III.9 Ecosystem Productivity and Carbon Flows: Patterns Across Ecosystems Julien Lartigue and Just Cebrian

III.9 Ecosystem Productivity and Carbon Flows: Patterns across Ecosystems Julien Lartigue and Just Cebrian

OUTLINE by fitting the following single exponential equation to the pattern of detritus decay observed in experi- 1. Nature of carbon budgets mental incubations, DM ¼ DM e k(t t0), where 2. Rationale and approach for studying patterns of t t0 k is the decomposition rate, DM is the detrital ecosystem productivity and carbon flow t mass remaining in the experimental incubation at 3. Patterns in ecosystem productivity and carbon time t,DM is the initial detrital mass, and (tt )is flow t0 0 the incubation time 4. Conclusion detrital production. The amount (in g Cm2year1)of net primary production not consumed by herbi- The characterization and understanding of carbon flows in vores, which senesces and enters the detrital com- aquatic and terrestrial ecosystems are topics of paramount partment importance for several disciplines, such as ecology, bio- detritus. Dead primary producer material, which nor- geochemistry, oceanography, and climatology. Scientists mally becomes detached from the primary producer have been studying such flows in many diverse ecosystems after senescence for decades, and sufficient information is now available to herbivory. The amount (in g Cm2year1) of net investigate whether any patterns are evident in how carbon primary production ingested or removed, including flows in ecosystems and to determine the factors respon- primary producer biomass discarded by herbivores sible for those patterns. In particular, a wealth of infor- net primary production. The amount (in g Cm2 mation exists on the movement of carbon through the year1) of carbon assimilated through photosyn- activity of herbivores and consumers of detritus (i.e., de- thesis and not respired by the producer composers and detritivores), two of the major agents of nutrient concentration (producer or detritus). The per- carbon flows in ecosystems. This chapter analyzes the centage of nitrogen and phosphorus within pro- transference of carbon through herbivory and decomposi- ducer biomass or detritus on a dry weight basis tion in aquatic and terrestrial ecosystems, documents the nature and implications of salient patterns, and explains why those patterns emerge. 1. NATURE OF CARBON BUDGETS Carbon enters the biotic component of an ecosystem when inorganic carbon, often carbon dioxide, is taken GLOSSARY up and converted into organic compounds. With the absolute decomposition. The amount (in g Cm2 rare exception of chemosynthetic organisms, the en- year1) of detritus consumed by microbial decom- ergy for this conversion comes from photosynthesis. posers (e.g., bacteria, fungi) and detritivores, which Once inorganic carbon has been converted into organic range from detritivorous micro-, macro-, and ge- compounds, it is considered fixed. This production of latinous zooplankton in pelagic systems to micro- fixed carbon is known as primary production, and (<100 mm), meio- (100–500 mm), and macrofauna those organisms that can fix carbon are primary pro- (>500 mm) in benthic and terrestrial systems ducers. Gross primary production is the entire amount decomposition rate. The proportion of detrital mass de- of carbon fixed by a primary producer. Net pri- composed per unit time (e.g., day), often estimated mary production is gross primary production minus Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 3/6/2017 3:26 PM via UNIV OF SOUTH ALABAMA AN: 331277 ; Carpenter, Stephen R., Levin, Simon A..; The Princeton Guide to Ecology Account: s4595122 Productivity and Carbon Flows 321

