L
Journal of Experimental Marine Biology and Ecology 245 2000 215–224
www.elsevier.nl locate jembe
Precision of different methods used for estimating the abundance of the nitrogen-fixing marine cyanobacterium,
Trichodesmium Ehrenberg
Jeng Chang
Institute of Marine Biology , National Taiwan Ocean University, Keelung 202-24, Taiwan
Received 25 July 1999; received in revised form 3 September 1999; accepted 25 October 1999
Abstract
Variances involved in estimating the abundance of the nitrogen-fixing marine cyanobacterium, Trichodesmium Ehrenberg, were evaluated by repeated sampling using bottle casts and plankton
net tows at two stations in the southern East China Sea. The variance among individual samples and the variance arising from subsampling processes were separated by the method of analysis of
variance, and the coefficient of variation C.V. of an abundance estimate based on a single subsample was calculated. For bottle-collected samples, the major source of variation came from
taking subsamples from a water bottle. The C.V. of a single subsample estimate ranged from 57 to 184. For net-collected samples, variance in abundance estimation was mainly caused by
distinctive net tows, and when distributing materials in the receiving bucket into smaller containers. The C.V.s of single subsample estimates were 34 and 40, respectively. Tri-
chodesmium abundance estimated with bottle- and net-collected samples were further compared using data obtained from 17 stations in the East China Sea. Although a general distribution pattern
was supported by both methods, the correlation coefficient between them was 0.461, not significantly different from 0.
2000 Elsevier Science B.V. All rights reserved.
Keywords : Abundance estimation; Nitrogen fixation; Plankton nets; Sampling design; Subsampling; Tri-
chodesmium; Water bottles
1. Introduction
Trichodesmium Ehrenberg comprises a group of filamentous cyanobacteria distributed
Tel.: 1886-2-2462-2192 ext. 5308; fax: 1886-2-2463-3152. E-mail address
: b0176ntou66.ntou.edu.tw J. Chang 0022-0981 00 – see front matter
2000 Elsevier Science B.V. All rights reserved.
P I I : S 0 0 2 2 - 0 9 8 1 9 9 0 0 1 6 3 - X
216 J
. Chang J. Exp. Mar. Biol. Ecol. 245 2000 215 –224
mainly in tropical and subtropical oceans for a review, see Capone et al., 1997. Each filament, or trichome, is composed of about 100 cells, and trichomes often aggregate
together to form spherical or fusiform colonies Nagasawa and Marumo, 1967. Trichodesmium is the major organism that fixes dinitrogen in oceanic environments
Capone et al., 1997. Recent evidence indicates that the fixed nitrogen as a source of nutrient input supports a significant portion of global new production Karl et al., 1997.
An accurate assessment of spatial and temporal distribution of Trichodesmium is important to calculate regional and global nitrogen fixation rates.
To date, although a remote sensing algorithm is under development Subramaniam and Carpenter, 1994, the great majority of information about Trichodesmium dis-
tribution comes from microscopic examination of water samples. Water samples with a volume from 500 ml to 8 l are routinely used for such purposes, and a process of
sedimentation or filtration is required to concentrate trichomes before the actual counting Marumo and Asaoka, 1974a; Carpenter and Romans, 1991. A drawback of this
approach is that large, but sparse, colonies are easily missed due to small sample volumes Capone et al., 1997. In addition, unless the intracellular gas vesicles are
destroyed by acid treatment, the positively buoyant Trichodesmium trichomes may float to the surface during the sedimentation procedure Capone et al., 1997.
Another consideration of bottle-collected samples is that a sample with very limited volume may not be representative enough to reflect Trichodesmium abundance at a
sampling station. The highly heterogeneous distribution of Trichodesmium is clearly seen during a bloom event Karl et al., 1992, and similar distributions are very likely to
occur during nonbloom conditions. To overcome this difficulty, sampling with a plankton net is another commonly used method for the estimation of Trichodesmium
abundance e.g., Marumo and Asaoka, 1974b. The general recommendation is to use a mesh size of less than 100 mm and to tow the net at a low speed Capone et al., 1997.
However, single trichomes may escape through gauze apertures, and the use of a flowmeter to measure the volume filtered can be inaccurate Tangen, 1978.
Regardless of the tools used to collect water samples, some subsampling procedures are often involved so that only trichomes in an aliquot of the original sample are
enumerated. For example, out of materials collected in the receiving bucket the cod-end jar of a plankton net, only a small portion is preserved and counted. Similarly, in a
counting chamber containing concentrated water samples, only trichomes in limited fields are enumerated. These subsampling processes inevitably introduce variations and
affect the precision of the final counts Venrick, 1978.
When abundance data are reported, it is difficult to make comparisons and perform further analyses if the range of variation is unknown. For Trichodesmium, although its
unique cellular organization and distribution pattern are known to cause large variations in abundance estimation, no statistically sound evaluation of the magnitude of uncertain-
ty has been performed. In this report, multiple samples were obtained from a fixed location, and multiple subsamples were taken from each sample to estimate the variances
contributed by each step of sample handling Venrick, 1978. Once these variances are known, the standard error of the trichome concentration estimated from a single
subsample can be determined Woelkerling et al., 1976. In addition, trichome concentrations estimated simultaneously by bottle-collected and net-collected samples
were compared to detect if discrepancies exist between the two sampling methods.
J . Chang J. Exp. Mar. Biol. Ecol. 245 2000 215 –224
217
2. Materials and methods
