Understanding temporal response patterns and the underlying mechanisms that control them will be fundamental to making longer-term predictions of ecosystem response to a changing environment. As illustrated in Figure 2, a perturbation, such as elevated CO2 or a change in climate can move a system from one state (A) to another state (B), but the trajectory of the response may vary.
If the response is assumed to be approximately linear (i.e., a change in the perturbation results in a uni-directional change in the response over time; hypothetical line 1 in Figure 2), when, in fact, the response is non-linear or cyclic (hypothetical lines 2-6 in Figure 2), then extrapolations from measurements of the initial response may lead to false conclusions concerning the future state of the system. The following sections describe issues that should be considered when evaluating and modeling temporal patterns of ecosystem response to global change.
The direct effects of CO2 enrichment, warming, and changes in moisture on ecosystem processes are relatively well understood. Indirect effects, which will likely regulate long-term changes in ecosystem response, are more complex and will require considerably more effort to accurately predict and model because they can involve a complex web of interactions (Shaver et al. 2000). An example of direct vs. indirect effects is the influence of warming on soil respiration. In general, and up to a temperature optima, warming directly increases both autotrophic and hetero-trophic soil respiration (Rustad and Norby 2002). However, if higher temperatures increase evapotranspiration and thereby reduce soil moisture, then warming can indirectly result in a decrease in soil respiration. This was demonstrated by Rustad et al. (2001) who showed that experimental soil warming (either with electrical heating cables, infra-red heaters, or glasshouses) at 16 different research sites generally increased soil respiration (Figure 3). However, at one site (The Rocky Mountain Biological Laboratory, or 'RMBL' in Figure 3), a warming-induced decline in soil moisture resulted in lower rates of soil respiration in the heated plots compared to the controls (Rustad et al. 2001; Figure 3).
Interestingly, soil carbon also declined over time in the heated plots compared to the controls at the
Figure 2. Hypothetical trajectories as a system moves from state A to state B. The lines represent the following hypothetical responses: 1=a linear response, 2 = lag, 3 = acclimation, 4 = resource limitation, 5 = homeostasis, and 6 = threshold. (Diagram courtesy of Gus Shaver, TERACC Workshop, 2002.)
Figure 2. Hypothetical trajectories as a system moves from state A to state B. The lines represent the following hypothetical responses: 1=a linear response, 2 = lag, 3 = acclimation, 4 = resource limitation, 5 = homeostasis, and 6 = threshold. (Diagram courtesy of Gus Shaver, TERACC Workshop, 2002.)
Figure 3. Percent change in soil respiration at 16 ecosystem warming experiments (from Rustad et al. 2001).
RMBL (Saleska et al. 2002b, Figure 4). This decline in soil carbon was not due to an increased loss of carbon through soil respiration (since soil respiration had declined) but rather was due indirectly to a change in plant community dynamics, with a shift from forbs (characterized by high productivity) to shrubs (characterized by low productivity), and consequent declines in above and belowground plant detrital quantity and quality (deValpine and Harte 2001). Results from observations across an associated climate gradient, however, suggest that the temporal response patterns for this study will be even more complicated, and that the observed decline in soil carbon is a transient response that will eventually be reversed as lower quality detrital inputs from the increasingly dominant shrub species reduce soil respiration losses (Saleska et al. 2002a).
At a larger scale, direct effects of warming on snow and/or ice cover, LAI, and/or changes in disturbance frequency such as fire may alter local, regional, or even global albedo, or the fraction of incoming solar radiation that is reflected back to the atmosphere (Ingram et al. 1989; Betts et al. 1997; Lynch and Wu 1999; Betts 2000). Decreases in albedo will increase the radiation absorbed by the region which will amplify warming whereas increases in albedo will cause a greater amount of radiation to be reflected back to the atmosphere, and will thus have a negative feedback to warming.
'Lags' in response
Lags in response occur when some responses take longer to come to a new equilibrium with the environment than others because of either internal (e.g., life span, seed dispersal, vegetative propagation, etc) or external (e.g., fire, pathogens, etc.) factors (hypothetical line 2 in Figure 2). For example, Vetaas (2002) showed that although mature individuals of long-lived Rhododendron species were not able to survive outside the cold limit of their realized niche, they could survive and continue to reproduce vegetatively when planted outside their natural high temperature range. The future distribution of this species may thus show a lag in response to gradual increases in mean annual temperature. 'Lags' due to limitations to seed dispersal and changes in disturbance regimes have been suggested by Chapin and Starfield (1997) who modeled a lag of 150 -250 years in forestation of an arctic tundra following climatic warming due to (1) slow tree establishment and growth under slow climatic warming and (2) higher frequencies of fire and insect attack under more rapid climatic warming.