Net primary ducers to herbivores is not complete, and only a frac- production tion of producer biomass ingested becomes herbivore biomass. When herbivory is considered as the percentage of net primary production removed, its implications for the impact of herbivores on carbon and Producer Herbivory nutrient recycling and storage as producer biomass in biomass Detrital the ecosystem become apparent. If herbivores remove a production Decomposition large percentage of net primary production, only a small percentage of the carbon fixed and nutrients Degradable Export Import detrital mass taken up by producers is available for accumulation as Refractory producer biomass. In such cases, herbivores have the accumulation potential to exert significant control on carbon and Refractory detrital mass nutrient storage by producers, which is commonly re- ferred to as top-down regulation (see chapter III.6). Likewise, as the percentage of net primary production Figure 1. Diagram of carbon flow into and out of the producer and consumed increases, so does consumer-driven recycling detrital pools in an ecosystem. (Adapted from Cebrian, 1999) of carbon and nutrients in the ecosystem. It is important to mention that, when diverse ecosystems are compared, absolute consumption and percentage of the organic compounds that have been broken down net primary production consumed are not always re- during respiration to fuel cellular processes within the lated. Ecosystems with high net primary production primary producer. may support large absolute consumption by herbi- It is the fixed carbon measured by net primary pro- vores, which may still represent a small percentage of duction that becomes primary producer biomass and that high net primary production, in comparison with part of the producer carbon pool (figure 1). This fixed ecosystems with lower net primary production sup- carbon will then either remain as producer biomass, be porting less absolute consumption but a larger per- consumed by herbivores, or enter the detrital pathway centage of net primary production lost to herbivores. and become part of the detrital carbon pool. The im- As is the case for herbivory, decomposition can also port or export of detritus can also alter the amount of be viewed as an absolute flux or as a proportion of carbon in the detrital pool, but regardless of the source detrital mass decomposed per unit time (i.e., decom- of the detritus, detrital carbon will either be recycled by position rate). When considered as an absolute flux, decomposers and detritivores or stored as refractory decomposition corresponds to the amount of detritus carbon. consumed by microbial decomposers and detritivores. In both aquatic and terrestrial ecosystems, the This consumption leads to the reduction of particulate transfer of fixed carbon from primary producers to and dissolved detritus into simpler and simpler con- herbivores and decomposer/detritivores provides ma- stituents and, ultimately, to nutrient mineralization. jor pathways for the flow of energy and nutrients. As a Much like herbivory, decomposition, when regarded as result, these transfers have consequences not only for an absolute flux, is indicative of the potential levels of carbon storage but also for nutrient recycling and decomposer and detritivore production maintained in herbivore and decomposer/detritivore populations. the ecosystem because only a fraction of the carbon In assessing these transfers, it is important to rec- ingested by decomposers and detritivores is metabo- ognize that they can be viewed in absolute as well as lized into biomass of these organisms. When decom- proportional terms. Absolute size refers to the amount position is viewed as the proportion of detrital mass or magnitude of the transfer measured in units of decomposed per unit time, its implications for how fast producer carbon often over space and time (i.e., g C m2 carbon and nutrient flow through the detrital pathway year–1), whereas proportional size refers to the per- become apparent. Ecosystems whose decomposition centage of net primary production consumed by her- rate is high tend to have faster nutrient and carbon bivores or the percentage of detrital mass consumed recycling rates and store less carbon in their detrital per unit time by decomposers and detritivores. pools regardless of any large differences in detrital pro- When regarded as an absolute flux, herbivory sets duction. It is worth mentioning that, when diverse limits to the level of herbivore production maintained ecosystems are compared, higher values of absolute in an ecosystem. Because of herbivore respiration and decomposition do not always equate to higher de- herbivore egestion of nonassimilated producer bio- composition rates. Ecosystems with low detrital pro- mass, the transfer of fixed carbon from primary production may have high decomposition rates, yet small Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

