Heard D.E. (editor) Analytical Techniques for Atmospheric Measurement
Подождите немного. Документ загружается.
12 Analytical Techniques for Atmospheric Measurement
updating, through improved kinetic measurements in the laboratory. A powerful method
is to use the model to calculate atmospheric composition for our current atmosphere or
past trends in key species, and compare these to field measurements. Care must be taken
to choose an appropriate model to compare with an appropriate measurement (usually
defined by the lifetime of the trace gas or aerosol).
A commonly used molecule to test the accuracy of chemical mechanisms in a variety
of environments is the hydroxyl radical, OH. Its short chemical lifetime, due to its high
reactivity, means that its budget (and hence concentration) is only controlled by local
in situ chemistry, and not by transport processes. Hence zero-dimensional or box models,
which consider a box in which the sample is well mixed, can be used to describe the
chemistry of OH under specific conditions without having to incorporate transport into
or out of the box. The rate of change of OH concentration is given by:
dOH/dt = POH −LOH = POH −OH
i
k
i+OH
i (1.2)
where i is the concentration of a given sink (photolysis or heterogeneous losses of OH
are not important loss terms for OH). OH is generated and destroyed quickly and so
its concentration rapidly assumes that of a steady state dOH/dt = 0 with a balance
between production and loss, and hence
POH = LOH
OH = POH/
i
k
i+OH
i (1.3)
Comparison of measured values in the field with OH calculated from Equation 1.3 will
provide a sensitive test of how well a mechanism has described the production and loss
terms (i.e. the chemistry!). The lifetime of OH is less than 1 second, and hence is made
within a few metres of the inlet of an instrument designed to measure OH. As input, the
model requires the concentration of all sinks, i, to be measured, as well as all terms
contributing to the production term P(OH) (including radiation), ideally within a few
metres of the OH instrument inlet. In reality many of the species required as input
either cannot be or are not measured (e.g. O
1
D HO
2
), and must be calculated, but
themselves are coupled closely to the concentrations of many other species. In general,
if the number of chemical species in the mechanism is N , then N differential equations
must be solved simultaneously to calculate [OH]. There are now many examples of
model comparisons with OH measurements from a variety of platforms. The agreement
is variable, depending on the environment under study, pointing to inadequacies in our
understanding of the underlying chemistry. For remote environments, where the air is
clean and relatively few reactions and species are required to describe the chemistry, there
is now good agreement between measured OH and concentrations predicted using box
models, as shown in Figure 1.5 for the data taken during a field campaign at Mace Head,
Ireland, in 2002. If the chemical mechanism were to be used in a chemistry-transport
model (CTM) to predict the future composition in remote areas (e.g. of ozone), then we
can be more confident that the results are accurate.
Field Measurements of Atmospheric Composition 13
Figure 1.5 Model–measurement comparison for OH radicals for 9–10 August 2002 during the NAMBLEX
campaign at Mace Head, Ireland. The measurements were made in clean westerly air using the Leeds
FAGE instrument (laser-induced fluorescence at low-pressure), and have been averaged up to 15 min
time intervals to facilitate comparison. For the model calculation, mechanisms with different degrees
of chemical complexity were used. The ‘clean’ model was constrained to reaction of OH with CO
and CH
4
only, the ‘full’ model also included 23 hydrocarbons and chloroform, and the ‘fulloxy’ model
also included 23 hydrocarbons, chloroform, acetone, acetyaldehyde and methanol. The calculated OH
concentration decreases as more and more of its sinks are included in the mechanism.
1.2 Scope, structure and content of this book
1.2.1 Scope and structure of the book
This book will primarily address instruments used to make quantitative measurements
of the chemical composition of the atmosphere. Instruments to measure the structure of
the atmosphere – for example, micro-meteorological determinations of the instantaneous
three-dimensional wind velocity (although important in order to allow interpretation of
chemical data) or the measurement of deposition velocities to the earth’s surface – are
not covered. However, as the majority of atmospheric processes in the atmosphere are
initiated by the absorption of sunlight, at a rate dependent upon the rate of photolysis
of certain trace gases, the book includes a chapter on techniques to measure photolysis
frequencies directly or to measure radiative properties of sunlight to enable the calculation
of photolysis frequencies. Also, the book will concentrate on real-time in situ methods,
or those sampling a long path through the atmosphere, and that have short integration
times on the order of minutes or less. Methods that require the accumulation of a sample,
perhaps via a denuder, will typically not be covered.