Acclimation is the often misused term that refers to a non-heritable, reversible change in the physiology or morphology of an organism in response to changing environmental conditions (Ricklefs 1990; hypothetical line 4 in Figure 2). Plants, for example, can acclimate to changing conditions by various mechanisms including changing enzyme concentrations (e.g., Maroco et al. 1999; Watling et al. 2000; Gesch et al. 2002), altering shoot:root ratios (e.g., Equiza et al. 2001; Kozlowski and Pallardy 2002; Horacio 2003; Matsuki et al. 2003), or changes in phenology (e.g., Campbell and Sorensen 1973; Adam et al. 2001). Evidence is accumulating that many ecosystem processes acclimate to elevated CO2 and warming at the physiological level, thereby reducing their sensitivity to these perturbations, and invalidating many predictions of future responses. Considerable effort must be made to (a) understand the mechanisms underlying physiological acclimation at the organism level and (b) incorporate acclimation into existing ecosystem models.
Three examples of physiological acclimation that have received considerable attention in recent years are the acclimation, or down regulation, of photosynthesis in response to elevated CO2, the acclimation of photosynthesis to elevated temperature, and the acclimation of autotrophic respiration to elevated temperature. Photosynthetic down regulation in response to elevated CO2 was initially reported in dozens of CO2 enrichment studies (e.g., Gunderson and Wullscheleger 1994; Luo et al. 1994; Drake et al. 1997; Egli et al. 1998; Rey and Jarvis 1998; Ziska 1998; Medlyn et al. 1999; Sims et al. 1999; Hymus et al. 2002b; Rogers and Ellsworth 2002) and was generally attributed to decreases in leaf nitrogen and ribulose 1,5-bi-phosphate carboxylase/oxygenase (Rubisco) which lead to declines in photosynthesis (Rogers and Humphries 2000). More recently, however, the role of photosynthetic down regulation has been questioned, and its prevalence, particularly in earlier pot or chamber studies has been attributed, in part, to root restriction within experimental pots (e.g., Stitt 1991; Farage et al. 1998), inadequate N supply (e.g. Webber et al. 1994; Drake et al. 1997; Kubiske et al. 2002; Ainsworth et al. 2003), or the age class of needles in conifers (Medlyn et al. 1999).
Photosynthetic acclimation to increased temperature including both shifts in temperature optima and uniform shifts across all temperatures, has been long recognized (e.g., Barry and Bjork-man 1980; Ferrar et al. 1989; Read 1990; Gunderson et al. 2000), and has been attributed to various factors including different thermal properties of key photosynthetic enzymes, different temperatures at which membranes are damaged, and differential thermal stability of photochemical reactions (Nilsen and Orcutt 1996). The acclimation of autotrophic respiration to elevated temperature has also been demonstrated (e.g., Kirshbaum and Farquhar 1984; Tjoelker et al. 1999, 2001; Atkin et al. 2000a,b; Will 2000; Griffin et al. 2002; Bolstad et al. 2003), and has been attributed variously to decreased number of mitochondria (Miroslavov and Kravkina 1991), decreased respiratory capacity per mitochondria (Klikoff 1966), limitations in substrate supply (Lambers et al. 1996), changes in the concentration of plant soluble sugars (Atkin et al. 2000a), changes in demand for respiratory energy (Atkin and Lambers 1998), and/or changes in enzymatic capacity (Atkin et al. 2002).
Although the acclimation of both photosynthesis and autotrophic respiration to warming is well established, the potential acclimation of 'soil respiration' (i.e., the combined respiration of roots and soil micro- and macro-biota) to warming is more controversial. Historically, dozens of studies have demonstrated strong positive relationships between soil respiration and temperature (for syntheses see Raich and Nadelhoffer 1989; Raich and Schlesinger 1992; Raich and Potter 1995; Kirschbaum 1996; Rustad et al. 2001), and soil respiration is typically and effectively modeled with an exponential or Arrehenius function (Rus-tad et al. 2000). Recently, however, the temperature dependence of soil respiration has been challenged by Luo et al. (2001) who suggested that soil respiration 'acclimates' to elevated temperature. They conducted a warming x grazing experiment in a tall grass prairie in Oklahoma, USA using overhead infra-red lamps and clipping, and reported a decline in the respiration quotient Q10 from 2.70 in the unheated, unclipped plots to 2.43 in the heated, unclipped plots, and from 2.25 in the unheated, clipped plots to 2.10 in the heated, clipped plots. However, a physiological mechanism for the acclimation of soil respiration to temperature has yet to be elucidated. This is in part because, unlike photosynthesis, soil respiration is not a single process but is instead the sum of the combined respiration of plants and the complex community of micro- and macro-heterotro-phic soil organisms, and includes several alternate chemical pathways. In addition, the direct effect of warming on soil respiration is complicated by a host of indirect effects, including warming-induced changes in above and belowground biomass, soil moisture, N mineralization, substrate quality and/ or quantity, and microbial community activity, biomass, and composition.