absolute decomposition, when compared with other grams of carbon or ‘‘g C.’’ Last, researchers need to ecosystems with high detrital production, low decom- ensure that the conclusions obtained from multistudy position rates, and large absolute decomposition. data sets are not compromised by the uncertainty that results from compiling values from studies that use different methods, assumptions, and sample sizes. Meta- 2. RATIONALE AND APPROACH FOR STUDYING analysis and estimation of error propagation are two PATTERNS OF ECOSYSTEM PRODUCTIVITY examples of techniques that allow researchers to test the AND CARBON FLOW robustness of conclusions obtained from literature The first studies that measured productivity and the comparisons. flow of carbon focused on individual ecosystems. These studies sought to characterize the transfer of carbon 3. PATTERNS IN ECOSYSTEM PRODUCTIVITY between trophic levels with the goal of understanding AND CARBON FLOW how energy moved through an ecosystem. The pio- neering studies of Howard T. Odum in freshwater Having discussed the nature of carbon budgets and the springs in Florida and John M. Teal in the salt marshes rationale behind developing ecosystem carbon budgets of Georgia are classic examples of this early work. as well as assembling this information into larger data Later studies sought to understand what factors sets, we now consider the general patterns in produc- limited net primary production and decomposition by tivity and carbon flow that emerge from the analysis of investigating differences in these processes across en- these larger data sets. In this section, we analyze these vironmental gradients within the same type of ecosys- patterns and flow by first exploring the overall differ- tem. These studies led to now-well-known patterns ences between aquatic and terrestrial ecosystems being established. Annual precipitation is a major de- and then exploring patterns within each type of eco- terminant of net primary production in grassland system. ecosystems in arid regions across the Great Plains of the United States. In eastern deciduous forests in the Uni- Aquatic and Terrestrial Ecosystems: ted States, net primary production increases as the General Differences length of the growing season increases. In the mountains of Hawaii, net primary production and decomposition Net primary production in aquatic and terrestrial rates are positively associated with temperature along systems is highly variable, but production in both is an elevation gradient. In aquatic ecosystems, light and similar (figure 2A). Aquatic ecosystems, however, do nutrient availability frequently limit net primary pro- support greater carbon flow to herbivores, both as an duction. absolute carbon flux and as a percentage of net primary These studies and their successors have led to a production (figure 2D,E). Because herbivore produc- growing body of work measuring net primary pro- tion efficiency, the ratio of herbivore growth to carbon duction, herbivory, detrital production, decomposi- ingested, does not seem to vary significantly between tion, and the nutritional content of both producers and aquatic and terrestrial systems, this greater absolute detritus across a variety of aquatic and terrestrial eco- flux of producer carbon to herbivores implies that systems. Such a wealth of data can be extremely useful aquatic systems should support higher levels of herbi- for detecting more general trends in productivity and vore production compared to terrestrial systems, as has how carbon flows through ecosystems. recently been demonstrated. Researchers have compiled published values of net The higher percentages of net primary production primary production, herbivory, decomposition, and removed by herbivores in aquatic ecosystems suggest producer and detrital nutrient content for aquatic and that herbivores are more influential in carbon and nu- terrestrial ecosystems into large data sets. In assembling trient recycling and accumulation of producer biomass such comprehensive data sets, the researchers need to in aquatic ecosystems relative to their role in terrestrial ensure that the values compiled reflect adequately the ecosystems. Indeed, many of the examples of herbi- ecosystems examined. To do so, the values entered into vores controlling producer biomass (i.e., top-down the data set must include the most abundant species of control) are from aquatic ecosystems, although on producers and consumers in the ecosystem and encom- occasion herbivores are found to regulate producer pass at least a year or the entire growing season. In biomass in terrestrial ecosystems as well. The evidence addition, when making comparisons of productivity and for herbivores as important agents of nutrient recycling carbon flow across a wide range of ecosystems and using in aquatic ecosystems is also abundant, whereas there data from multiple studies, we must deal in a common is considerably less evidence for such a role for herbi- currency or unit. The most common unit of choice is vores in terrestrial ecosystems, which tend to channel a Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

- 100 1.0

year 6 2

- 10 0.8 0.6 1 4 (g C m Producer nitrogen Producer 0.4 Producer phosphorus Producer Net primary production 0.1 2 concentration (% dryconcentration mass) concentration (% dryconcentration mass) 0.2

D. E. F. 100

80 1000 80 ) 1

- 100 60 60 year 2

- 10 40 40

consumed (%) consumed 1 (g C m Detrital production Absolute consumption Absolute Net primary production 20 0.1 20 (% net primary production)

0.01

G. H. I. 1.8 1.6 1000 8 1.4 ) 1

- 100 1.2 6 1.0 year 2

- 10 0.8 4 1 0.6 (g C m Detritus nitrogen Detrital production 2 Detritus phosphorus 0.4 0.1 concentration (% dryconcentration mass) concentration (% dryconcentration mass) 0.2

