1
Twenty Years of Litter and Root Manipulations: Insights into Multi-Decadal SOM Dynamics
2
3
Kate Lajtha1
4
Fox Peterson1,a
5
Knute Nadelhoffer2
6
Richard D. Bowden3
7
8
Affiliations:
9
1
Dept. Crop and Soil Sciences, Oregon State University, Corvallis, OR 97331
10
2
Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI
11
48109
12
3
Department of Environmental Science, Allegheny College, Meadville, Pennsylvania, 16335
13
a
Corresponding author: fox.peterson@oregonstate.edu, (541)-737-5751
14
15
Acknowledgements
16
This study was supported in part by NSF grants 0817064 and 1257032, as well as NSF 1237491
17
to the Harvard Forest LTER research program. We thank Mel Knorr, Brian Godbois, and Eric
18
Morrison for field leadership and assistance with sample collection, Jennifer Wig, Sarah
19
Wurzbacher, Bruce Caldwell, and Lauren Deem for laboratory assistance, and Serita Frey
20
for overseeing and maintaining the Harvard Forest DIRT plots.
21
22
1
23
Twenty Years of Litter and Root Manipulations: Insights into Multi-Decadal SOM Dynamics
24
25
Abstract
26
An understanding of the controls on carbon (C) stored in soil organic matter is of critical
27
importance to models of biospheric C sequestration. Although most models of ecosystem C
28
balance assume a strong relationship between plant litter inputs and soil C accumulation, there is
29
little experimental evidence in support of this relationship. The Detritus Input and Removal
30
Treatment (DIRT) experiment at Harvard Forest was designed to assess how rates and sources of
31
plant litter inputs control the accumulation and dynamics of organic matter and nutrients in soils
32
over decadal time scales. After 20 years of litter manipulation, doubling litter inputs did not
33
increase bulk soil C content, light or heavy fraction pools of C, or measures of labile C including
34
laboratory-based C respiration rates, all suggestive of priming by added litter early in the
35
experiment. The activities of two key enzymes were increased by 30% with litter additions,
36
suggesting that soil enzyme activities are sensitive indicators of changes in residue management
37
or detrital inputs. Sustained doubling of litter inputs did not increase soil N concentrations in
38
either 0-10cm or 10-20cm mineral soil layers, suggesting that these soils are at or close to their N
39
retention capacity under current climatic conditions at the site. Exclusion of either aboveground
40
litter or roots, or both litter and roots, resulted in significant but slight (5-10%) decreases in total
41
soil C, light fraction C, and soil respiration, but there were no differences among removal
42
treatments. Soil C pools were remarkably resistant to destabilization over two decades, even in
43
plots without roots and thus root-derived aggregation. Similarly, there appeared to be slow
44
recovery of soil C and N pools after removal of O and A soil horizons but with normal litter
45
inputs. Clearly soil C pools are not tightly coupled over decadal time-frames to changes in
46
productivity or management that affect litter inputs.
2
47
Introduction
48
Globally, soils store more than three times the amount of carbon as the atmosphere, and four and
49
a half times the amount of carbon as the world’s biota (Lal 2004). Soil degradation, land use
50
change, and unsustainable forest management have substantially decreased soil carbon stocks
51
(Lal 2004, Vagen et al. 2005). Although carbon sequestration in soil is often suggested as a
52
management technique to reduce the rate of atmospheric CO2 increases, mechanisms of soil
53
carbon sequestration, the amount of carbon that may potentially be sequestered, and the long-
54
term implications for carbon sequestered in soils are poorly understood (Baldock and Skjemstad
55
2000, Six et al. 2002, von Luztow et al. 2006, 2008).
56
Increasing aboveground biomass necessarily sequesters C from the atmosphere, but
57
changes in C masses stored in biomass do not necessarily lead to immediate or long-term
58
changes in soil C storage (Sulzman et al. 2005, Crow et al. 2009a). For example, there is
59
evidence that soil organic carbon content does not change in the short-term after forest harvest
60
(Olsson et al. 1996, Johnson and Curtis 2001, Johnson et al. 2002, Yanai et al. 2003), even
61
though forest harvest clearly causes an immediate decrease the C content of the stand. The
62
response of soil organic matter (SOM) to changes in forest biomass and litter inputs might have a
63
significant temporal lag; Holub et al. (2005) found no effect of above- and below-ground organic
64
inputs on SOM in two different forest soils after four or eleven years. Although most models of
65
ecosystem C balance assume a strong linkage between NPP, litter inputs, and soil C
66
accumulation (Gottschalk et al. 2012), soils have a finite capacity to sequester C, and might
67
“saturate”, or achieve maximum equilibrium levels under different combinations of soil texture,
68
mineralogy, and climatic regimes (Chung et al. 2009, Stewart et al. 2009, Six et al. 2002,
69
Mayzelle et al. 2013). The C saturation deficit is the difference between current C content and
3
70
the point of C saturation (Hassink 1997), and thus C accumulation potential likely depends on
71
litter inputs as well as on current C content and soil saturation capacity.
72
Altering above and belowground inputs may also have complex effects on SOM pools
73
due to changes in microbial respiration rates (Fontaine et al. 2003, 2004, Sulzman 2005, Brant et
74
al. 2006, Crow et al. 2009a). Increases in labile carbon inputs to soil can cause disproportionate
75
increases in microbial respiration rates, known as positive priming. Sulzman et al. (2005) saw a
76
positive priming effect of 187 % in response to litter additions in an old growth forest after 13
77
years of litter manipulations, which agrees with other studies in forest ecosystems (Nottingham
78
et al. 2012, Phillips et al. 2012, Sayer et al. 2007, Langley et al. 2009). The quality of litter
79
inputs and the source of the material can have disproportional effects on soil C fluxes and
80
microbial composition (Cheng et al. 2012). The priming effect could, however, be short-lived
81
(Hoosbeek and Scarascia-Mugnozza 2009), and thus not be evident in longer-duration litter
82
manipulation studies. There is a clear need for more long-term, comprehensive studies to
83
elucidate plant litter controls on soil C dynamics over both short and long time scales.
