Efficiency of different mesh sizes for isolating fossil chironomids for stable isotope and radiocarbon analyses
- First Online:
- Cite this article as:
- van Hardenbroek, M., Heiri, O. & Lotter, A.F. J Paleolimnol (2010) 44: 721. doi:10.1007/s10933-009-9327-6
- 315 Views
We examined the effects of sieving with different mesh sizes on the efficiency of processing fossil chironomids from lake sediments for isotope analyses. Results obtained for three different sediments indicate that each of the studied sieve fractions (100–150, 150–200, 200–250, 250–300, >300 μm) contain a similar proportion of the overall mass of chironomid fossils in a sample. However, the sorting time needed to separate chironomids from other sieve residue is disproportionately large for smaller mesh sizes. Employing sieves with a 200-μm rather than the 100-μm mesh commonly used for standard palaeoecological analyses of fossil chironomids decreased processing time for a given mass of fossils by 30–58% in our study. For optimizing the efficiency of chironomid sample processing for stable isotope and radiocarbon analysis we therefore recommend a 200-μm mesh size sieve, although the sorting of all >100-μm fractions may be necessary in sediments with low chironomid abundances. Excluding certain small taxa from isotope analysis, may structurally bias isotope values of samples. Therefore, further studies on taxon-specific isotope analysis are required to quantify these effects.
KeywordsFossil chironomidsStable isotopesRadiocarbonPalaeolimnologyLake sediments
Non-biting midges (Insecta: Diptera: Chironomidae) are sensitive indicators for a variety of environmental variables. The chitinous remains of chironomid larvae preserve well, are ubiquitous in lake sediments and have been used to reconstruct physical and chemical variables such as air or water temperature (Walker and Cwynar 2006; Brooks 2006; Heiri et al. 2007), total phosphorus (Brooks et al. 2001; Langdon et al. 2006), chlorophyll a (Brodersen and Lindegaard 1999), oxygen availability (Quinlan et al. 1998), or lake depth (Korhola et al. 2000).
The potential of fossil chironomids for isotope studies has first been shown for 14C dating (Jones et al. 1993; Fallu et al. 2004). High-latitude or high-altitude sites are often devoid of terrestrial plant remains and bulk 14C dates from lake sediments are often too old due to either contamination by allochtonous material or hard-water effects (Olsson 1991; Abbott and Stafford 1996). In such circumstances chironomids can be one of the few reliable sources of carbon available for dating. Recently, chironomid fossils have also been used in stable isotope studies, such as δ18O-based temperature reconstruction (Wooller et al. 2004, 2008) or the reconstruction of lake productivity using stable carbon and nitrogen isotopes (Wooller et al. 2008). One of the major difficulties in all attempts to measure isotopes in chironomid fossils is to attain the required minimum sample mass for isotope analyses. The amount of chironomid material necessary for Accelerator Mass Spectrometry (AMS) 14C dates was reported by Jones et al. (1993) as being 250–400 μg carbon (~800 head capsules), and Fallu et al. (2004) used between 180 and 370 μg chironomids (in their case equivalent to ~1,300–2,500 head capsules from unsieved sediment). The amount of larval chironomid head capsules necessary for an oxygen isotope measurement is approximately 100 μg (300–700 head capsules; Wooller et al. 2004) with a minimum of 50 μg (~120 head capsules) reported by Wang et al. (2008).
The most commonly used method to isolate head capsules from other sediment components is to wash the sediments through a 90–115-μm sieve and subsequently hand pick the remains under a dissecting microscope. The choice of the commonly used ~100-μm mesh size for sieving sediments is based on the observation by Walker and Paterson (1985) that most head capsules are larger than 100 μm in diameter and thus retained on a 100-μm sieve. Even the sorting of 50–100 head capsules per sample commonly used in palaeoecological analyses can take an analyst several hours for sediments with large amounts of obscuring debris. Therefore, the time needed for sorting and isolating chironomid remains is an important constraining factor for the number of samples that can be processed in chironomid-based isotope studies.
Missing certain small taxa due to sieving with a too wide a mesh and excluding them from numerical analyses may not only have considerable effects on the palaeoecological interpretation of the assemblage but can also significantly bias quantitative chironomid-based environmental reconstructions (Heiri and Lotter 2001; Quinlan and Smol 2001). For isotope analyses, however, mass is often more relevant than the number of individuals, at least when the isotopic composition of chironomids is expected to be similar within a lake basin and between species of different size. It may thus be beneficial to select a processing method that optimises the preparatory process, i.e., that yields the greatest mass of chironomid fossils in the shortest processing time.
A method to concentrate chitinous fossils from lake and stream deposits is the floatation of insect remains in a denser, apolar organic liquid. Using kerosene, a grade mineral oil, insects can be concentrated from any sediment material (Coope 1986). However, Rolland and Larocque (2007) recently demonstrated that this method yields reduced amounts of large head capsule types such as the fourth instars of Chironomus because these are often filled with sediment and are therefore heavier. This is unfortunate because large, heavily sclerotized head capsules provide a disproportionately large share of the chironomid sample mass available for isotope analyses. An additional drawback for carbon and hydrogen isotope analysis is the introduction of carbon and hydrogen from kerosene. Although it may be possible to remove the kerosene by chemically cleaning the samples, the additional time needed for such a cleaning step and the introduction of potential contaminants makes this approach unattractive for isotope studies.