Aquatic Terrestial J. K. ) 1 - 0.1 1000 ) 1 - 0.01 year 2

- 100 0.001 (g C m 10

0.0001 decomposition Absolute Decomposition rate (day rate Decomposition

1 Aquatic Terrestial Aquatic Terrestial Figure 2. Box plots comparing aquatic and terrestrial ecosystems: phorus concentration, (J) decomposition rate, and (K) absolute (A) net primary production, (B) producer nitrogen concentration, (C) decomposition. Boxes encompass 25th and 75th percentiles, and producer phosphorus concentration, (D) percentage of net primary the central line is the median. Bars are 10th and 90th percentiles production consumed, (E) absolute consumption, (F) detrital pro- with measurements outside of these percentiles indicated by duction as a percentage of net primary production, (G) detrital closed circles. Data set used to generate the box plots is from production, (H) detritus nitrogen concentration, (I) detritus phos- Cebrian and Lartigue (2004). Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

higher percentage of net primary production into the rate by the nutrient concentrations of their diets. With detrital pathway. higher-quality detritus to consume, aquatic decompos- There are several possible explanations for the ers and detritivores should generally have higher met- greater herbivory measured as an absolute flux of car- abolic and growth rates than their terrestrial counter- bon or as a percentage of net primary production parts leading to faster decomposition rates in aquatic consumed in aquatic versus terrestrial ecosystems. One than in terrestrial ecosystems. In other words, the explanation is that primary producers in aquatic eco- greater nutritional detritus found in aquatic ecosystems systems tend to have higher nutrient concentrations leads to more active decomposers and detritivores and than those in terrestrial ecosystems (figure 2B,C). There faster decomposition rates in comparison with terres- is growing evidence that the growth rates of aquatic trial ecosystems. And indeed, comparisons have often and terrestrial herbivores are limited by the nutrient found faster decomposition rates in aquatic than in content of their diets (see chapter III.15). Under such a terrestrial ecosystems despite substantial environmen- premise, herbivore metabolism and growth in aquatic tal variability between the two types of ecosystem. ecosystems are promoted by a diet of higher nutritional Faster decomposition rates in aquatic systems indi- quality, and higher rates of absolute consumption and cate faster rates of nutrient recycling through the de- larger percentages of net primary production removed trital pathway. This is supported by evidence of faster by herbivores result. Indeed, aquatic ecosystems sup- turnover rates of nutrients through the detrital pool in port greater herbivore standing stocks than do terres- aquatic ecosystems compared to terrestrial ecosystems. trial ecosystems. Higher concentrations of structural, In addition, faster decomposition rates in aquatic refractory compounds, such as lignin, in terrestrial ecosystems along with lower detrital production imply producers could also lead to lower rates of herbivory in the accumulation of smaller standing stocks of detritus terrestrial ecosystems. Other compounds in the pro- in comparison with terrestrial ecosystems. ducers, such as fatty acids and digestible carbohydrates, However, aquatic and terrestrial ecosystems show and differences in the availability of nutrients in the similar values of absolute decomposition despite the producer for herbivore digestion may also affect the higher decomposition rates found in the former sys- growth rates of herbivores and the intensity of her- tems (figure 2K). The reason for this lies in the interplay bivory, but their role in explaining the differences in between detrital production and decomposition rates; herbivory observed between aquatic and terrestrial aquatic ecosystems produce less detritus than terres- ecosystems requires further research. Herbivore be- trial ecosystems, but it decomposes faster. As a conse- havior, size, energy demands (endothermy versus ec- quence, absolute decomposition, which corresponds to tothermy), and predation intensity as well as other the product between detrital production and decom- factors may also have an impact on herbivory intensity position rate, remains similar between the two types of and supersede the expected effects of producer nutri- ecosystem. Because the efficiency of decomposer and tional quality, especially when only few ecosystems are detritivore production does not differ between aquatic being compared. and terrestrial ecosystems, aquatic and terrestrial eco- Because net primary production differs little be- systems should support similar amounts of decomposer tween aquatic and terrestrial ecosystems, and herbi- and detritivore production. Interestingly, aquatic eco- vory is greater in aquatic ecosystems, aquatic ecosys- systems feature lower standing stocks of decomposers tems tend to transfer a smaller flux of producer carbon, and detritivores than do terrestrial ecosystems, point- both in absolute terms and as a percentage of net pri- ing to higher rates of predation on decomposers and mary production, to the detrital pathway than do ter- detritivores in the former ecosystems. restrial ecosystems (figure 2F,G). However, most net primary production in both types of ecosystem is not Patterns within Aquatic and Terrestrial Ecosystems consumed by herbivores and enters the detrital compartment. We now move from discussion of differences between The higher nutrient concentrations of aquatic pro- aquatic and terrestrial ecosystems to consideration of ducers compared to terrestrial producers carry over patterns within each type of ecosystem. Net primary into the detrital compartment as well, where aquatic production and producer nutrient concentrations detritus has higher nutrient concentrations than ter- are uncorrelated within aquatic and within terrestrial restrial detritus (figure 2H,I). Aquatic detritus also ecosystems. The identification of such a general inde- decomposes at a faster rate than does terrestrial detri- pendence may seem surprising at first (figure 3A,B). tus (figure 2J), possibly because microbial decomposers Indeed, many fertilization experiments have shown and invertebrate and vertebrate detritivores, like her- that increased nutrient availability often leads to in- bivores, appear limited in their metabolism and growth creased nutrient concentrations in producer biomass Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

