Standard Methods for Sampling Diatoms for the Monitoring and Assessment of Streams

Bruce Chessman, Department of Land and Water Conservation, NSW, Australia
Peter Gell, University of Adelaide, SA, Australia
Peter Newall, CRC-Freshwater Ecology, Environment Protection Authority, VIC, Australia
Jason Sonneman, Monash University, VIC, Australia 

Introduction

The standard methods presented here for the sampling of diatoms for bioassessment and biomonitoring of streams are those being used in the collection of several large data sets across south-eastern Australia (Chessman unpublished data, Gell unpublished data, Newall unpublished data, Sonneman unpublished data), covering several hundred sites.

It is expected that most or all of these data sets will be combined and form the basis of a national diatom data base. Consequently, researchers using these methods will be able to compare their results with the national data base and its accompanying environmental data. Similarly, gathering diatom data using these methods will enable those data to be added to the data base.

Diatom samples for the data bases have generally focussed on natural substrata, wherever available. Future additions to the data base may include artificial substrata.

The Methods

Substratum Selection

Wherever possible, two samples should be taken:

i) hard surface; and

ii) mud/detritus

Hard-surface samples are expected to be the most consistently collected, are faster to analyse and less likely to be contaminated with dead cells than the mud/detritus samples. It is expected that hard-surface samples will be the most readily used in the development of transfer functions and in general site assessment. Therefore, if it is at all possible to get a hard-surface sample, it is import ant to do so.

Hard-Surface Samples

 The preferable substratum is a rock that is big enough to have remained stable under most flows, yet small enough to be picked out of the water (e.g. approx 10-30 cm diameter). The key will be to "make do" where the preferred situation doesn't exist. So, if there are no rocks of the right size, use larger or smaller ones. If there are no rocks at all, then the preferred substrata are logs, othe r woody debris, and macrophytes (in that order). The type of substratum used should be recorded.

Wherever possible, these samples should be from riffle or run sections of the streams. It is preferable to sample along the thalweg of the stream (i.e. the deepest part of the channel where the main current flows). The sample should be a composite of scrapings from three rocks (or logs, etc.) at the site, with each individual substratum being at least 20 metres apart (where practical) from the o thers being sampled. Where it is not practical to sample from 20 metres apart, the three substrata should be as far apart as practical and the distance should be recorded.

Diatoms are collected from the hard surface by scraping with a blade (e.g. from a pocket knife) or from a stick, such as a sharpened ice-cream stick. When a pocket knife is used, it must be thoroughly washed between sites. When sharpened ice-cream sticks are used, a new one should be used for each site.

It is important to scrape from an area of the rock that was exposed to light when it was in the stream (i.e. the top or side of the rock). Also, it is preferable to scrape an area that does not have other growths (such as moss, lichen, or filamentous algae). Sometimes the options may be limited and a compromise will be needed on some of these preferences. If there is some detritus on the rock ( or other substratum), the sample may actually be part "soft surface" community as well as "hard surface" community. To avoid this contamination of the hard surface sample, shake any hard surfaces (under the water) prior to sampling. For consistency, a vigorous shake for 3 seconds is recommended.

It is preferable (if possible) to sample from approximately 15 cm depth. This measure is also selected purely for maintaining consistency. If the substratum in the thalweg of the stream is deeper than this, then deeper is fine. Two exceptions apply: 

1. If the substratum is too deep to sample in the thalweg, then sample where practical; and

2. If the water is too turbid to see the bottom of the stream in the middle, then keep to the 15 cm rule.

If the stream has no riffles or runs then slow-flowing or still sections are all that can be sampled. If there are no hard substrata, then there cannot be a hard-surface sample.

Mud/Detritus Samples

Wherever possible these should be from the edges of pools or protected areas (such as behind large rocks or logs). The sample is collected using a pipette. Press the bulb of the pipette, then gradually release it while dragging the nozzle over the surface of the sediment. This should be done at three locations in the stream reach. These three locations should be at least 20 metres apart and ma y be on opposite sides of the stream. A new pipette should be used for each site (plastic disposable pipettes are ideal for this).

The preferred depth of sampling is 5 cm - again this is just for consistency and more shallow is preferable to going deeper.

Two important points to keep in mind when selecting sample points are: 

1. The soft-substratum sampling should be done at a point in the stream that has been inundated for several weeks. In streams where the water depth fluctuates considerably, it is better to sample a little deeper if this means that the sample point is more likely to have been inundated for the required length of time; and

2. Although quiet sections of the stream may offer the most suitable habitat for detritus accumulation, diatom assemblages from areas that do not have substantial movement of water over them may reflect the internal nutrient dynamics of those areas rather than the water quality of the main stream. Therefore isolated pools or backwaters without significant streamflow should not be sampled as strea m assessment sites.

As for the hard-surface samples, if there is not a suitable place to sample, then there cannot be a soft-surface sample from that site.

Warnings

1. Diatoms are often sampled as part of a larger biological monitoring program. When macroinvertebrates are also being sampled, the associated kicking and sweeping activity can disturb large areas of the stream site. It is important to collect the diatom samples from undisturbed areas. Because diatom sampling requires only three rocks (or other substratum types) and three sheltered areas it is best to sample the diatoms prior to the invertebrate sampling. Otherwise, collect the diatom samples upstream of the macroinvertbrate samples. 

2. There is currently some debate as to the best method of preservation of the diatoms and whether the frustules dissolve in the various preservatives. Preservatives most commonly used are ethanol (approx 70%) and Lugols iodine solution. Regardless of which preservative is used, it is best to process the samples as soon as possible after collection to minimise the risk of dissolution. 

3. Because diatoms are minute and virtually ubiquitous (often occurring in tap water, for example), and their frustules (shells) are resistant to decay, it is very easy to contaminate samples with either living or dead diatoms from elsewhere. Extreme cleanliness is therefore required during sampling. Dirty hands, used and inadequately washed sampling gear and containers, and preservative dispens ers that have been dipped into previous samples are all potential sources of contamination. Procedures must be followed that prevent contamination from such sources. 

 

Diatom Sampling Field Notes

Stream and site:

Hard Surface Sample

Detritus Sample

Surface

Type

Rock / Log / Debris / Other (describe) N/A

Depth

15 cm / other (measurement)

5 cm / other (measurement)

Habitat

Riffle / Run / Other (describe)

Pool / backwater / Other (describe)