182
studies investigating size distribution found generally
increasing abundances with decreasing size classes (Imhof
et al. 2016). Even though the applied methodology will
define the lower size limit, the post-sampling procedures will
allow for classification into different size classes. Thereby,
Hanke et al. (2013) recommended to allocate MP particles
into size bins of 100 μm. Although this recommendation
would provide high resolution datasets, in practice this is
almost not feasible, as the preparation of microplastic samples is already very time consuming and, for instance, additional sieving steps would further increase analysis time.
Further, depending on the research question different size
categories are of importance. If, for example, pictures of the
microplastic particles are taken during analysis, it is possible
to obtain data on the size at a later time point in case the data
would be requested for comparative analysis.
Standardization of reporting units is a further necessity to
increase comparability among data sets. So far, different
sampling strategies have led to various reporting units (e.g.,
m
2
, m
3
, ml, l, g, kg) (Hidalgo-Ruz et al. 2012; Löder and
Gerdts 2015; Costa and Duarte 2017). For MPs in the environment (excluding biota samples) either bulk or volume
reduced samples are taken. Thus, a volume measurement can
always be obtained and should be the minimum information
reported. Additional reporting of sampling depth as well as
weight measurements for sediment samples will further
increase data quality.
Finally, reporting of meta data like prevailing wind direction, sea state, beach morphology, rainfall, and so on would
improve the interpretation of the data collected (Barrows
et al. 2017). In the current literature, missing information
range from unreported size ranges, replication, detected
numbers of particles to sampling locations (Filella 2015;
Besley et al. 2017). Comprehensive reporting of the applied
methods is a crucial part and not only a requirement for
reproducibility, but further gives the reader the ability to
judge about the representativeness of the study, as well as the
conclusions drawn from the results.
Sampling Equipment
Further considerations should be made on the sampling
equipment, as this will define the size range of MPs in the
study, as well as reporting units. For beach sediments, sampling equipment is well established (Hidalgo-Ruz et al.
2012; Hanvey et al. 2017), it only remains important to consider, whether to collect a bulk or a volume reduced sample.
For the latter, a lower size limit is defined. For bottom sediments corers, Van Veen or Ekman grabs can be used, however, grabs disturb the surface layer of the sediment and
corers do not only take the sediment but also the water layer
above the sediment (Löder and Gerdts 2015).
For water samples, nets of various types have been used
(Table 1 gives an overview of the used equipment found in
the current literature). Most commonly, manta nets are the
device of choice (Costa and Duarte 2017), where the reduced
sample volume limits the lowest size class of investigated
MPs mostly to 300–350 μm (Filella 2015). Thus, some
researchers used bottles to take bulk samples of the water
surface (Dubaish and Liebezeit 2013; Barrows et al. 2017),
which, however, results in small sample volumes.
Nevertheless, sampling lower size ranges, Barrows et al.
(2017) found MP concentrations were several orders of magnitude higher in bottle samples than manta samples. To
obtain larger sample volumes, others took several bottles or
buckets of surface water and concentrated the material on
filters with smaller mesh sizes on board (hand-nets; Chae
et al. 2015; Kang et al. 2015). Moreover, contamination
issues through high air exposure times during a manta trawl,
as well as filtering samples on board, motivated researchers
to further develop pumping systems (Desforges et al. 2014;
Lusher et al. 2014; Enders et al. 2015). One of the first studies comparing different methodologies for the same size
class (300–5000 μm) was conducted by Setälä et al. (2016)
comparing their custom-made pump to manta trawls.
Preliminary results from the pump (collecting surface water
in a depth of 0–0.5 m) did not significantly differ from the
results obtained by the manta net. Another interesting solution to decrease sampling effort has been published by Edson
and Patterson (2015). They designed an automated sampling
device (MantaRay), which automatically pumps sea surface
water at a depth of 30 cm, while drifting through the water.
Thereby, particles are concentrated on a filter and 28 successive samples can be taken. For the prototype, 500 μm stainless steel sieves were used. Such an instrument can decrease
sampling effort and airborne contamination, which is often a
challenge when conducting trawls. One drawback could be
the autonomous operation of the MantaRay, which limits the
control over the area sampled. Moreover, an optical sensor is
implemented to ensure that only water containing particulate
matter is filtered. Thereby, especially small MP particles
could be overlooked so the influence on the obtained results
must be further evaluated.
Independent of the applied method, decreasing mesh
sizes will increase the content of organic and inorganic material, which could lead to smaller sample sizes as meshes will
become clogged faster, but also to increased sample preparation time in the laboratory. In any case, negative controls
should be run, as most of the used methods may contain
polymer materials which are a further source for
contamination.