Large head capsules yield several times more mass per specimen than small head capsules. Therefore, selectively isolating large head capsules from the sediments will concentrate a large proportion of the total chironomid mass available in a sample. This can be done in a standardized way by sieving with mesh sizes >100 μm which allows smaller head capsules and many other sedimentary particles to be washed through the sieve. Previous studies have examined the effect of different mesh sizes on chironomid sample processing. However, these studies mainly examined the effect of mesh size on the representativeness of chironomid assemblages identified and enumerated under the light microscope (Walker and Paterson 1985; Verschuren and Eggermont 2007) and no studies are available that document the effects of the mesh size used during sieving on the sorting time of chironomid samples. In this study we assess the effect of mesh size on fossil chironomid sample mass and processing time, with the aim of providing a recommendation for the most time-efficient mesh size to be used to concentrate fossil chironomids for stable isotope analyses and AMS radiocarbon dating.
Main characteristics of the three lake sediment samples used in this study
Unnamed tundra pond, Russia
ca. 12,800–14,800 cal. years BP
Altitude (m a.s.l.)
Water content (%)
Total number of head capsules >100 μm in sample
Dry weight of analyzed sample (g)
Number of head capsules per g dry weight
Total weight of head capsules >100 μm in sample (μg)
Number of larval head capsules (HC), HC mass, and sorting time for 1.0 g dry weight of sediment samples A, B, and C
Mesh size (μg)
Number of HC
Cumulative number of HC
Cumulative weight (μg)
Sorting time (min)
Cumulative sorting time (min)
Efficiency of sorting (μg/min)
Efficiency of sorting (HC/min)
HC weight (μg/HC)
Results and discussion
The number and mass of chironomid head capsules in each size fraction vary between the three analyzed sediment types. The total mass of the head capsules per gram dry weight of sediment is very similar in sediment samples A and C (1,317 and 1,539 μg, respectively), but only 43 μg in sediment sample B (Table 2). The highest concentration of head capsules was found in sediment sample C, which contains 980 head capsules per gram dry weight. This is twice as much as in sediment sample A and more than 15 times the concentration found in sediment sample B (Table 2). Furthermore, the average weight of a head capsule is higher in sediment sample A (4.1 μg) compared with sediment samples B and C (0.9 and 1.6 μg, respectively), indicating that the average mass of individual head capsules is site-specific.
Chironomid mass (μg)
Time needed for sediment sample A (h)
Time needed for sediment sample B (h)
Time needed for sediment sample C (h)
Time reduction relative to 100 μm fraction
Sample preparation with larger mesh sizes will require larger quantities of sediments to retrieve the same mass of head capsules. This is not necessarily problematic if the concentration of head capsules is high in the sediment record of interest, but it may decrease temporal resolution of palaeoenvironmental reconstructions if concentrations are low and adjacent samples have to be pooled. When preparing chironomid samples for isotopic studies, we therefore recommend to pre-screen sediment records to see if the concentrations of chironomids in the >200-μm fraction are sufficient before deciding on a certain mesh size. If chironomids are abundant, we recommend using 200-μm sieves to process samples rather than the 90–115-μm mesh commonly used for palaeolimnological studies as this may shorten sorting time by 30–58% (Table 3). However, we also recommend to retain the fractions <200 μm until the samples have been weighted, so that additional chironomids can be isolated from the 150–200- or 100–150-μm fractions if necessary to obtain the required minimum mass for analysis. In that case all samples should also include the 150–200- or 100–150-μm fractions to prevent size-dependent bias.
Our results indicate that sieving of chironomid samples with mesh sizes in the range of 150–200 μm can significantly reduce processing time compared with samples sieved with the standard mesh size of 90–115 μm. Selection of coarser sieve for chironomid sample preparation will therefore enhance the temporal resolution that can be achieved in studies of the isotopic composition of fossil chironomid assemblages. An important caveat, however, is that mesh size will potentially affect results if different chironomid size classes or taxa are characterized by different isotopic values. For example, stable carbon isotope values can be very variable in different chironomid taxa within a lake basin (Grey et al. 2004; van Hardenbroek et al. 2009), with strongly depleted values reported for some chironomids. A selective enrichment or elimination of head capsules of 13C-depleted chironomids associated with choosing a certain mesh size would therefore lead to biased isotopic measurements on fossil chironomid samples. In contrast, Fallu et al. (2004) assumed that for 14C analyses the isotopic composition of chironomid fossils indiscriminately reflects isotope concentrations in the lake water. Similarly, Wooller et al. (2004) demonstrated that chironomid δ18O is equilibrated with the δ18O of lake waters in which larvae live, if lakes with short residence times are examined. In these situations it can be expected that mesh size will have a minor effect on isotopic measurements of fossil chironomid assemblages, although it remains to be demonstrated whether chironomid δ18O is unaffected by the vital effects (e.g., habitat, instar effects, temperature fractionation), which have been described for inorganic remains of lacustrine invertebrates (Ito 2001).
We would like to thank M.F. Mortensen for providing sediments from Slotseng for our experiments and K. van Huissteden for the sediments from a Siberian tundra pond. Comments from anonymous reviewers greatly improved this manuscript. This research has been supported by the Darwin Centre for Biogeology. This is Netherlands Research School of Sedimentary Geology (NSG) publication no. 20081202.
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.