Given that gross primary productivity (GPP), aboveground respiration, and soil respiration represent three of the largest fluxes in the terrestrial global carbon cycle (estimated at ~120, 60, and 60 Pg C yr-1, respectively; Schlesinger 1997), it is imperative to understand if and to what degree these processes will acclimate to changing environmental conditions such as CO2 enrichment and global warming. Even slight changes in the direction and or magnitude of these fluxes could equal or exceed the annual input of CO2 to the atmosphere via combined fossil fuel combustion and land-use changes (estimated at Pg C yr-1), and could therefore significantly accelerate - or decelerate - the rate of atmospheric build-up of CO2, with consequent feedbacks to climate change.
Resource limitation/initial conditions
The sustainability of the magnitude and even direction of a response may be governed by the availability of resources which will be governed in part by initial conditions. For example, ecosystems with large stocks of relatively labile C may show a larger and more sustained increase in soil respiration in response to warming than an ecosystem with low initial labile C stocks, or an N-rich ecosystem may show a more sustained increase in photosynthesis and NPP under CO2-enrichment then a N-poor ecosystem. In either case, if the systems receive no new inputs of labile C or atmospheric or fertilizer N, the magnitude of the response will decline over time as either labile C or N are depleted (hypothetical line 2 in Figure 2). For example, at the Harvard Forest soil warming experiment, Peterjohn et al. (1994) initially reported an approximately 40% increase in soil respiration during the first six months of the experiment. However, the magnitude of this increase diminished over time such that after 10 years of warming soils at 5 °C above ambient, soil respiration rates in the heated plots were not significantly different than rates in the control plots (Figure 5; Melillo et al. 2002). One explanation is that labile C supplies were depleted during the course of the experiment suggesting resource limitation. The lack of a treatment effects
in the latter years of the experiment also lend support to the hypothesis that different carbon fractions have different temperature sensitivities, with labile carbon fractions (consistently predominantly of simple sugars and amino acids) being highly temperature sensitive but recalcitrant carbon fractions (consisting of more complex aromatic compounds) being relatively temperature insensitive (Liski et al. 1999; Giardina and Ryan 2000; Melillo et al. 2002; Gu et al. 2004).
Homeostasis is the 'maintenance of or 'return to' constant internal conditions in the face of a varying external environment (Ricklefs 1990). Classic examples include (a) the thermal regulation of homeotherms despite external fluctuations in temperature, (b) the ability of organisms to maintain their internal chemical composition despite fluctuations in the chemical content of their environment or food source, and (c) predator -prey cycles where as the population of prey increases so does the population of predators, thereby decreasing the population of prey and consequently the population of predators (hypothetical line 5 in Figure 2). Local, regional, and even global ecological systems also exhibit homeostatic behavior. An example is elevated atmospheric CO2 and the global carbon cycle. Within limits, as atmospheric CO2 increases, leaf level photosynthesis and NPP should increase, thereby removing CO2 from the atmosphere and stabilizing atmospheric CO2 concentrations. Concerns exist, however, that the current anthropogenic input of carbon to the atmosphere from fossil fuel combustion and land-use changes, particularly in combination with possible positive (rather than negative) feedbacks from warming-induced increases in the release of soil carbon to the atmosphere or decreases in albedo, may exceed the capacity of the earth's systems to maintain this homeostatic balance.