The chapters are organised chapter-wise according to experimental technique (e.g.
absorption, fluorescence, chromatography, mass spectrometry, etc.), rather than by class
of species measured. It is realised that instruments may be highly specific for the
measurement of one species only – and under a particular technique there may be a
large number of instrument types. Clearly it is not possible to describe all of these, and
14 Analytical Techniques for Atmospheric Measurement
examples selected will be the most commonly used variants. Each chapter will contain
the following common elements:
(1) The fundamental physical principles of each technique will be covered, together
with a description of apparatus commonly deployed in the field, which may include
a variety of platforms to probe different regions of the atmosphere. Sampling is
given particular attention, as reactive or labile species may change their chemical
composition during sampling and prior to detection.
(2) The detection limits, accuracy, precision, sources of noise, degree of selectivity and
possible interferences for monitoring of different species using a given technique will
be discussed.
(3) A flavour of field-data obtained using the technique, including examples of time-
series, and key breakthroughs that have resulted from field observations of key speices.
In some cases comparison with model calculations are included to demonstrate how
our understanding of the atmosphere can be probed via the use of field measurements.
(4) Use of the technique, where appropriate, in routine air quality monitoring, perhaps
as a network of instruments that are relatively inexpensive to set up and operate, and
which can be run continuously with little or no user interaction.
(5) A discussion of quality control, which is of paramount importance as field observa-
tions are increasingly used to support legislation controlling emission of substances
harmful to health or the environment. Calibration of instruments and intercom-
parison of methods are important components of quality control and are covered
extensively in each chapter.
(6) A description of the future requirements and the latest developments involving a
given technique.
(7) Suggestions for further reading in general texts, as well as a comprehensive bibliog-
raphy of the research literature covering the historical development of the technique,
descriptions of state-of-the-art instruments used worldwide, and examples of their
deployment and results obtained. It is hoped that the book will provide a one-stop
shop to quickly obtain information, both basic and in detail, for all the commonly
used analytical methods for atmospheric measurement.
Of course, a particular atmospheric species can sometimes be measured by several
methods, with some being more fit-for-purpose than others for a particular application
or deployment platform. The organisation of the book via technique is less convenient
for readers who are interested in learning about alternative methods for detection of a
given species. In Section 1.3.3, atmospheric measurements are organised according to
class of species, with detailed cross-referencing to sections of various chapters of the book
that cover detection of that species via various methods. A brief discussion of the pros
and cons of the methods available is given for some key species.
1.2.2 Previous texts describing methods for determining
atmospheric composition
There are very few books dedicated to the measurement of atmospheric composition,
and even fewer that are current at the time of writing. A few books in edited series
Field Measurements of Atmospheric Composition 15
volumes cover measurements in the atmosphere with an emphasis on the techniques
used. For example, Measurement Challenges in Atmospheric Chemistry (Newman, 1993)
is the result of a symposium at an American Chemical Society meeting held in 1990, and
contains a series of technically detailed chapters organised according to the measurement
of a variety of key trace species. Selective Detectors (Sievers, 1995), applicable to the
determination of atmospheric composition is also rather specialised. Handbook of Air
Pollution Analysis (Perry & Young, 1977) is one of the few books dedicated to the
measurement of atmospheric trace gases. The structure is largely organised by class of
species, but there are chapters on general issues such as inlets and sampling methods,
as well as remote monitoring techniques and the planning and execution of an air
pollution study. However, the book was published almost 30 years ago, and hence the
methods described are somewhat antiquated, and many species at that time had not been
recognised as important to measure in the atmosphere. In 2003, a synthesis of a decade of
tropospheric chemistry research was published by the International Global Atmospheric
Chemistry Project (IGAC, report completed in 2001) (Brasseur et al., 2003), and contains
a succint discussion of advances in field measurements in both the gas and aerosol
phases, together with the results from field campaigns in the context of atmospheric
photooxidants and tropospheric aerosols. Methods for measurements of atmospheric
trace gases are reviewed by several authors for selected species in Calvert (1994).
