainous areas in the Terrain view, but it does not show elevation by default and not in cities or areas where there are no natural elevation gradients. In Google Earth one can access elevation information everywhere and it is visible with the [latitude] and [longitude] in the lower right of the view screen. Elevation in Google Earth is based on the [mean-sea-level] model of the EGM96 [geoid]. Note that this can vary by up to 200 meters from the [WGS84] reference [ellipsoid] in some areas (see [img-mean-sea-level-wgs84-ellipsoid]). As noted under [Google Earth], we recommend using the values extracted from the work of Wang et al. 2017 as estimates of elevational uncertainty when the source is the Google Earth terrain model.

Uncertainties in Marine Maps

Harbor charts are generally produced at a scale of 1:10,000, and coastal charts at 1:50,000 to 1:150,000, and often in the Mercator projection. A page on Navigation – finding [location] on nautical maps can be seen at Coastal Navigation 2020. A majority of new maps (post-2019) are only being produced digitally (NOAA 2020, personal communication, 25 Jan), with paper maps being produced from the digital product.

For most marine or nautical charts, the [accuracy] and reliability of the information used to compile the chart is recorded as Zones of Confidence (ZOC) (Prince 2020). ZOC categories warn mariners which parts of the chart are based on good or poor information and which areas should be navigated with caution. The ZOC system consists of five categories for assessed [data quality], with a sixth category for data which has not been assessed (Marine mapping Zones of Confidence (ZOC) categories and their associated accuracy. Derived with permission from AHP20 (Australian Hydrographic Office 2020) and NOAA 2016.).

Positional accuracy refers to the horizontal accuracy of a [depth] or [feature]. Depth accuracy refers to the vertical accuracy of individual recorded depths, of which those shown on the chart are a subset designed to best represent the sea floor as it is known or estimated.

Table 2. Marine mapping Zones of Confidence (ZOC) categories and their associated accuracy. Derived with permission from AHP20 (Australian Hydrographic Office 2020) and NOAA 2016.

ZOC	Positional Accuracy	Depth Accuracy	Seafloor Coverage
A1	± 5m (16 ft)	=0.50m (1.6 ft) + 1% depth	All significant seafloor features detected.
A2	± 20m (66 ft)	=1.0m (3.2 ft) + 2% depth	All significant seafloor features detected.
B	± 50m (160 ft)	=1.0m (3.2 ft) + 2% depth	Uncharted features hazardous to surface navigation are not expected but may exist.
C	± 500m (1600 ft)	=2.0m (6.5 ft) + 5% depth	Depth anomalies may be expected.
D	Worse than ZOC C	Worse than ZOC C	Large depth anomalies may be expected.
U	Unassessed. The quality of bathymetric data has yet to be assessed.

Uncertainty due to GPS

The uncertainties inherent in various Global Navigation Satellite Systems and [GPS]/GNSS devices are discussed in detail in Section [GPS Accuracy]. The most common way of getting [coordinates] in the field is from a GNSS-enabled device, which includes most smartphones. Most user interfaces on hand-held GPS/GNSS devices and applications on smartphones show a "GPS Accuracy". The figure shown as "Accuracy" isn’t true [accuracy]. It is the EPE (Estimated Position Error) (Herries 2012). In other words, it is the probability that the location the GPS is displaying is within the "accuracy" distance from the true location. Keep in mind that a GPS receiver doesn’t actually know its true location. It calculates a location, based on the data received from the satellites. However, if the instrument has a [bias], it still may give a low reported "Accuracy" (i.e. the repeated measurements may be close together) but they may be some distance from the true location (see [img-accuracy-vs-precision]). While most GPS manufacturers don’t tell you how they calculate "accuracy", you can consider it a figure that says "most of the time, the displayed location coordinates are within X distance of the GPS receiver" (where X is the "accuracy" figure).