Buffers are mechanisms or attributes that allow systems to resist change in response to external perturbation or impact. In chemistry, solutions that contain a weak acid and its salt or a weak base and its salt, and which thereby can resist changes in pH, are called buffers. Similarly, ecological systems have certain attributes that allow them to resist moderate changes in environmental variables. Examples of ecosystem properties that may provide 'buffers' against impacts of CO2 fertilization and climate change may include soil C quality and quantity [i.e., systems with more protected, chemically stable C will be less vulnerable to soil C loss than systems with less stable C e.g. (Collins et al. 1997; Paustian et al. 1997, 2000; Six et al. 2000)], soil depth and water holding capacity (i.e., ecosystems with deeper soils with better water holding capacity will be less sensitive to fluctuations in precipitation than those with shallower soils with limited water holding capacity), albedo [i.e., ecosystems with higher albedo will reflect more solar radiation back to the atmosphere and will thus be less sensitive to warming than systems with lower albedo (e.g., Betts et al. 1997; Betts 2000; IPCC 2001; Berbet and Costa 2002)] and biodiversity [i.e., ecosystems with greater species or functional group diversity may be more resistant and resilient to environmental change than those with lower diversity (e.g., Naeem and Li 1997; Walker et al. 1999; Chapin et al. 2000; Ives and Cardinale 2004)].
A process is said to have a threshold if below that threshold there is either no change or proportionate change in the response of the process to a perturbation and above that threshold there is a dramatic, non-proportional response (hypothetical line 6 in Figure 2). Arnold et al. (1999) provide an example of the former, where, using laboratory incubations under controlled temperature and moisture conditions, they showed no difference in microbial biomass at gravimetric soil moisture contents between 120 and 320%, but a dramatic reduction of almost 95% of the microbial biomass when gravimetric soil moisture was decreased to 20%. They suggest a soil moisture threshold exists between 20 and 120% for their soils above which moisture is not limiting and temperature largely controls microbial biomass dynamics, and below which moisture is too low to sustain viable microbial biomass, regardless of temperature.
At a larger scale, thresholds also appear to exist in the climate system. Reconstruction of past climates, for example, show gradual changes in climate over geologic time scales, punctuated by dramatic changes in temperature and precipitation on time scales as small as decades (IPCC 2001). Examples include a 5-10 °C increase in temperature and a doubling of snowfall that occurred in Greenland over a period of 40 years following the last glaciation and the rapid transition from shrubland to desert that occurred in the Sahara approximately 5500 years ago (Rahmstorf 2002). The causes of these rapid changes are uncertain but may be associated with thresholds in ocean circulation and sea ice dynamics, or vegetation-induced changes in albedo (Rahmstorf 2002). Concerns exist, including those expressed by the National Academy of Science Committee on Abrupt Climate Change (2001) and by Gregory et al. (2004), that similar mechanisms will come into play such that CO2-induced global warming will lead to increased precipitation in high northern latitudes, which, combined with melting of the polar ice sheets, will increase freshwater input to the North Atlantic Ocean, leading to a precipitous reduction in the global ocean's thermohaline circulation, thereby shutting down the Gulf Stream, and resulting in decreases in temperature, particularly over much of Europe.
Ecosystem stoichiometry is based on principals of (1) the conservation of matter, (2) the stoichiom-etry of chemical reactions, and (3) the observation that plants, animals and even ecosystems are constructed of multiple elements in relatively fixed forms (Sterner and Elser 2002). Ratios between elements are therefore also relatively fixed, which puts constraints on element distribution and cycling, and implies that a change or disruption in the ecosystem- or global-scale cycle of one element, such as C, N, or P, will necessarily impact the cycling of other elements. For example a CO2 enrichment-induced increase in photosynthesis and NPP will require an increase in belowground nitrogen acquisition in order to maintain leaf C:N ratios within a relatively fixed range, and will thereby impose a change in the nitrogen cycle. Or, as pointed out by Nadelhoffer et al. (1999) and
Hungate et al. (2003), the amount of C that can be sequestered by an ecosystem with increasing N deposition will depend largely on whether the N is immobilized in bacteria (C:N ratios typically between 5 and 15) or soil organic matter (C:N ratios typically between 10 and 50), or whether the added N is taken up and stored in foliage (C:N ratios typically 30-100) or wood (C:N ratios typically >300). Results from the decadal-scale N fertilization experiment at the Bear Brook Watershed in Maine and the decadal-scale soil warming experiment at the Harvard Forest in Massachusetts both show that most of the added or warming-induced mineralized N is stored in soil organic matter with relatively low C/N ratios, thus limiting the potential for these systems to sequester large amounts of additional carbon (Nadelhoffer et al. 1999; Melillo et al. 2002).