There are also a number of advanced undergraduate textbooks that contain very
useful descriptions of the fundamental physical methods of techniques used to make
composition measurements, together with some details on the instruments and platforms
themselves, and examples of field data and their interpretation (Brasseur et al., 1999;
Finlayson-Pitts & Pitts, 2000). More specialised review articles are cited in individual
chapters of this book, but there are a number of articles that review a wide range of instru-
mental methods rather than just for individual techniques or class of species. A review of
a very wide range of instrumentation and measurement techniques for ground-based and
airborne field studies of gas-phase tropospheric chemistry can be found in Clemitshaw
(2004). The coverage of techniques and their deployment is very comprehensive indeed,
but due to the nature of the review, each method is covered fairly briefly and does not
discuss the fundamental principles upon which the methods are based. In a short review,
Roscoe and Clemitshaw provide a selective view of the past, present (1997) and future
measurement techniques in gas-phase tropospheric chemistry (Roscoe & Clemitshaw,
1997). As discussed in Section 1.8, there have been a number of large intensive field
campaigns, and for some the detailed results are disseminated in the open literature via
special issues of journals, which may contain up to 20–30 papers in succession, many of
which contain quite detailed accounts of field instruments for composition measurements
(see Further reading). Encyclopaedia of Atmospheric Sciences (Holton et al., 2003) (2780
pages) is a comprehensive collection of review articles that covers all of atmospheric
science, and several of the articles discuss methods for atmospheric composition measure-
ments (for example, (Clemitshaw, 2003; Heard, 2003).
1.2.3 Content of the book: Summary of individual chapters
Chapter 1 provides an introduction to the determination of atmospheric composition,
and an overview of the various techniques used to measure trace gases, aerosols and
16 Analytical Techniques for Atmospheric Measurement
radiation. As each chapter covers the commonly used techniques, it is inevitable that some,
although important, have not been covered, and this chapter (Section 1.5) provides a brief
description of methods not covered elsewhere in the book. In order to assist navigation
through the remainder of the book, and for reference, detailed tables are provided
according to experimental technique and by class of trace species measured. The various
platforms available for the deployment of instruments are described, including satellites,
with examples. Other issues, which are recurring themes in later chapters, are also briefly
introduced, for example the calibration of instruments, detection limits, instrument
accuracy and precision, and intercomparison of various techniques. The general require-
ments when designing a major field campaign are also discussed, examples given of major
field campaigns mounted around the time of writing, and a case study is provided of the
instrument portfolio deployed during a campaign at Mace Head, Ireland. A discussion is
also given of long-term measurements, as well as monitoring networks, which provide the
backbone of data which regulatory bodies rely on heavily to direct future policy and that
provide evidence about changes in our atmosphere and climate. There is also discussion
of the use of simulated environments in outdoor environmental chambers, where the
composition of the sampled air can be carefully controlled. Finally, the chapter concludes
with a brief discussion of the future directions for field measurements of atmospheric
composition.
Chapters 2–4 cover measurement techniques that involve excitation of a molecule
to an excited state through absorption of a photon of light. The source of radiation
can be natural, for example the sun, moon, stars, or the earth itself, or artificial, for
example lamps or lasers. In Chapters 2 and 3 the concentrations of atmospheric species
are derived using the Beer–Lambert law from measurement of the amount of light
absorbed. In Chapter 2, Alan Fried and Dirk Richter describe how the excitation of
vibrations within molecules via the absorption of infrared (IR) radiation is used to
quantitatively measure the concentration of atmospheric molecules. The method is widely
applicable over a range of spatial scales. In Chapter 3, John Plane and Alfonso Saiz-
Lopez describe how the excitation of electrons within molecules via the absorption
of visible and UV radiation is used to measure the concentration of the molecules.