The "Accuracy" value is affected by the current satellite configuration (the number of satellites that are visible and their positions in the sky (satellite ephemeris)), and a vast host of environmental variables between the device and the satellites that affect the signal trajectories and signal-to-noise ratios. Without access to a [SBAS] (see [Satellite Based Augmentation System]), this value can be used only as an indicator of relative accuracy, but it is statistically always less than the real value. This is easy to demonstrate with sufficient repeated measurements of coordinates and purported accuracy at the same well-known location over time. The mean accuracy value will be less than the mean distance shift between the mean coordinate given by all readings (a statistical proxy for the true coordinates) and the individual coordinate readings. Herries 2012 recommends doubling the Accuracy (EPE) reported by the GPS Receiver (including smartphones) to get a more realistic representation of true accuracy.

In summary, the EPE (‘accuracy’ given on a GPS) is not a maximum uncertainty, but an equal (50 per cent) chance that your position lies with a radius of that value. To get a 95 per cent confidence level that your measurement is within a circle of a fixed radius, you have to multiply the EPE value by two as an absolute minimum. For details on georeferencing GPS coordinates see [GPS Accuracy], and {gqg}#coordinates-geographic-coordinates[Coordinates – Geographic Coordinates^] in the {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^].

Uncertainty due to using previously georeferenced localities

Using previously georeferenced localities – whether from your own database, or from an external source can introduce uncertainties. If the source is previously georeferenced localities from your own database, then it is important that you retain all the metadata associated with that previously georeferenced locality with all subsequent records. Similarly, if using an external source, try and record a DOI reference or similar if possible, so that any subsequent changes can be traced.

Note	When using previously georeferenced localities as a source, if an [error] was made with the original georeferencing, then it will be perpetuated through all subsequent georeferences.

Uncertainty Related to Coordinate Precision

Geographic coordinates should always be recorded using as many digits as possible; the [precision] of the [coordinates] should be captured separately from the coordinates themselves, preferably as a distance, which conserves its meaning regardless of [location] and coordinate transformations. Recording coordinates with insufficient [precision] can result in unnecessary uncertainties. The magnitude of the uncertainty is a function of not only the precision with which the data are recorded, but also of the [datum] and the coordinates themselves. This is a direct result of the fact that a degree does not correspond to the same distance everywhere on the surface of the earth.

Table showing metric uncertainty due to precision of coordinates based on the WGS84 datum at varying latitudes. Uncertainty values have been rounded up in all cases. From Wieczorek 2001. shows examples of the contributions to uncertainty for different levels of precision in coordinates using the [WGS84] reference [ellipsoid]. Calculations are based on the same degree of imprecision in both coordinates and are given for several different latitudes. Approximate calculations can be made based on this table, however, more accurate calculations can be obtained using the Georeferencing Calculator (Wieczorek & Wieczorek 2020) – see further discussion below.

From Table showing metric uncertainty due to precision of coordinates based on the WGS84 datum at varying latitudes. Uncertainty values have been rounded up in all cases. From Wieczorek 2001., it can be seen that an observation recorded in degrees, minutes, and seconds ([DMS]) has a minimum [uncertainty] of between 32 and 44 metres.

Table 3. Table showing metric uncertainty due to precision of coordinates based on the WGS84 datum at varying latitudes. Uncertainty values have been rounded up in all cases. From Wieczorek 2001.

Precision	0 degrees Latitude	30 degrees Latitude	60 degrees Latitude	85 degrees Latitude
1.0 degree	156,904 m	146,962 m	124,605 m	112,109 m
0.1 degree	15,691 m	14,697 m	12,461 m	11,211 m
0.01 degree	1,570 m	1,470 m	1,246 m	1,121 m
0.001 degree	157 m	147 m	125 m	112 m
0.0001 degree	16 m	15 m	13 m	12 m
0.00001 degree	2 m	2 m	2 m	2 m
1.0 minute	2,615 m	2,450 m	2,077 m	1,869 m
0.1 minute	262 m	245 m	208 m	187 m
0.01 minute	27 m	25 m	21 m	19 m
0.001 minute	3 m	3 m	3 m	2 m
1.0 second	44 m	41 m	35 m	32 m
0.1 second	5 m	5 m	4 m	4 m
0.01 second	1 m	1 m	1 m	1 m

Caution

False precision can arise when transformations from degrees minutes seconds to [decimal degrees] are stored in a database (see Glossary for expanded discussion).

Caution

Never use [precision] in a database as a surrogate for the coordinate uncertainty; instead, record the uncertainty explicitly, preferably as a distance.

