2
Osvaldo E. Sala, Robert B. Jackson, Harold A. Mooney, and Robert W. Howarth
a challenge. Methods change quickly and are
adapted to the needs of researchers locally. This
local need sometimes conflicts with the need of our
discipline to synthesize results across systems. If
every researcher uses a different technique to measure an ecosystem process, we may be unable to
compare results quantitatively across studies. This
inability hinders some aspects of ecosystem science
today and limits our ability to synthesize data for
use by other disciplines. Another goal of this book,
then, is to clarify differences and similarities among
methods and suggest the extent to which results collected using different tools could be compared.
The book discusses tools to address similar questions in aquatic and terrestrial ecosystems. Differences between methods used in these two ecosystem types stem from differences in their natures and
differences in the histories of the disciplines.
Aquatic and terrestrial scientists have often worked
in isolation as a result of the barriers imposed by
the structure of institutions, such as funding agencies, universities, or scientific societies. Although
structure is necessary and fosters efficiency, it
sometimes leads to duplication and intellectual isolation. This book therefore attempts to present, in
one volume, methods for aquatic and terrestrial
ecosystems because we think that this intellectual
interaction will be fruitful. Developments in one
area may stimulate ideas in the other. The book
attempts to cover the breadth of aquatic and terrestrial ecosystem science in a structure that is roughly
parallel, but the reader will undoubtedly find gaps.
For example, description of the manipulations of
nutrients in terrestrial ecosystems does not have an
aquatic counterpart in the book, and the analysis of
animal manipUlations belowground lacks a counterpart of the aboveground manipulations. Similarly, the book describes most types of manipUlations of the abiotic environment but it misses some
important manipulations, such as carbon dioxide.
Ecosystem science is a complex discipline spanning different spatial and temporal scales, from the
square meter to the region and from minutes to centuries. As temporal and spatial scales change, not
only the researcher's perspective changes but also
the nature of the process and its controls change.
Important controlling variables at one scale become
white noise at a coarser scale, and what may have
been considered constant now becomes a driving
variable (Allen and Starr 1982; O'Neill et al. 1986).
For example, in the case of carbon fixation, the
same process can be studied at the scale of patches
with a diameter of a few meters, at the landscape
level, or at the regional level. Whereas differences
in soil texture, through changes in water and nutrient availability, account for most of the variability
in terrestrial carbon fixation at the finer levels of
the hierarchy, they are overshadowed by climate
variability at coarser scales (Sala et al. 1988). In
contrast, while climate could be easily considered
as a constant at the finer scale, it is certainly one of
the major determinants of primary production patterns at the regional scale.
Important basic and applied questions in ecosystem science need to be addressed at each of these
levels. The rate of primary production at the scale
of a paddock or lake and its variability in space and
time are critical tools when managing commercial
fish stocks, herds of grazers, or forests stands. Carbon balance at the regional level is currently an
important item in global negotiations for carbon
trade among countries. Different tools and methods
are necessary to address the same kind of questions
at different scales. This book attempts to describe
methods to address this nested hierarchy of ecosystem questions.
Chapters in this book do not prescribe a method
or technique but emphasize tradeoffs among tools.
A clear description of advantages and disadvantages will aid scientists in how to test hypotheses
and answer questions. When discussing tradeoffs,
the authors point out not only issues of scale but
cost-accuracy or accuracy-understanding tradeoffs. Methods that yield the best results at a fine
scale may not be the most appropriate at the regional scale. Similarly, the errors associated with
methods sometimes are negatively associated with
its costs. Consequently, the question and the accuracy needed for the estimate of the variable help to
determine the best methodological option. Some
chapters make a thorough analysis of the errors associated with each technique. In some cases, methods that provide estimates closest to the real value
are not necessarily those that are most detailed (for
example, see Chapters 2 and 4). A very detailed
estimate of a large number of process components
certainly yields a better understanding of the role
of the different pieces. This bottom-up approach
and the error inevitably associated with aggregation
yield, in many cases, results that are farther away
Osvaldo E. Sala, Robert B. Jackson, Harold A. Mooney, and Robert W. Howarth
a challenge. Methods change quickly and are
adapted to the needs of researchers locally. This
local need sometimes conflicts with the need of our
discipline to synthesize results across systems. If
every researcher uses a different technique to measure an ecosystem process, we may be unable to
compare results quantitatively across studies. This
inability hinders some aspects of ecosystem science
today and limits our ability to synthesize data for
use by other disciplines. Another goal of this book,
then, is to clarify differences and similarities among
methods and suggest the extent to which results collected using different tools could be compared.