Although similar in concept, measurement of the absorption of IR or UV/visible
radiation requires very different approaches. The converse of absorption is emission
of radiation from excited states, and, in Chapter 4, Ezra Wood and Ronald Cohen
describe the use of fluorescence from electronically excited atomic or molecular states
to monitor atmospheric species. A major advantage is the observation of photons on a
null background, rather than the measurement of a very small absorption of radiation
as in some cases, but set against this is the absence of a convenient fundamental law
to relate signal to concentration, and calibrations must be carried out. The use of lasers
as radiation sources is commonplace, their narrow spectral linewidths and wavelength
tunabilty making them ideal for highly selective molecular excitation. The observation
of IR or microwave emission from vibrationally or rotationally excited molecules as
a method of monitoring is used from space-borne or high altitude platforms, and is
covered briefly in Section 1.5.4. Furthermore, Section 1.4.8 discusses the use of satel-
lites for global measurements of trace species, using remote sensing techniques based on
differential optical absorption spectroscopy of scattered or transmitted sunlight, or the
detection of radiation in the IR and far-infrared/microwave regions emitted by excited
molecules.
Field Measurements of Atmospheric Composition 17
Chapter 4 is restricted to the generation of fluorescence via direct optical excitation,
rather than fluorescence (chemiluminescence) generated as a result of an electronically
excited state being produced as the product of an exothermic chemical reaction or an
inelastic collision involving energy transfer from another molecule. The chapter does
not consider fluorescence detection following a previous derivatisation or chromato-
graphic separation of the desired analyte (covered in Chapters 7 and 8, respec-
tively), but does discuss fluorescence detection following simple chemical conversion,
for example HO
2
detection via HO
2
+NO → OH +NO
2
with detection of OH by
LIF, or thermolysis, for example N
2
O
5
−→NO
3
+NO
2
with detection of NO
2
by LIF.
N
2
O
5
has also been monitored in the atmosphere via detection of the NO
3
product,
using cavity ring-down spectroscopy (Chapter 3). Free-radical species, for example BrO
and ClO, have been quantified in the atmosphere through their conversion to Br
and Cl atoms, with detection by resonance fluorescence at vacuum UV wavelengths
(Chapter 4).
Chapter 5 by Jonathan Williams is dedicated to the monitoring of atmospheric species
following their ionisation, with selectivity achieved using subsequent separation by mass
prior to detection. Ions can be spatially controlled and detected with exquisite sensitivity,
and mass spectrometry is the most widely applicable of all methods. The chapter is
primarily concerned with methods of ionisation and the subsequent separation by mass,
and the detection of ions, rather than with any prior separation, for example via chromato-
graphic methods (Chapter 8), that can lead to additional selectivity. Ionisation can be
achieved through collisions with electrons or ionising radiation, or through ion–molecule
reactions with positively or negatively charged reagent ions, with detection of either
positively or negatively charged product ions. An important reagent ion is the hydronium
ion H
3
O
+
, which can be used to protonate a wide range of species. Very high resolving
power mass spectrometers can distinguish molecules containing different isotopes within
a given molecule, and measurement of isotope ratios enables much to be learnt about the
sources and chemistry of that molecule. Indeed isotope ratios for molecules trapped within
ice-cores are used as proxies for the temperature of the atmosphere going back millions of
years.
In Chapter 6, by James Allan and Hugh Coe, methods for the measurement of the
physical and chemical properties of atmospheric aerosols are covered. Aerosols come in a
wide variety of sizes, shapes and chemical composition, and in order to understand their
toxic properties, their ability to absorb or scatter radiation, or their ability to act as sinks
(via surface uptake) for gaseous intermediates, it is crucial to have a detailed knowledge
of the number density and chemical composition as a function of size. Understanding,
and hence being able to predict, how aerosols are formed, and subsequently grow, with
associated changes in composition, is crucial to understanding their role in, for example,
global warming and the toxicity of urban air pollution. A focus of Chapter 6 is the
discussion of mass spectrometric methods to determine the chemical composition of size-
segregated aerosols. A major difference compared with the detection of gaseous species
(Chapter 5) is the requirement to volatilise the aerosols prior to or at the same time as
ionisation, and to take account of the different efficiencies of volatisation as a function
of composition.