Note	Details of calculations used to determine uncertainties in coordinate precisions can be found in Wieczorek 2001 and Wieczorek et al. 2004.

Example 3. Coordinate precision

Lat: 10.27° Long: −123.6° Datum: WGS84

In this example, the lat/long precision is 0.01 degrees. Thus, latitude error = 1.1061 km, [longitude] error = 1.0955 km, and the uncertainty resulting from the combination of the two is 1.5568 km.

Lat: 10.00000° Long: −123.50000° Datum: WGS84

In this example, the lat/long precision is 0.5 degrees because neither coordinate demonstrates more specificity than that. Thus, latitude error = 55.6 km, longitude error = 54.75 km, and the uncertainty resulting from the combination of the two is 77.87 km.

Uncertainty from Unknown Datum

It is important to record the [datum] used for the coordinate source ([GPS], map sheet, [gazetteer]) if it is known, or to record the fact that it is not known. Coordinates without a [coordinate reference system] are ambiguous. Geographic coordinates with a datum constitute a coordinate reference system (see [Coordinate Reference System]), but seldom do natural history collections have complete coordinate reference system information. Even with a [GPS] being used to record coordinates in the field, the [geodetic datum] is typically ignored.

The ambiguity from a missing datum varies geographically and adds greatly to the [error] inherent in the georeferencing. Differences between datums may cause an error in true [location] from a few centimeters up to kilometers (Wieczorek 2019). Note that the difference between datums is not a simple function that can be calculated on the fly. The values have to be pre-calculated comparing all datums to a reference datum of choice (e.g. [WGS84]) at every point of interest over the earth’s surface and stored in a way that can be looked up by geographic coordinates. The Georeferencing Calculator (Wieczorek & Wieczorek 2020) is capable of doing such a lookup (see Using the Georeferencing Calculator). In the absence of looking up the actual value by coordinates, the worst case scenario of 5359 m (Wieczorek 2019) can be used.

Uncertainty Related to Heading

The calculation of [uncertainty] from the [precision] in which a [direction] is recorded depends on the distance from the starting reference [feature]. The uncertainty will increase with increasing distance from the source. For simple determinations of angular precision due to direction – see Calculating uncertainty using the precision of the recorded direction (derived from Wieczorek et al. 2004)..

Note	The uncertainty due to directional imprecision increases with distance, so it can only be calculated from the combination of distance and direction (see below).

Table 4. Calculating uncertainty using the precision of the recorded direction (derived from Wieczorek et al. 2004).

Precision	Interpretation	Example	Heading Uncertainty
N	Between NW and NE	10.6 km N of Lambert Centre	45°
NE	Between NNE and ENE	10.5 mi NE of Lambert Centre	22.5°
NNE	Between N of NNE and E of NNE	10 km NNE of Lambert Centre	11.25°

Figure 1. Diagram showing directional precision for the interpretation of NE between ENE and NNE. Uncertainty (x and y) increases with distance from the feature

Using the example

10 km NE of Lambert Centre

and if we ignore distance imprecision, uncertainty due to the direction imprecision (Diagram showing directional precision for the interpretation of NE between ENE and NNE. Uncertainty (x and y) increases with distance from the feature) is encompassed by an arc centered 10 km (d) from the center of Lambert Centre (at x,y) at a heading of 45 degrees (θ), extending 22.5 degrees in either direction from that point. At this scale the distance (e) from the center of the arc to the furthest extent of the arc (at x′,y′) at a heading of 22.5 degrees (θ′) from the center of Lambert Centre can be approximated by the Pythagorean Theorem,

\$e = sqrt( (x′-x)^2 + (y′-y)^2)\$

where x=dcos(θ), y=dsin(θ), x′=dcos(θ′), and y′=dsin(θ′). The uncertainty in the above example would be 3.90 km.

This shows just one simple example. For details and formulae for calculating more complicated uncertainties, see Wieczorek 2001 and Wieczorek et al. 2004. Because of the complicated nature of these calculations, it is best to use the Georeferencing Calculator (Wieczorek & Wieczorek 2020) – see Using the Georeferencing Calculator.