T. Hamm et al.
studies investigating size distribution found generally
increasing abundances with decreasing size classes (Imhof
et al. 2016). Even though the applied methodology will
define the lower size limit, the post-sampling procedures will
allow for classification into different size classes. Thereby,
Hanke et al. (2013) recommended to allocate MP particles
into size bins of 100 μm. Although this recommendation
would provide high resolution datasets, in practice this is
almost not feasible, as the preparation of microplastic samples is already very time consuming and, for instance, additional sieving steps would further increase analysis time.
Further, depending on the research question different size
categories are of importance. If, for example, pictures of the
microplastic particles are taken during analysis, it is possible
to obtain data on the size at a later time point in case the data
would be requested for comparative analysis.
Standardization of reporting units is a further necessity to
increase comparability among data sets. So far, different
sampling strategies have led to various reporting units (e.g.,
m
2
, m
3
, ml, l, g, kg) (Hidalgo-Ruz et al. 2012; Löder and
Gerdts 2015; Costa and Duarte 2017). For MPs in the environment (excluding biota samples) either bulk or volume
reduced samples are taken. Thus, a volume measurement can
always be obtained and should be the minimum information
reported. Additional reporting of sampling depth as well as
weight measurements for sediment samples will further
increase data quality.
Finally, reporting of meta data like prevailing wind direction, sea state, beach morphology, rainfall, and so on would
improve the interpretation of the data collected (Barrows
et al. 2017). In the current literature, missing information
range from unreported size ranges, replication, detected
numbers of particles to sampling locations (Filella 2015;
Besley et al. 2017). Comprehensive reporting of the applied
methods is a crucial part and not only a requirement for
reproducibility, but further gives the reader the ability to
judge about the representativeness of the study, as well as the
conclusions drawn from the results.
Sampling Equipment
Further considerations should be made on the sampling
equipment, as this will define the size range of MPs in the
study, as well as reporting units. For beach sediments, sampling equipment is well established (Hidalgo-Ruz et al.
2012; Hanvey et al. 2017), it only remains important to consider, whether to collect a bulk or a volume reduced sample.
For the latter, a lower size limit is defined. For bottom sediments corers, Van Veen or Ekman grabs can be used, however, grabs disturb the surface layer of the sediment and
corers do not only take the sediment but also the water layer
above the sediment (Löder and Gerdts 2015).
For water samples, nets of various types have been used
(Table 1 gives an overview of the used equipment found in
the current literature). Most commonly, manta nets are the
device of choice (Costa and Duarte 2017), where the reduced
sample volume limits the lowest size class of investigated
MPs mostly to 300–350 μm (Filella 2015). Thus, some
researchers used bottles to take bulk samples of the water
surface (Dubaish and Liebezeit 2013; Barrows et al. 2017),
which, however, results in small sample volumes.
Nevertheless, sampling lower size ranges, Barrows et al.
(2017) found MP concentrations were several orders of magnitude higher in bottle samples than manta samples. To
obtain larger sample volumes, others took several bottles or
buckets of surface water and concentrated the material on
filters with smaller mesh sizes on board (hand-nets; Chae
et al. 2015; Kang et al. 2015). Moreover, contamination
issues through high air exposure times during a manta trawl,
as well as filtering samples on board, motivated researchers
to further develop pumping systems (Desforges et al. 2014;
Lusher et al. 2014; Enders et al. 2015). One of the first studies comparing different methodologies for the same size
class (300–5000 μm) was conducted by Setälä et al. (2016)
comparing their custom-made pump to manta trawls.
Preliminary results from the pump (collecting surface water
in a depth of 0–0.5 m) did not significantly differ from the
results obtained by the manta net. Another interesting solution to decrease sampling effort has been published by Edson
and Patterson (2015). They designed an automated sampling
device (MantaRay), which automatically pumps sea surface
water at a depth of 30 cm, while drifting through the water.
Thereby, particles are concentrated on a filter and 28 successive samples can be taken. For the prototype, 500 μm stainless steel sieves were used. Such an instrument can decrease
sampling effort and airborne contamination, which is often a
challenge when conducting trawls. One drawback could be
the autonomous operation of the MantaRay, which limits the
control over the area sampled. Moreover, an optical sensor is
implemented to ensure that only water containing particulate
matter is filtered. Thereby, especially small MP particles
could be overlooked so the influence on the obtained results
must be further evaluated.
Independent of the applied method, decreasing mesh
sizes will increase the content of organic and inorganic material, which could lead to smaller sample sizes as meshes will
become clogged faster, but also to increased sample preparation time in the laboratory. In any case, negative controls
should be run, as most of the used methods may contain
polymer materials which are a further source for
contamination.
T. Hamm et al.