As mentioned above, electronically excited species can be generated as a result of
exothermic chemical reactions, in either the gas or the liquid phase, or occurring upon
18 Analytical Techniques for Atmospheric Measurement
surfaces, or via energy transfer from collisions with other molecules, and Chapter 7, by
Andrew Weinheimer, describes methods relying on the observation of chemilumines-
cence. As for directly excited fluorescence, the method is very sensitive on account of
a null background, but has the advantage of not requiring a light source, rather the
mixing of another reagent with the analyte of interest. Chemiluminescence detectors
are very common, due to their relative simplicity and low expense, and are deployed
very widely, for example in the detection of nitric oxides. Chapter 7 also covers
techniques that convert the analyte to a species that can be detected via other methods,
but mainly restricted to chemiluminesence or fluorescence. The conversion may be
chemical, either in the gas phase (e.g. peroxy radicals via a chemical amplifier) or in the
liquid phase following dissolution in a scrubber (e.g. HCHO), or following photolytic
conversion (e.g. NO
2
→ NO). Chemical conversion of analytes is also covered in other
chapters; for example, HO
2
can be converted easily to OH via the addition of NO,
and thus HO
2
detection is considered together with OH in Chapter 4 on fluorescence
methods.
A well-established method of separating analytes is chromatography, and when
combined with a variety of detectors, is widely used for measurements of atmospheric
composition. In Chapter 8, Jacqueline Hamilton and Alastair Lewis describe a number
of gas and liquid chromatographic systems in common use, giving a comprehensive
account of practical issues for their operation, including sampling methods, separation
strategies, types of detectors, instrument calibration and quality control. For some
detection methods, for example mass spectrometry, the details of the fundamental
principles and field instrumentation are covered in other chapters, and the emphasis
in Chapter 8 is on the separation of analytes and the coupling between chromato-
graphic column and detection system. As well as the sampling from the gas phase,
the composition of aerosols can also be analysed using chromatographic methods,
with collection of material on filter paper by high-volume sampling, followed by
extraction or sample volatisation. For polluted environments, containing many thousands
of volatile organic compounds, conventional single-column chromatography is only
able to separate and identify a small fraction of species. Developments in the use
of multidimensional chromatography (the use of more than one column) to further
separate complex mixtures, are given, in particular when coupled to a time-of-flight
mass spectrometer, which gives an excellent time response and enables a further
dimension of separation to be achieved. Chromatographic methods have been applied
widely to study a range of atmospheric problems, and a selection of examples is
given.
The chemistry of the atmosphere is driven to a large extent by reactive species photo-
chemically generated following the interaction of atmospheric constituents with radiation
of sufficient energy to break chemical bonds. An example is the hydroxyl radical,
the most important atmospheric oxidant, which is mainly generated by the photolysis
of ozone to generate electronically excited oxygen atoms, followed by reaction with
abundant hydrogen atom donors (e.g. H
2
O vapour). The composition of the atmosphere
is greatly influenced by photochemical processes, and in Chapter 9, Andreas Hofzumahaus
describes techniques used to determine the rate (or frequency) at which molecules are
photodissociated in the atmosphere. The photolysis frequency of a molecule at a given
wavelength (to generate a specific product) is dependent upon the product of three main
Field Measurements of Atmospheric Composition 19
parameters: (1) the absorption coefficient, (2) the photodissociation quantum yield to
generate that product, and (3) the amount of light available to excite the molecule in
the first place. It is necessary to know these for all wavelengths over which the molecule
absorbs light, and both (1) and (2) are also pressure and temperature dependent. For
many molecules (1) and (2) have been determined in the laboratory, and the amount
of direct radiation from the sun can be calculated under clear sky conditions for any
location/altitude/time on the planet. However, molecules are photolysed not only by
direct solar radiation but also by radiation scattered by air molecules, aerosols and
clouds, or by radiation reflected from the ground, and (3) is therefore very difficult
to calculate. In Chapter 9 several methods to determine the rate of photolysis are
described. Measurement is either taken directly using an actinic flux filter radiometer,
equipped with a suitable optical filter/detector combination, or by a chemical actinometer,
whereby the concentration of a photoproduct is measured as a result of photodissoci-
ation of a known precursor. Alternatively, the solar flux can be measured as a function
of wavelength (parameter (3) above) and values of (1) and (2) – measured in the
laboratory – are used to calculate the rate of photolysis. A number of practical systems
are described, and where appropriate, their calibration, together with their deployment
on various platforms in atmospheric chemistry field experiments, is also described. The
temporal and spatial variations of photolysis frequencies are illustrated for a number of
important photolabile species, together with the controlling influence these rates have on
the concentration of photochemically generated intermediates, for example the hydroxyl
radical.