A. B. 10,000 10,000 ) ) 1 1 - - 1000 1000 year year 2 2 - -

100 100

10 10 Net primary production (g C m Net primary production (g C m

1 1 0.01 0.1 1 10 0.001 0.01 0.1 1 10 Producer nitrogen concentration (%DW) Producer phosphorus concentration (%DW) Figure 3. Net primary production from aquatic (gray symbols) down-pointing triangles), and seagrass (gray up-pointing trian- and terrestrial (black symbols) ecosystems plotted against pro- gles). The terrestrial communities represented are tundra shrubs ducer (A) nitrogen concentration and (B) phosphorus concentration. and grasses (black down-pointing triangles), freshwater and ma- The aquatic communities represented are freshwater phytoplank- rine marshes (black diamonds), temperate and tropical shrublands ton (gray circles), marine phytoplankton (gray squares), freshwater and forests (black circles), temperate and tropical grasslands benthic microalgae (gray left-pointing triangles), marine ben- (black squares), and mangroves (black up-pointing triangles). thic microalgae (gray right-pointing triangles), marine macro- (Adapted from Cebrian and Lartigue, 2004) algae (gray diamonds), freshwater submerged macrophytes (gray

and higher levels of net primary production in aquatic with higher nutrient concentrations lead to faster her- and terrestrial ecosystems. This independence likely bivore growth rates and larger percentages of net pri- stems from the large environmental variability en- mary production consumed. It follows that, regardless compassed when a large range of ecosystems are com- of whether aquatic or terrestrial ecosystems are con- pared within the aquatic or terrestrial realm. Growth sidered, herbivores should exert a greater control on limitation by light and temperature, water availability producer biomass accumulation and carbon and nutri- (in terrestrial ecosystems), wave action (in aquatic eco- ent recycling in ecosystems composed of producers with systems), and other types of environmental stress may higher nutrient concentration. very well prevent a positive association between pro- Although there is a positive association between ducer nutrient concentrations and net primary produc- producer nutrient concentration and the percentage of tion over a broad range of ecosystems. net primary production consumed by herbivores in Aquatic and terrestrial ecosystems composed of pro- aquatic ecosystems, herbivory measured as an absolute ducers with higher nutrient concentrations do tend to flux of producer carbon to herbivores is only poorly have a greater percentage of net primary production associated with producer nutrient concentration (figure removed by herbivores (figure 4A,B) despite contrasting 4C,D). Absolute consumption is, however, strongly natures (invertebrate versus vertebrate), metabolic pat- associated with the absolute magnitude of net pri- terns (ectothermy vesus endothermy), behavior (mi- mary production (figure 4E) in aquatic ecosystems. In gratory versus resident), and feeding specificity (spe- other words, more productive aquatic ecosystems, but cialized versus generalized) of the herbivore populations not aquatic ecosystems having producers with higher in the ecosystems compared. As is the case for the gen- nutrient concentrations, support greater absolute con- eral comparison between aquatic and terrestrial eco- sumption by herbivores. This pattern stems from the systems, this relationship within each type of ecosystem interaction between the variability in net primary pro- is likely fueled by the growth rates of aquatic and ter- duction and the variability in the percentage consumed restrial herbivores, which are often limited by the nu- within aquatic ecosystems (figure 4F). A higher per- trient content of their diets. On this basis, producers centage of net primary production is lost to herbivores Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