Uncertainty Related to Offset Precision

Precision can be difficult to gauge from a [locality] description as it is seldom, if ever, explicitly recorded. Further, a database record may not reflect, or may reflect incorrectly, the precision inherent in the original measurements, especially if the locality description in the database has undergone normalization, reformatting, or secondary interpretation of the original locality description.

There are a number of ways of calculating uncertainty from distances. In this document, we recommend a conservative approach, which assumes that many records have undergone a certain amount of interpretation or transformation when being entered into the database. Thus, a record of "10¼ mi" may be entered into the database as 10.25 mi. The precision implied in the value 10.25 is thus a [false precision] and the real precision should not be assumed to be between 10.24 and 10.26 or between 10.2 and 10.3. The method of Wieczorek et al. 2004, adapted here, bases the estimate of uncertainty on the fractional part of the distance, calculated by dividing 1 by the fractional denominator. The uncertainty would just be half of the precision. For example, 10.5 mi N of Bakersfield could reasonably be expected to mean 10½ mi with a precision of half a mile between 10.25 and 10.75 mi, or 10.5 with an uncertainty of 0.25 mi.

For distance measurements that are positive integer powers of 10, the precision should be ten to the next lower power. This calculation differs from Wieczorek et al. 2004, which recommended that the precision should be based on ten to the same power. Upon reconsideration, that seems excessive (see Calculating uncertainty related to the [precision] of a distance measurement. The table shows examples of distance measurements, the recommended uncertainty due to the precision in the example adapted from Wieczorek et al. 2004, and a comparison to the rules applied for uncertainty by Frazier et al. 2004).). This same reasoning can be used for precision in verbatim elevations and depths. Recommended values for uncertainty related to offset precisions are shown in Calculating uncertainty related to the [precision] of a distance measurement. The table shows examples of distance measurements, the recommended uncertainty due to the precision in the example adapted from Wieczorek et al. 2004, and a comparison to the rules applied for uncertainty by Frazier et al. 2004)..

Table 5. Calculating uncertainty related to the [precision] of a distance measurement. The table shows examples of distance measurements, the recommended uncertainty due to the precision in the example adapted from Wieczorek et al. 2004, and a comparison to the rules applied for uncertainty by Frazier et al. 2004).

Distance	Recommended Uncertainty	Uncertainty sec. Frazier et al.
10.1 km	0.05 km	0.1 km
10.25 mi	0.125 mi	0.01 mi
10.5 km	0.25 km	0.1 km
10.6 mi	0.05 mi	0.1 mi
10.75 km	0.125 km	0.01 km
10 mi	0.5 mi	1.5 mi
15 km	0.5 km	1 km
30 mi	0.5 mi	4.5 mi
33 km	0.5 km	1 km
100 mi	5 mi	15 mi
140 km	5 km	21 km
200 mi	5 mi	30 mi
1000 m	50 m	150 m
2000 m	50 m	300 m

Precision can also be masked or lost when measurements are converted, such as from feet to meters, or from miles to kilometers.

Caution

Be careful that the value you are using for precision when calculating the uncertainty is a true precision and not a false precision. For example, converting a collector’s recording of 16 miles (with a precision of 1 mile) to 25.6 km (with a precision of 0.1 km) leads to an unwarranted level of precision that is more than 16 times higher than the original.

Example of a locality b as offsets x and y in orthogonal directions (from the corrected center a of a feature (i.e. stock watering point). The coordinates b (8 km E and 6 km N of a are surrounded by a bounding box 1 km square c showing the uncertainty due to distance precision of 1 km. The net uncertainty from distance precision is represented by a circle d that circumscribes the bounding box and which has a radial of 0.707 km. By convention the headings for localities with offsets in orthogonal directions are exactly in the specified directions and contribute no uncertainty due to direction precision. shows an example of two orthogonal distances measured from a [feature], each with the uncertainty due to distance precision. If we ignore all sources of uncertainty except those arising from distance precision, the uncertainty is a bounding box centered on the point 8 km E and 6 km N of the [corrected center] of the feature. Each of the distance measurements demonstrates a precision of 1 km. Thus, each side of the box is a total of 1 km in length (0.5 km uncertainty in each cardinal direction from the center). Since we are characterizing the precision as a single distance measurement (1 km), we need the circle that circumscribes the above-mentioned bounding box to get the uncertainty due to the combined distance precisions. The radius of this circle is half the length of the distance precision bounding box, which is equal to one half the square root of two times the distance precision. So, for the above example the uncertainty associated with only the distance precision is one half the square root of two, or 0.707 km.