The concepts of turnover rates and turnover times are fundamental to understanding and modeling ecosystem response to global change. Assuming a steady state, turnover 'rate' is defined as the net mass of a material entering or leaving a system or reservoir in a given time period (i.e., flux) divided by the total mass of the material present in that system or reservoir (i.e., pool; units are percent/ time period). Turnover 'time' is the inverse, or the total mass of a material in a system or reservoir (i.e., pool) divided by the net mass of the material going into or out of that system or reservoir over a given time (i.e., flux; units are time). Turnover times can also be interpreted as the mean life span of a system or component of a system (e.g. mean tree or root lifespans) or the mean residence time of material in a system or component of a system (e.g., mean residence times for greenhouse gases in the atmosphere and for the amount of carbon in a particular soil carbon pool).
Within the global change literature, concepts of turnover rates and times have been most frequently applied to greenhouse gas concentrations, above- and below-ground biomass pools, and carbon and nutrient cycles, and questions have arisen as to whether global change will alter the fluxes of material into or out of atmospheric, biomass or nutrient pools, or the pool sizes themselves. For example, over the long term, the amount of carbon that can be sequestered by an ecosystem will depend on both the size of the carbon pool in that system and its turnover time. More carbon can be stored in an ecosystem only if either the same amount of carbon is retained for a longer time (longer turnover times) or more carbon is added to the total pool than is lost from the pool (larger pool size). Elevated CO2 and warming will generally increase photosynthesis and will thus increase the flux of carbon going into an ecosystem. However, if this carbon is stored in labile carbon pools with fast turnover times, and if elevated CO2 and temperature directly or indirectly increase the turnover time of these labile carbon pools, then little or no carbon will be sequestered. Mitigation efforts to reduce the rise in atmospheric CO2 must therefore be focused not just on stabilizing or increasing terrestrial or oceanic carbon pool sizes but also either decreasing or slowing turnover rates of existing pools (for example, by increasing the chemical and physical protection of soil carbon through better soil management practices) or transferring carbon from pools with short turnover times to pools with longer turnover times (for example, converting pasture land to forest and forest to wood products).
Community composition, biodiversity and ecosystem function
It is widely accepted that species composition and community dynamics will be strongly affected by the combined effects of elevated CO2, warming, and changes in precipitation, and that these community changes will, in turn, have significant feedbacks on ecosystem function. However, despite this consensus, the underlying mechanisms driving plant community responses to global change are not well understood, and it has been difficult to accurately predict both community response to global change and the ecosystem consequences of these responses. This is due, in part, to the variable influence in time and space of global change on individual species, functional groups, and/or entire communities.
The responses of plant communities to simulated global change can be strongly influenced by individual plant species. A few CO2 enrichment experiments have shown that even a single species can dominate responses of an entire plant community. For instance, Griinzweig and Korner (2001) reported significant ecosystem-scale changes in aboveground biomass, reproduction, and plant nitrogen content in response to CO2 enrichment in semi-arid grassland assemblages from Israel. Surprisingly, these ecosystem-scale responses were attributable to CO2-induced changes in just one out of 32 plant species. Morgan et al. (2004a) also reported that CO2-induced increases in aboveground biomass in native Colorado short-grass steppe were driven primarily by one of 36 plant species, and that enhanced seedling recruitment appeared to be an important mechanism behind this response. How would these plant communities have responded without the CO2-responsive species, and what would have been the long-term implications for the ecosystems? These questions are difficult to answer, since species interact complexly in plant communities where microclimatic feedbacks and competition for resources occur. Absence of the CO2-sensitive plant species would not necessarily result in a non-responsive plant community since more resources would be available to the remaining plants, and the reaction of individual species to CO2 often interact with resource availability (Smith et al. 2000; Poorter and Perez-Soba 2001; Belote et al. 2003; Zavaleta et al. 2003). While a single species may drive a plant community response, plant community production and related responses to CO2 are generally enhanced by plant species diversity (Niklaus et al. 2001; Reich et al. 2001b). Species-rich plant communities are thus more likely to exhibit strong reactions to global changes. Greater responsiveness of species-rich over species-poor communities can involve one of several forms of synergy whereby the presence of one species enhances the capability of another species to respond to CO2 (Morse and Bazzaz 1994; Luscher et al. 1996; Reich et al. 2001b), or may simply be attributed to the greater likelihood of having global change-sensitive species in a community with more species.