1.3 The measurement of atmospheric composition
1.3.1 Units of concentration
Concentrations of the most important atmospheric trace gases range over more than 10
orders of magnitude. Concentrations are expressed in several units, and it is useful to
define their relationship. The most common method to report levels is the mixing ratio,
which is the ratio of the number of molecules of the species of interest to the number
of molecules in the volume of air of interest. An advantage of using mixing ratios is that
even if the temperature or pressure of the air changes, the mixing ratio is conserved.
Examples of common units of mixing ratio are the following: ppmv (parts per million
by volume)–1in10
6
molecules of air; ppbv (parts per billion by volume)–1in10
9
molecules of air; pptv (parts per trillion by volume)–1in10
12
molecules of air; and
ppqv (parts per quadrillion by volume)–1in10
15
molecules in air.
Another unit of concentration, often used by legislators, is micrograms of species per
cubic metre of air gm
−3
, but for comparisons it is necessary to correct values to a
standard temperature and pressure. For NO
2
, at 101.3 kPa (1 atm at sea level), and 293 K,
1 ppb is equivalent to 1913 gm
−3
, whereas for NO it is 1248 gm
−3
as it is a lighter
molecule. It is not possible to express particulate matter concentrations as a mixing ratio,
and their measurements are given in gm
−3
.
20 Analytical Techniques for Atmospheric Measurement
The number density of an atom or molecule (X) in the gas phase, commonly referred
to as concentration, [X], is usually reported in units of molecules cm
−3
, and is linked to
the total pressure, p (in Pa), and its mixing ratio,
X
, by the expression:
X = N
a
X
p/RT ×10
−6
(1.4)
where N
a
= 6022 ×10
23
mol
−1
is the Avogadro constant, the factor 10
−6
is to convert
from the SI unit of m
−3
to cm
−3
and R is the gas constant 8314 J mol
−1
K
−1
. At sea
level, using air as an example (i.e. mixing ratio
air
= 1), p = 10
5
Pa, T = 298 K, gives
air = 243×10
19
molecule cm
−3
. As another example, a measurement of HO
2
radicals
at high altitude (p = 025 atm = 25 ×10
4
Pa, T = 200 K) yielded a mixing ratio of
HO
2
= 4 pptv, which corresponds to a concentration of:
HO
2
= 4 ×10
−12
×N
a
p/RT ×10
−6
= 362 ×10
7
molecules cm
−3
1.3.2 Selection criteria for instruments
There are a number of selection criteria to consider when it comes to choosing the most
suitable instrument for a given task.
(a) The universality of the method is. It is a clear advantage to be able to measure several
species simultaneously.
(b) What is the sensitivity of the instrument? Is this significantly less than the atmospheric
concentration of the desired analyte?
(c) Is the instrument specific for a given species, and are there interferences from other
species? Optical methods have an advantage in that spectra can be used to fingerprint
a given molecule, but it needs to be proven that the spectra are due to ambient
sources of the species and not generated within the instrument.