10 10

1 1 production consumed production consumed

Percentage of net primary Percentage 0.1 of net primary Percentage 0.1

0.01 0.01 0.01 0.1 1 10 0.01 0.1 1 10 Producer nitrogen concentration (%DW) Producer phosphorus concentration (%DW) C. D. 1000 100

100 10 ) ) 1 1 - - year year

2 10 2 1 - - (g C m (g C m

Absolute consumption Absolute 1 consumption Absolute 0.1

0.1 0.01 0.01 0.1 1 10 0.01 0.1 1 10 Producer nitrogen concentration (%DW) Producer phosphorus concentration (%DW) E. F. 10000 100

1000 10 100 ) 1 - year

2 10 1 -

(g C m 1 production consumed Absolute consumption Absolute

Percentage of net primary Percentage 0.1 0.1

0.01 0.01 0.01 0.1 1 10 100 1000 10000 0.01 0.1 110100 1000 10000 Net primary production (g C m-2 year-1) Net primary production (g C m-2 year-1) Figure 4. The relationship between herbivory and producer nutri- (C) absolute consumption versus producer nitrogen concentration ent concentrations or net primary production in aquatic and ter- (solid line, terrestrial ecosystems regression, R 2 ¼ 0.38), (D) ab- restrial ecosystems: (A) percentage consumed versus producer solute consumption (solid line, terrestrial ecosystems regression, nitrogen concentration (dashed line, aquatic ecosystems regres- R 2 ¼ 0.64), (E) absolute consumption versus net primary produc- sion, R 2 0.37; solid line, terrestrial ecosystems regression, tion (dashed line, aquatic ecosystems regression, R 2 0.66; solid Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. ¼ ¼ R 2 ¼ 0.40), (B) percentage consumed versus producer phosphorus line, terrestrial ecosystems regression, R 2 ¼ 0.25), and (F) abso- concentration (dashed line, aquatic ecosystems regression, lute consumption versus net primary production. Symbols are the R 2 ¼ 0.44; solid line, terrestrial ecosystems regression, R 2 ¼ 0.65), same as in ﬁgure 3. (Adapted from Cebrian and Lartigue, 2004) EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 3/6/2017 3:26 PM via UNIV OF SOUTH ALABAMA AN: 331277 ; Carpenter, Stephen R., Levin, Simon A..; The Princeton Guide to Ecology Account: s4595122 Productivity and Carbon Flows 327