Less work has been done on the role of below-ground biological diversity in global change experiments (Pendal et al. 2004). Linkages of aboveground and belowground biota indicate that global change may indirectly affect a number of belowground biological activities that will have powerful potential to feedback on plant communities, invoking both positive and negative responses (Wardle et al. 2004). Belowground biotic diversity will likely be important in determining the long-term reactions of plant communities to global change which are expected to be strongly conditioned by soil nutrient cycling (Zak et al. 2000).
Functional groups may also show differential responses to global change, and may be useful in streamlining approaches to understanding plant community responses to global change. However, contradictory results from field studies show that more work is needed to elucidate these differences (Morgan et al., 2004b; Nowak et al., 2004). For example, it has generally been predicted that C3 species will show greater photosynthetic response to CO2 enrichment compared to C4 species (Strain and Bazaaz 1983). In a higher CO2 world, an increase in the ecosystem abundance of C3 relative to C4 species over time would thus be accompanied by increased ecosystem productivity (Arp et al. 1993). Although numerous studies have demonstrated the greater photosynthetic response to CO2 enrichment in C3 compared to C4 species (e.g., Bazaaz 1990; Bowes 1993; Ehrlinger and Monson 1993; Poorter 1993; Reich 2001a), other studies have shown few differences between species with these very different photosynthetic pathways, particularly under conditions of water or nutrient stress (e.g., Wand et al. 1999; Derner et al. 2003). Failure of the C3 vs. C4 functional group paradigm to manifest may be attributed, in part, to the fact that stomates of most herbaceous species close under elevated CO2, which induces a water relations benefit that minimizes differences among photosynthetic functional groups. This is especially important in dry environments where CO2-induced water relations responses often drive CO2 responses (Morgan et al. 2004b). Legumes are another functional group that has been predicted to respond strongly to elevated CO2, because of their capability to fix atmospheric N. While this has been confirmed in several studies (Hebeisen et al. 1997; Tissue et al. 1997; Luscher et al. 1998; Griinzweig and Korner 2001), other experiments show little or advantage of N-fixing capability under elevated CO2 (Niklaus et al. 1998; Stocklin and Korner 1998; Nowak et al. 2004). In some cases, lack of a legume CO2 response may be attributable to insufficient soil P levels such that N fixation capacity is impaired (Korner 2000), or to super-optimal N levels (Poorter et al. 1996). However, in many cases, failure of legumes, C3
plants and other functional groups to respond simply indicates that one response mechanism may be insufficient to account for a species response in a plant community and other factors may need to be considered (e.g. water relations, nutrition, plant morphology, phenology). The temporal and spatial variability of the environment, which can interact with species and plant communities in complex ways, may also be important to determine species responses.
Different plant communities are also expected to show different responses to global change. Up-degraff et al. (2001), for example, reported greater seasonal CH4 emissions, aboveground net primary productivity, and dissolved N retention in bog compared to fen mesocosms under conditions of warming and water table manipulation. All of these examples underscore the linkage between species composition and ecosystem function, and illustrate that temporal patterns of ecosystem response to global change will be determined, in part, by the changing assemblages of species within that ecosystem.
For pragmatic reason, much of the experimental work on the effects of global change on species diversity has been done on species with short life spans such as annuals and short-lived perennials (Wand 1999; Reich et al. 2001a,b; Morgan et al. 2004a,b). Exceptions include the work on tree species response to (a) elevated CO2 in a coastal scrub-oak community in Florida, USA (Hymus et al. 2002a,b), (b) elevated CO2 and warming for two species of maple in Tennessee, USA (Norby et al. 1997, 2000), and (c) elevated CO2 and ozone in a northern forest ecosystem in Wisconsin, USA (Karnosky et al. 2003). The time scale of response will vary directly with the life span of the biota, with changes occurring on the scale of days to months for soil microbes, to years for annual plants, to decades and even centuries for longer-lived perennials and woody species. Although experimental manipulations will continue to be useful to evaluate the effects of changes in species composition on ecosystem function for short-lived species, alternative approaches, using space-for-time substitutions such as gradients and chronosequences, along with ecosystem- and regional-scale models may be necessary to elucidate species change and ecosystem consequences over time for species with longer life spans (e.g. decade or greater).
Was this article helpful?
You Might Just End Up Spending More Time In Planning Your Greenhouse Than Your Home Don’t Blame Us If Your Wife Gets Mad. Don't Be A Conventional Greenhouse Dreamer! Come Out Of The Mould, Build Your Own And Let Your Greenhouse Give A Better Yield Than Any Other In Town! Discover How You Can Start Your Own Greenhouse With Healthier Plants… Anytime Of The Year!