Collection Techniques and Logistics

Page Contents

• Site selection
• Activities at each station
• Specimen collection
• Location and tracking with GPS
• Image documentation
• Initial sample processing upon return
• Making data available
  • Site numbers and notes
  • Images
  • Web site

These detailed notes are intended for researchers interested in the specific techniques we used (whether to evaluate the usefulness of our data or for help in planning similar expeditions). Please feel free to contact us with any further questions.

Site selection Back to Top

We planned our route (thanks to generous help from Dr. Khayankharvaa Terbish at the National University of Mongolia) to pass through diverse environments that might contain temporary (ephemeral) bodies of water. Broadly, we left the capital Ulaanbaatar headed south to the Gobi Desert. Turning west and north, we headed through hills and up to higher elevations in Arkhangai Aimag. Finally, we headed back east to Ulaanbaatar. See the route map for the detailed track.

Along our route, we sampled wherever we saw water that could possibly hold small crustacea that we could collect. In many cases consultation with local Mongolians enabled us to find small ponds, or let us know that mapped lakes and rivers were currently dry. The only situations in which we did not sample observed water were cases where the water was flowing too quickly to make it plausible that we would find slow-swimming small crustacea, or if the water was clearly contiguous with (and near to) a previously-sampled body of water.

Hence, along the track of our route, lengths with no sampling sites indicate lengths where we found no water. Wherever there was water, we sampled.

Activities at each station Back to Top

We quickly developed a routine set of activities performed at each sampling station on the expedition. Two researchers (Wetzer and Boyce) immediately went to the the water to look for any signs of small crustaceans near the edge. One researcher (Pentcheff) recorded the site coordinates, and a general description. We measured the water temperature using a small thermometer and measured the salinity using a temperature-compensated optical refractometer. While the specimens were being collected, the site's appearance was recorded using several digital photographs. Following collection, a general description of what was collected (and the number of bags of specimens) was recorded.

Specimen collection Back to Top

At each collecting site Wetzer and Boyce assessed the depth of the water. Shallow water pools, puddles, or streams were most common and were ideally suited for sampling with a 63 μm mesh plankton net with ca. 15 cm diameter opening.

Larger bodies of water (e.g. rivers, lakes) were sampled with a standard 240 μm mesh plankton net. Sampling effort was put into collecting from as many microhabitats at each sampling site as possible.

Specimens were concentrated and rinsed with freshwater into the center of the net and then transferred to Nasco Whirl-Pak bags. Clumped vegetation and debris was transferred to the Whirl-Pak bag with a large serving spoon or a pair of forceps used in chopstick-fashion not to squash specimens. A plastic funnel was then placed over the opening of the bag, the plankton net was inverted and placed into the funnel, and the contents of the net were washed into the bag with 95% ethanol dispensed from a plastic liter-size soda bottle. All samples were preserved and stored in 95% ethanol. The bag was closed, a collection label was prepared, and the sample was double-bagged with its label contained between the outer and inner bags.

Location and tracking with GPS Back to Top

A Garmin Legend GPS (Global Positioning System) receiver was used during the expedition. At each sampling site, the coordinates of the site were recorded both on paper in the field notes and as an electronic waypoint in the GPS receiver. The waypoint number was recorded in the written notes, serving as a unique identifier for the site during the trip (but see Making data available below).

As well as point coordinates, the GPS reciever was used to record the entire track of the expedition. The Track Log feature was set to record locations Automatic / Least Often. We chose to avoid storing tracks as "Saved Tracks", since doing so discards time and altitude information from each point along the track. Hence we were limited to the "Active Track" log. At our rate of travel, this usually filled by late afternoon and required a brief stop to download into our laptop before continuing.

On occasion we also used the GPS receiver as an aid to navigation. The internal electronic map (from Garmin's MapSource WorldMap) proved to be reasonably accurate and was occasionally useful in determining our location relative to rivers and river valleys.

Image documentation Back to Top

At each site several documentary digital photographs were taken to record the general topography of the site. Other images were sometimes taken to show the water clarity or appearance as well as local vegetation.

At the close of sampling for a site, a digital image of the field notes for that site was taken. These images proved to be very useful. They served as markers in the digital image stream that allowed us to unequivocally assign the correct site for every image. They also served as a backup in case the data notebook was lost (since all images were downloaded to a laptop each evening).

Initial sample processing upon return Back to Top

Upon return to the laboratory, specimen data were transcribed from field notebooks. GPS and locality data were verified and specimen labels were prepared. Data were entered into a FileMaker Pro database and collection labels were printed on 100% cotton rag, acid-free archival paper. Labels were printed in Arial font, 8-10 points in size on a Hewlett Packard LaserJet printer. Each sample was processed by cutting off the top of the Whirl-Pak bag with scissors. The contents were poured into the same size mesh (either 63 or 240 μm) as was used to originally collect the sample over a funnel resting in a jar for stability. The sample was gently rinsed under running cold water to remove excess fine sediments which could not be removed in the field.

The sample was then transferred to museum grade glassware and the jar was filled to the top with fresh 95% ethanol. Samples are now ready for long term storage or further processing by sorting organisms to various taxonomic levels.

Making data available Back to Top

A challenge with any expedition such as this is making the data available to the people who can use it. We chose to make the information (textual and image-based) available through a WWW interface.

Site numbers and notes Back to Top

Upon arrival back in the U.S., the text notes from the expedition describing each sampling site were entered into the Natural History Museum of Los Angeles County's Marine Biodiversity Center's database. (The administrator of the database graciously allowed us to use it despite the non-marine nature of our samples.)

In the context of that database, each sampling site was assigned a Processing Center site/lot number. That number became the definitive unique key for the site's information. From that point onward, the in-field use of GPS waypoint numbers was discontinued and Processing Center Site Numbers were used.

Images Back to Top

At each site, a number of digital (and sometimes film) photographs had been taken. Each digital image's site assignment and internal photographic data were entered into an image database (separate from the Processing Center's site/lot database). Each film photograph's site assignment and a brief description were also entered. In many cases, comparisons with the digital photographs were useful in assigning film photographs to specific sites. Selected slides were digitized and added to the collection of digital photos. Images that were taken at locations other than the numbered sampling sites were assigned the numbers of the sampling site prior and the sampling site following the photograph.

All digitized photographs were processed using the ImageMagick suite of programs to add copyright information both to the internal "comment" field and explicitly to the image itself. The same package was also used to generate smaller versions of each image (640 and 96 pixels wide) for use on the web site.

Web site Back to Top

The interface to the databased textual and image information is provided by an Apache web server running on a Red Hat Linux server. The online versions of the databases are kept in a MySQL database. Processing Center data are transferred from Filemaker Pro databases on Macintosh computers to the MySQL database using DB::CSV2db. The Mongolia web site's infrastructure is created in the Mason web development system.

All maps on the web site were generated using the GMT mapping package. The relief maps use the USGS GTOPO30 30-second resolution digital elevation data. The tiny index maps use GMT's built-in coastline database.

Place names within Mongolia and their locations come from the U.S. National Imagery and Mapping Agency GEOnet Names Server. Because of language and character set variation, there are often multiple variants for an individual place name within Mongolia. In this case, the GEOnet "Native" name spelling was generally used.