in aquatic ecosystems comprised of producers with why the independence between net primary production higher nutrient concentrations. However, this percent- and producer nutrient concentration drives the inde- age varies little compared to the much larger differences pendence between detrital production and detritus nu- in net primary production within aquatic ecosystems. trient concentration. As a result, absolute consumption, which is the prod- Detritus with higher nutrient concentrations tends uct of net primary production and the percentage con- to exhibit faster decomposition rates within both sumed by herbivores, remains more closely associated aquatic and terrestrial ecosystems, although the trend with net primary production and only poorly associ- is not always strong (figure 5A,B). Yet this association ated with the percentage consumed and producer nu- is relevant given the substantial environmental vari- trient concentrations. An implication of these patterns is ability that may exist among ecosystems and the con- that aquatic ecosystems with higher net primary pro- trasting effects on decomposition rates that result duction, and not those composed of more nutritional from differing levels of temperature, soil or sediment producers, transfer more producer carbon to herbivores reduction–oxidation reaction conditions, and, in ter- and, because the efficiency of herbivore production does restrial systems, moisture. not seem to vary consistently across ecosystems, also The association between faster decomposition rates support higher herbivore production. and more nutritional detritus found within ecosystems, A different situation exists within terrestrial eco- regardless of whether these are aquatic or terrestrial, systems, where absolute consumption is positively as- probably results from the limitation exerted by the sociated with producer nutrient concentration but less nutrient content of the detritus on the metabolic and so with net primary production. Again, the explanation growth rates of decomposers and detritivores; higher lies in the interaction between the variability in net pri- nutrient concentrations in the detritus stimulate the mary production and the variability in the percentage metabolic and growth rates of these organisms, re- consumed within terrestrial ecosystems (figure 4F). sulting in faster decomposition rates. Two important Within terrestrial ecosystems, net primary production corollaries follow. First, ecosystems with more nutri- varies to a lesser degree than does the percentage con- tional detritus, regardless of whether they are aquatic sumed, and, as a result, absolute consumption remains or terrestrial, should feature faster nutrient recycling more closely associated with the percentage consumed rates through the detrital pathway. Second, ecosystems and, by extension, with producer nutrient concentra- with more nutritional detritus should also accumulate tion than with net primary production. Therefore, ter- smaller detrital pools provided the differences in de- restrial ecosystems composed of producers with higher composition rates among ecosystems exceed the dif- nutrient concentrations, in addition to supporting a ferences in detrital production. greater impact by herbivores on the accumulation of Despite the association between faster decomposi- producer biomass and carbon and nutrient recycling, tion rates and higher detritus nutrient concentrations transfer a greater flux of producer carbon to herbivores within both aquatic and terrestrial ecosystems, decom- and should have higher levels of herbivore production. position when viewed as an absolute flux is independent Recent work, however, has shown that herbivore pro- of detritus nutritional quality (figure 5C,D). Instead, duction is positively related to net primary production absolute decomposition is strongly associated with de- in terrestrial ecosystems because absolute consumption trital production and net primary production within and net primary production are positively related, al- either type of ecosystem (figure 5E). The reason for this beit not strongly, within these ecosystems. pattern lies in the interaction between the variability in Because most net primary production enters the de- detrital production and the variability in the percentage trital compartment in both aquatic and terrestrial sys- of detrital production decomposed within a year among tems, detrital production is strongly associated with net ecosystems (figure 5F). In aquatic ecosystems, and to a primary production within each type of system. Con- lesser degree in terrestrial ecosystems, detrital produc- versely, detrital production is unrelated to detritus nu- tion varies to a larger extent among ecosystems than trient concentration within each type of ecosystem. This does the percentage decomposed. As a consequence, lack of association stems from the independence be- absolute decomposition, which corresponds to the prod- tween net primary production and producer nutrient uct of detrital production and the percentage decom- concentration. Within both aquatic and terrestrial eco- posed, remains closely associated with detrital produc- systems, the nutrient concentration of producers changes tion and net primary production and independent of the little through senescence in relation to the variability percentage of detrital production decomposed. Because among producers. This, along with the strong asso- the percentage of detrital production decomposed is a ciation between detrital production and net primary surrogate for decomposition rates, absolute decompo- production within each type of ecosystem, explains sition also remains unrelated to decomposition rates Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

50% 50%

0.001 0.001 Decomposition rate (day rate Decomposition (day rate Decomposition 10% 10%

0.0001 0.0001 0.01 0.1 1 10 0.01 0.1 1 10 Detritus nitrogen concentration (%DW) Detritus phosphorus concentration (%DW) C. D. 10000 10000

1000 1000 ) ) 1 1 - - year year

2 100 2 100 - - (g C m (g C m

Absolute decomposition Absolute 10 decomposition Absolute 10

1 1 0.01 0.1 1 10 0.01 0.1 1 10 Detritus nitrogen concentration (%DW) Detritus phosphorus concentration (%DW) E. F. 10000 0.1

1000 )