The book discusses tools to address similar questions in aquatic and terrestrial ecosystems. Differences between methods used in these two ecosystem types stem from differences in their natures and
differences in the histories of the disciplines.
Aquatic and terrestrial scientists have often worked
in isolation as a result of the barriers imposed by
the structure of institutions, such as funding agencies, universities, or scientific societies. Although
structure is necessary and fosters efficiency, it
sometimes leads to duplication and intellectual isolation. This book therefore attempts to present, in
one volume, methods for aquatic and terrestrial
ecosystems because we think that this intellectual
interaction will be fruitful. Developments in one
area may stimulate ideas in the other. The book
attempts to cover the breadth of aquatic and terrestrial ecosystem science in a structure that is roughly
parallel, but the reader will undoubtedly find gaps.
For example, description of the manipulations of
nutrients in terrestrial ecosystems does not have an
aquatic counterpart in the book, and the analysis of
animal manipUlations belowground lacks a counterpart of the aboveground manipulations. Similarly, the book describes most types of manipUlations of the abiotic environment but it misses some
important manipulations, such as carbon dioxide.
Ecosystem science is a complex discipline spanning different spatial and temporal scales, from the
square meter to the region and from minutes to centuries. As temporal and spatial scales change, not
only the researcher's perspective changes but also
the nature of the process and its controls change.
Important controlling variables at one scale become
white noise at a coarser scale, and what may have
been considered constant now becomes a driving
variable (Allen and Starr 1982; O'Neill et al. 1986).
For example, in the case of carbon fixation, the
same process can be studied at the scale of patches
with a diameter of a few meters, at the landscape
level, or at the regional level. Whereas differences
in soil texture, through changes in water and nutrient availability, account for most of the variability
in terrestrial carbon fixation at the finer levels of
the hierarchy, they are overshadowed by climate
variability at coarser scales (Sala et al. 1988). In
contrast, while climate could be easily considered
as a constant at the finer scale, it is certainly one of
the major determinants of primary production patterns at the regional scale.
Important basic and applied questions in ecosystem science need to be addressed at each of these
levels. The rate of primary production at the scale
of a paddock or lake and its variability in space and
time are critical tools when managing commercial
fish stocks, herds of grazers, or forests stands. Carbon balance at the regional level is currently an
important item in global negotiations for carbon
trade among countries. Different tools and methods
are necessary to address the same kind of questions
at different scales. This book attempts to describe
methods to address this nested hierarchy of ecosystem questions.
Chapters in this book do not prescribe a method
or technique but emphasize tradeoffs among tools.
A clear description of advantages and disadvantages will aid scientists in how to test hypotheses
and answer questions. When discussing tradeoffs,
the authors point out not only issues of scale but
cost-accuracy or accuracy-understanding tradeoffs. Methods that yield the best results at a fine
scale may not be the most appropriate at the regional scale. Similarly, the errors associated with
methods sometimes are negatively associated with
its costs. Consequently, the question and the accuracy needed for the estimate of the variable help to
determine the best methodological option. Some
chapters make a thorough analysis of the errors associated with each technique. In some cases, methods that provide estimates closest to the real value
are not necessarily those that are most detailed (for
example, see Chapters 2 and 4). A very detailed
estimate of a large number of process components
certainly yields a better understanding of the role
of the different pieces. This bottom-up approach
and the error inevitably associated with aggregation
yield, in many cases, results that are farther away