Figure 2. Example of a locality b as offsets x and y in orthogonal directions (from the corrected center a of a feature (i.e. stock watering point). The coordinates b (8 km E and 6 km N of a are surrounded by a bounding box 1 km square c showing the uncertainty due to distance precision of 1 km. The net uncertainty from distance precision is represented by a circle d that circumscribes the bounding box and which has a radial of 0.707 km. By convention the headings for localities with offsets in orthogonal directions are exactly in the specified directions and contribute no uncertainty due to direction precision.

Combined Uncertainties

When combining uncertainties from different sources, it is not as simple as taking the average or adding them together. Uncertainties inherent in the [location] of the [feature], in its [extent], in the direction of the [offset], and the distance of the offset, are just four sources that need to be combined to get an overall uncertainty. A detailed discussion of the calculations involved can be found in Wieczorek 2001 and Wieczorek et al. 2004. For a practical way of calculating uncertainties in [locality] descriptions, we recommend the Georeferencing Calculator (Wieczorek & Wieczorek 2020). To understand how each source of uncertainty contributes to the net overall uncertainty, see {gcm}#understanding-uncertainty[Understanding Uncertainty Contributions^] in the {gcm}[Georeferencing Calculator Manual (Bloom et al. 2020)^].

Using the Georeferencing Quick Reference Guide

The {gqg}[Georeferencing Quick Reference Guide (Zermoglio et al. 2020)^] is a practical guide for georeferencing giving step-by-step instructions on how to georeference a wide variety of locality types (see Georeferencing Workflow – Localities) following the best practices in this document and with specific reference on what to enter into the Georeferencing Calculator (Wieczorek & Wieczorek 2020).

Using the Georeferencing Calculator

The Georeferencing Calculator (Wieczorek & Wieczorek 2020) (A snapshot of the Georeferencing Calculator (Wieczorek & Wieczorek 2020) showing maximum uncertainty calculation for the locality: ‘10 mi E (by air) Bakersfield’.) is a tool to aid in georeferencing descriptive localities such as those found in museum-based natural history collections. It was originally designed for the Mammal Networked Information System (MaNIS) Project and has since been adopted by many other georeferencing initiatives. The current version and its Georeferencing Calculator Manual (Bloom et al. 2020) have been extensively upgraded to include new features and to bring it in line with this document.

The application makes calculations adapted from the methods originally described in the Georeferencing Guidelines (Wieczorek 2001) and later formalized in a peer-reviewed publication (Wieczorek 2004). We recommend its use generally by all natural history institutions to calculate [uncertainty] in [location] data without the need for a detailed understanding of the complicated underlying algorithms. The more institutions that use this one method, the more consistent will be the quality of data across and between institutions, making it easier for users to evaluate the quality of the data. We recommend reading both of the above-mentioned publications and the Georeferencing Calculator Manual (Bloom et al. 2020) for an understanding of the calculations involved and an understanding of how the Calculator works.

The Calculator can work online or locally in a browser (latest release available on GitHub). The source code is freely and openly available on GitHub.

Figure 3. A snapshot of the Georeferencing Calculator (Wieczorek & Wieczorek 2020) showing maximum uncertainty calculation for the locality: ‘10 mi E (by air) Bakersfield’.

Difficult Localities

Some localities are difficult to [georeference]. For some the recommendation is to not even try. These are generally localities without sufficient information, with conflicting or ambiguous information, or where the information is explicitly in question. Some localities reference a [feature] that can’t be found with easily available resources. For these it may be just a matter of applying enough effort, but if the project is on a budget that can not support lengthy investigations into difficult localities, they may need to be left for another time. Difficult localities are not uncommon. Don’t despair. Some interesting ones have been documented by the MaNIS project.

