This site will look much better in a browser that supports web standards, but it is accessible to any browser or Internet device.

Project

Table of Contents

Introduction

The Missouri Herpetological Atlas Project (MOHAP) was intiated in 1997 as a result of the desire to obtain and easily update detailed distribution maps for Missouri amphibians and reptiles.

A database was established to serve as the basis for summarizing existing collection specimens, verifying new distribution records, and tracking changes in individual species distributions within the state. Qualification for inclusion in the database generally requires a catalogued voucher specimen housed in an institutional collection. However, in order to be valuable in tracking changes in the distribution of individual species, it is important to document historical records. Many of the specimens collected prior to 1960, most notably those reported by Julius Hurter and Paul Anderson, were documented by specimens that have since been lost or destroyed. In order to provide the most accurate picture of the distribution of the state's herpetofauna we have included literature records that we believe to be valid. Currently, the MOHAP database contains 31,495 entries representing the specimens housed in 34 museum collections and cited in 32 historical literature sources; 5,118 documented county records; 6,884 unique localities; and 12,866 unique species / locality combinations. The state's herpetofauna consists of 113 species.

Using a database for all collections and GIS software to produce maps allow for an endless combination of maps. However, a basic paper atlas is useful for field work and reference and allows for a wider public distribution. The first paper version of the atlas was released to the public in April 2002 and contained all records through the 2001 collecting season. It was a rather basic publication without an introduction or a literature cited section. An annual update containing the previous year's collections plus other sources has been published in 2003, 2004, and 2005. An updated version that included the 2006 and 2007 collecting seasons was released in 2008.

Project Plan

Phases

The project was deemed a long-term project from the beginning. Three natural but distinct phases can be used to describe the progression of the project since its inception. Phases I and II are mostly complete although some museum collections are still needed for the database. Phase III is underway and includes the updates recently made to this site.

  1. Summarize existing knowledge in database format
  2. Publish a basic paper atlas for public consumption
  3. Publish extended distribution data and commentary on each species

Goals

Several long-term goals were established for the project. Of these goals, only the last four remain partially or completely unfulfilled.

Timeline

Sources

The following museums or collections are included in the database and atlas. Except for the specimens from the University of Missouri--Columbia, specimens have not been verified by the authors. Sources marked with an asterisk (*) are included from published accounts only and are presumed to represent only part of the collection's Missouri records. All Missouri material from these sources has not been included in the database or the printed atlas.

Technical Details

Database

The atlas database was converted in 2008 to a PostgreSQL relational database. The nature of a relational database reduces keying and other errors dramatically and allows the flexibility to include an almost unlimited number of data elements associated with each record with very little additional storage requirements. It's also possible to produce maps based on a data element other than species. For example, maps can be produced for collectors, institutions, or dates. Basically, the following data elements are stored for each physical collection record, although storage in a relational database is not necessarily straightforward and details for each element are often stored in a separate set of tables:

The reliability code can have six potential values. As stated elsewhere, a record is assumed to be good (G) unless we have a good reason to think it otherwise. Records that are still considered valid might be instead marked as a literature (L) record or a duplicate (D) record. Duplicates arise because we have included literature records that might also be included in a museum collection. Most notably, Anderson's records were recorded from The Reptiles of Missouri and also from the University of Kansas and Chicago Academy of Sciences collections. The duplicate code was always marked on the record entered from the literature source. A few records were marked as questionable (Q), erroneous (E), or unknown (U). Currently, questionable records show up in the atlas as a question mark (?) while erroneous and unknown records do not show up at all. However, all three of the latter categories will be discussed on a record-by-record basis for each species.

The Atlas

To create locality and county records maps, a simplified table is exported from the relational database into a PostGIS table, which handles the conversion of UTM coordinates to a spatial geometry that can be used with county base maps, provided by the Missouri Spatial Data Information Service. A Python plug-in is used in Quantum GIS to export individual taxon locality and county records maps as PNG images to a common directory. These images can be used for both the printed atlas and the web site. The final atlas document is produced in portable document format (PDF) by pulling all sections (including the PNG image maps) together using a Python script outfitted with the ReportLab PDF bindings.

Open-Source Software

The atlas project itself is non-commercial in nature. The final printed product is available for download on the publications site gratis (i.e., free of charge) and additional information is available elsewhere on this site. In addition, it is our desire to prevent the project or its data from being subjected to the restrictive licensing conditions often found with commercial software products. As a result, all technology-related aspects of the project are supported by a suite of open source software that is both gratis (free as in beer) and libre (free as in speech). The following packages all play a significant role in the project.