(d) Sampling efficiency needs to be good, and well understood, as some highly reactive
molecules (e.g. free-radicals like OH) are destroyed on surfaces, and others (e.g. NH
3
,
HNO
3
are sticky, and transmission may vary according to the age and condition of
the inlet. The measurement of HNO
3
is an example of where specialised inlets have
had to be designed to tackle this problem, especially for aircraft sampling (Ryerson
et al., 1999).
(e) The accuracy and precision of the instrument is important (see Section 1.6.1),
and there are different requirements depending on the application. High accuracy
is required in order to bring together data from a global network of monitoring
stations, whereas high precision is required to measure fluxes of gases using very short
integration periods or to monitor long-term trends for an individual trace species.
(f) Integration or averaging period (response time) is another important factor, and the
requirements vary widely, depending on the chemical lifetime and hence temporal
variability in the atmosphere. Hydroxyl radical measurements ideally require time
responses of a few seconds, whereas measurements of CFCs can have much longer
averaging periods.
Field Measurements of Atmospheric Composition 21
(g) Spatial resolution is one of the most important criteria, again because of the widely
differing atmospheric lifetimes and hence spatial variability away from source. If one
is interested in understanding the formation and destruction of the OH radical, for
comparison with a constrained box-model, then ideally all the sources and sinks
should be measured as close to the OH instrument inlet as possible. If, however, the
aim is to compare with the calculations of a global CTM, with a grid-square size
of 100km ×100 km, then a measurement representative of the average within that
footprint is required, perhaps provided from a satellite or a remote sensing device, a
mobile laboratory or aircraft, or a network of ground-based sensors. The two major
classes of methods are in situ and remote sensing, the former sampling from a single
point in the atmosphere (normally an inlet but this could be a small volume defined
by a folded optical path), the latter normally using a distant source of light, for use
in absorption spectroscopy. As a general rule, remote sensing methods have inferior
spatial resolution. In the limit, one only obtains a total column measurement of the
species of interest between the light source (e.g. the sun, or the earth’s surface for
scattered light) and the detector. However, in some cases the spatial resolution of
remote sensing methods can be very good, for example in Light Detection and Ranging
(LIDAR) methods (Section 1.5.1) when a pulsed source of light is used. Care must be
taken when comparing measurements of the same species taken with instruments that
sample with different spatial resolutions, or deciding upon the optimum dataset to
constrain a photochemical model. For example, during the NAMBLEX field campaign
(Section 1.8.2), NO
3
radicals were measured using an in situ method (broadband
cavity ring-down spectroscopy, see Chapter 3), with the instrument housed on the roof
of a container near the shore, and also by differential optical absorption spectroscopy
DOAS (Chapter 3), using a path length of 2 ×42 km, with the retro-reflector on
an offshore island, and the light source and spectrometer housed in a building on
the shore close to the CRDS instrument. At night, concentrations of NO
3
from the
DOAS instrument were considerably higher than from the CRDS instrument, but this
observation could be explained by the emission of small concentrations of NO, locally,
probably from the soil, that react quickly with NO
3
, reducing its concentration, but
which are not present over the open ocean.
(h) The method should be non-intrusive with respect to the operation of nearby instru-
ments, not blocking other inlets from the prevailing wind. The structure supporting
the instrument should also be taken into consideration. Remote sensing methods (e.g.
with lasers or other light sources) are ideal in this regard but suffer from problems
of spatial resolution.
(i) Weight, size, requirement of calibration, electrical power requirements, autonomy
(does it need to be left for months unattended?) and ease of operation (are skilled
operators always available?) are other operational criteria, and will vary mostly
according to the deployment platform.
(j) Ease of data reduction is a key criterion. Having to wait months for validated data is
a disadvantage.
(k) Is the instrument available commercially? The cost of an instrument is not just derived
from the hardware but also from the work-years of effort required in its development,
and in its upkeep and maintenance. If the running costs and requirement of a skilled
operator are taken into account, a commercial instrument may be a better long-
term solution, even if the performance in some of the other criteria is not quite as