Some {marine}marine localities can also provide difficulties – for example "Off Mar del Plata". The trouble is, one doesn’t know how far "off" Mar del Plata the [event] took place. In terrestrial localities one can generally make a decision that it is between the feature and the next feature, but in the marine environment, that may not be as easy. Does it mean "within sight of", 5km, 12km, the EEZ boundary, the continental shelf…? One does not reliably know the end point so it makes it difficult (if not impossible) to georeference accurately. One good resource for finding marine localities, boundaries, etc. is the website marineregions.org (VLIZ 2019).

Determining Spatial Fit

Spatial fit, first formalized as the Reock degree of compactness (Young 1988, Reock 1961), is a georeferencing concept designed to measure how well a given geometric representation matches the original spatial representation. This is useful when spatial transformations change the way a [locality] is represented, either to mask its detail, or to match an agreed upon schema for data sharing (such as fitting locations to a [grid] cell).

A spatial fit with a value of "1" is an exact match or 100 per cent overlap. If the [geometry] given does not completely encompass the original spatial representation, then the [spatial fit] is zero (i.e. some of the original is outside the transformed version, which we interpret as not being a fit). If the transformed [shape] does completely encompass the original spatial representation, then the value of the spatial fit is the ratio of the area of the transformed [geometry] to the area of the original spatial representation. Special case: If the original spatial representation is a point and the geometry presented is not a point, then the spatial fit is undefined. The range of values of spatial fit is 0, 1, greater than 1, or undefined (see A diagram illustrating the spatial fit of a location that can be described by a polygon, a bounding box, a circle, or a point. c is the corrected center, r1 is the radial of the circle encompassing the polygon, r2 is the radius of the circle encompassing the bounding box. (Modified from Chapman & Wieczorek 2006). and Calculations of spatial-fit where the original spatial representation of a locality given by the polygon in A diagram illustrating the spatial fit of a location that can be described by a polygon, a bounding box, a circle, or a point. c is the corrected center, r1 is the radial of the circle encompassing the polygon, r2 is the radius of the circle encompassing the bounding box. (Modified from Chapman & Wieczorek 2006)., with area A., Table 7, Table 8 and Calculations of spatial-fit where the original spatial representation of a locality was given as the point C (A diagram illustrating the spatial fit of a location that can be described by a polygon, a bounding box, a circle, or a point. c is the corrected center, r1 is the radial of the circle encompassing the polygon, r2 is the radius of the circle encompassing the bounding box. (Modified from Chapman & Wieczorek 2006).).).

An example of the applicability of the spatial fit is where a point representing a terrestrial collection lies close to the coast, and the calculated [uncertainty] radius encompasses some {marine}marine area. In this case the spatial fit would be greater than 1 as it represents an area greater than the real uncertainty (Example of using spatial fit on the results of both a point-radius method and a refined shape method of describing uncertainty. Assuming the blue-shaded area is the "true" locality as we know the species is terrestrial, and the red circle is the point-radius method of representing the uncertainty, the ratio of the area of the red circle (5.726 sq km) divided by the area of the blue shaded area (~4.1 sq km) gives a spatial fit for the point-radius of 1.39.). Spatial fit is also a valuable measure for describing the degree of [generalization] of a sensitive species, for example see [Generalizing Georeferences for Sensitive Taxa and Locations] and Chapman 2020.

Figure 4. A diagram illustrating the spatial fit of a location that can be described by a polygon, a bounding box, a circle, or a point. c is the corrected center, r1 is the radial of the circle encompassing the polygon, r2 is the radius of the circle encompassing the bounding box. (Modified from Chapman & Wieczorek 2006).

A diagram illustrating the spatial fit of a location that can be described by a polygon, a bounding box, a circle, or a point. c is the corrected center, r1 is the radial of the circle encompassing the polygon, r2 is the radius of the circle encompassing the bounding box. (Modified from Chapman & Wieczorek 2006). illustrates a few examples of the definition of spatial fit and these are elaborated in the Tables below:

Table 6. Calculations of spatial-fit where the original spatial representation of a locality given by the polygon in A diagram illustrating the spatial fit of a location that can be described by a polygon, a bounding box, a circle, or a point. c is the corrected center, r1 is the radial of the circle encompassing the polygon, r2 is the radius of the circle encompassing the bounding box. (Modified from Chapman & Wieczorek 2006)., with area A.