1 0.01

- 90% ) 1 - 50% year

2 100 0.001 -

10% (g C m

Absolute decomposition Absolute 10

Decomposition rate (day rate Decomposition 0.0001

1 0.00001 1 10 100 1000 10000 1 10 100 1000 10000 Net primary production (g C m-2 year-1) Detrital production (g C m-2 year-1) Figure 5. The relationship between decomposition and detritus decomposition versus detritus phosphorus concentration, (E) ab- nutrient concentration or detrital production in aquatic and ter- solute decomposition versus net primary production (dashed line, restrial ecosystems: (A) decomposition rate versus detritus nitro- aquatic ecosystems regression, R2 ¼ 0.84; solid line, terrestrial gen concentration (dashed line, aquatic ecosystems regression, ecosystems regression, R2 ¼ 0.76), and (F) decomposition rate R2 ¼ 0.21; solid line, terrestrial ecosystems regression, R2 ¼ 0.46), versus detrital production. Horizontal dashed lines and percent-

Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. (B) decomposition rate versus detritus phosphorus concentration ages indicate the percentage of detrital production that would be (dashed line, aquatic ecosystems regression, R2 ¼ 0.34; solid line, decomposed within a year at the given decomposition rate. Symbols terrestrial ecosystems regression, R2 ¼ 0.54), (C) absolute de- are the same as in ﬁgure 3. (Adapted from Cebrian and Lartigue, compositionEBSCO Publishing versus : detrituseBook Academic nitrogen Collection concentration, (EBSCOhost) (D) absolute - printed 2004)on 3/6/2017 3:26 PM via UNIV OF SOUTH ALABAMA AN: 331277 ; Carpenter, Stephen R., Levin, Simon A..; The Princeton Guide to Ecology Account: s4595122 Productivity and Carbon Flows 329

and, by extension, to detritus nutrient concentration centration because net primary production often varies when either aquatic or terrestrial ecosystems are com- to a larger extent than does the percentage consumed by pared. Therefore, aquatic and terrestrial ecosystems herbivores or decomposed across a broad range of with higher primary and detrital production, and not ecosystems. Therefore, net primary production, and not those having more nutritional detritus, transfer more producer nutrient concentration, is often the indicator detrital carbon to decomposers and detritivores and, of secondary production (i.e., production of herbivores because the efficiency of decomposer and detritivore and consumers of detritus) in ecosystems. production varies little across ecosystems, support higher decomposer and detritivore production. FURTHER READING Cebrian, Just. 1999. Patterns in the fate of production in 4. CONCLUSION plant communities. American Naturalist 154: 449–468. Producer nutritional quality and net primary produc- Cebrian, Just. 2004. Role of first-order consumers in ecosys- tion are two independent predictors of herbivory and tem carbon flow. Ecology Letters 7: 232–240. Investigates decomposition in aquatic and terrestrial ecosystems. herbivore and decomposer and detrivore biomass and Herbivory, expressed as the percentage of net primary their impact on the turnover of producer-fixed carbon in aquatic and terrestrial ecosystems. production consumed by herbivores, and decomposi- Cebrian, Just, and Julien Lartigue. 2004. Patterns of herbi- tion, expressed as the proportion of detrital mass con- vory and decomposition in aquatic and terrestrial eco- sumed per day by decomposers and detritivores, are systems. Ecological Monographs 74: 237–259. A more positively associated with producer nutrient concen- detailed and technical discussion of the patterns in pro- tration but independent of net primary production, ductivity and decomposition discussed in the chapter. regardless of whether the comparison is done between Odum, Howard T. 1957. Trophic structure and productivity aquatic and terrestrial ecosystems or within each type of Silver Springs, Florida. Ecological Monographs 27: 55– of ecosystem. Thus, producer nutrient concentration, 112. and not net primary production, stands out as a po- Sterner, Robert W., and James J. Elser. 2002. Ecological tential indicator of top-down regulation of the pools of Stoichiometry: The Biology of Elements from Molecules to the Biosphere. Princeton, NJ: Princeton University producer biomass and detritus and nutrient and carbon Press. An introduction to the discipline of ecological recycling rates by first-order consumers in ecosystems. stoichiometry—the study of the balance of elements and The reverse situation is often found when herbivory energy in ecological interactions. and decomposition are expressed as absolute fluxes, Teal, John. 1962. Energy flown in the salt marsh ecosystem of which are then positively associated with net primary Georgia. Ecology 43: 615–624. production but independent of producer nutrient con- Copyright © 2009. Princeton University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.