The [spatial fit] of the white circle (r₂)	\$(pi r_2^2)/A\$
The [spatial fit] of the bounding box	\$(2 r_2^2)/A\$
The [spatial fit] of the yellow circle (r₁)	\$(pi r_1^2)/A\$
The [spatial fit] of the polygon	1
The [spatial fit] of the point C	0

Table 7. Calculations of spatial-fit where the original spatial representation of a locality was given as the bounding box in A diagram illustrating the spatial fit of a location that can be described by a polygon, a bounding box, a circle, or a point. c is the corrected center, r1 is the radial of the circle encompassing the polygon, r2 is the radius of the circle encompassing the bounding box. (Modified from Chapman & Wieczorek 2006)., with area \$2r_2^2\$.

The [spatial fit] of the white circle (r₂)	\$(pi r_2^2)/(2r_2^2)\$
The [spatial fit] of the bounding box	1
The [spatial fit] of the yellow circle (r₁)	0
The [spatial fit] of the polygon	0
The [spatial fit] of the point C	0

Table 8. Calculations of spatial-fit where the spatial representation of a locality was given as the circle (\$r_1\$) in A diagram illustrating the spatial fit of a location that can be described by a polygon, a bounding box, a circle, or a point. c is the corrected center, r1 is the radial of the circle encompassing the polygon, r2 is the radius of the circle encompassing the bounding box. (Modified from Chapman & Wieczorek 2006)., with area \$pi r_1^2\$.

The [spatial fit] of the white circle (r₂)	\$r_2^2/r_1^2\$
The [spatial fit] of the bounding box	0
The [spatial fit] of the yellow circle (r1)	1
The [spatial fit] of the polygon	0
The [spatial fit] of the point C	0

Table 9. Calculations of spatial-fit where the original spatial representation of a locality was given as the point C (A diagram illustrating the spatial fit of a location that can be described by a polygon, a bounding box, a circle, or a point. c is the corrected center, r1 is the radial of the circle encompassing the polygon, r2 is the radius of the circle encompassing the bounding box. (Modified from Chapman & Wieczorek 2006).).

The [spatial fit] of the white circle (r2)	Undefined
The [spatial fit] of the bounding box	Undefined
The [spatial fit] of the yellow circle (r1)	Undefined
The [spatial fit] of the polygon	Undefined
The [spatial fit] of the point C	1

Example of using spatial fit on the results of both a point-radius method and a refined shape method of describing uncertainty. Assuming the blue-shaded area is the "true" locality as we know the species is terrestrial, and the red circle is the point-radius method of representing the uncertainty, the ratio of the area of the red circle (5.726 sq km) divided by the area of the blue shaded area (~4.1 sq km) gives a spatial fit for the point-radius of 1.39. shows an example of applying the spatial-fit concept of a [point-radius] method of describing uncertainty where it is restricted to a shape method representation. For example, the location of a plant along the coast of north-east Madagascar – marked with the yellow X (Example of using spatial fit on the results of both a point-radius method and a refined shape method of describing uncertainty. Assuming the blue-shaded area is the "true" locality as we know the species is terrestrial, and the red circle is the point-radius method of representing the uncertainty, the ratio of the area of the red circle (5.726 sq km) divided by the area of the blue shaded area (~4.1 sq km) gives a spatial fit for the point-radius of 1.39.) – has an uncertainty radius of approx 1.35 km, but we know the record is of a terrestrial plant species so we can calculate the true area of uncertainty by excluding the marine biome using the shape method, thus the spatial fit is the ratio of the area of the red circle (5.726 sq km) divided by the area of the blue shaded area (~4.1 sq km) giving a spatial fit of the uncertainty radius of 1.39.

Figure 5. Example of using spatial fit on the results of both a point-radius method and a refined shape method of describing uncertainty. Assuming the blue-shaded area is the "true" locality as we know the species is terrestrial, and the red circle is the point-radius method of representing the uncertainty, the ratio of the area of the red circle (5.726 sq km) divided by the area of the blue shaded area (~4.1 sq km) gives a spatial fit for the point-radius of 1.39.