84
The Detritus Input and Removal Treatment (DIRT) experiment was designed to assess
85
how rates and sources of plant litter inputs control the accumulation and dynamics of organic
86
matter and nutrients in forest soils over decadal time scales (Nadelhoffer et al. 2004, Lajtha et al
87
2005). DIRT sites consist of plots with additions and removal of litter and wood, cessation of
88
root inputs, and removal of the soil O and A horizon. The experiment is based on a study
89
designed by Francis Hole at the University of Wisconsin Arboretum in 1956. The DIRT network
90
consists of nine sites internationally, the oldest of which, besides the original Wisconsin site, was
91
founded in 1990 in the Harvard Forest, a transitional hardwood forest in MA, USA (Nadelhoffer
92
et al. 2004).
4
93
The objective of the study described here was to determine the effect of changing litter
94
inputs on the quality and quantity of SOM and on microbial activity in the DIRT plots at Harvard
95
Forest after 20 years. We hypothesized that: 1) as high-quality above-ground inputs (litter)
96
increased, C and N in the most labile SOM fractions would increase, and 2) with litter removals
97
total C would decrease, but the proportion of biochemically resistant C would increase.
98
However, we also considered the alternative hypothesis that these effects could be muted or
99
reversed by increases in microbial respiration via the priming effect, as we also hypothesized that
100
activities of key microbial enzymes associated with the initial stages of litter breakdown would
101
respond quickly to litter additions or removals; 3) stable SOM pools would not show similar
102
increases because they are controlled by processes functioning at timescales longer 20 years; 4)
103
decreasing below-ground inputs would have a larger effect on both labile and longer-cycling
104
SOM pools than above-ground litter inputs; studies have shown a greater impact of roots and
105
rhizosphere microbes than above-ground inputs on SOM stabilization (Rasse 2005, Wilson et al.
106
2009, Clemmensen et al. 2013).
107
108
Methods
109
Site Description
110
The Harvard Forest DIRT site located in the Tom Swamp tract in Petersham,
111
Massachusetts (42.29°N, 72.11°W), which is a transition/mixed hardwood-white pine-hemlock
112
forest. Dominant tree species at the DIRT site are northern red oak (Quercus rubra), red maple
113
(Acer rubrum), and paper birch (Betula papyrifera Marsh.), which represent 43, 19 and 15
114
percent of the total basal area of the stand, respectively. From 1733 to 1850 the site was
115
permanent pasture. In 1908 it was classified as old-field white pine, and then as a white pine
5
116
transition-hardwood in 1923 (Bowden et al. 1992). Soil are moderately well-drained sandy-loam
117
Inceptisols from the Charlton series. Forest floor depth is 3-8 cm, and average soil depth is 3m,
118
with a thin Oa horizon (1-3 cm). Roots have not been observed below 70 cm. Bedrock is
119
primarily granite, gneiss, and schist. Mean temperature ranges from -7°C in January to 20°C in
120
July, and mean annual precipitation is 110cm (Nadelhoffer et al. 1999).
121
Litter and root manipulations began in September 1990, and include five input/exclusion
122
treatments and a control (C), each replicated three times. Plots are 3m by 3m, and none include
123
trees or saplings. Core treatments are control (CO), double litter (DL), no litter (NL), no roots
124
(NR), no input (NI), and OA-less (Table 1).
125
Control plots receive normal aboveground and belowground inputs. Aboveground inputs
126
include leaves, twigs, seeds, flowering parts, and woody inputs <1cm diameter. Belowground
127
inputs include roots and root exudates. In No Litter (NL) and No Input (NI) plots, aboveground
128
litter is excluded using a plastic mesh fabric placed upon the plots from late September until late
129
October (when 95% of all litterfall occurs). After senescence is complete, the leaves are
130
removed from the plots. Any aboveground litter that falls upon the plots during the non-winter
131
or non-autumn period is removed occasionally by hand. Living roots were excluded from the No
132
Roots (NR) and NI treatments. Trenches were excavated around each plot to a depth of 1.4 m,
133
approximately 0.4 below the depth of tree roots at the site; roots entering the plots were thus
134
severed from nearby trees, but roots within the plots themselves were not removed. Trenches
135
were lined with fiberglass or with impervious plastic root barriers, and then refilled. Double
136
Litter (DL) plots receive twice the annual input of aboveground litter; litter collected monthly or
137
during autumn senescence from a nearby NL or NI plot is placed upon the DL plots. No trees
138
exist within any of the plots. The surfaces of all plots are kept free of ground vegetation via hand
6
139
weeding. In the NI and NR plots, shade cloths or light burning with a propane torch was used to
140
control mosses that grew directly upon mineral soil.
141
litter exclusion has resulted in the complete loss of the O horizon. We also used OA-less (OA)
142
plots, established to follow the trajectory of SOM formation from soils with treated to lower
143
organic matter contents to levels similar to impoverished soils, for some analyses as well. OA-
144
less plots were created by removing the O and A horizons, and then replacing this soil with 0-10
145
cm B-horizon soil obtained immediately nearby.
On these plots, 20 years of aboveground
146
147
148
Collection methods
Soils were collected in October 2010. Two O-horizon samples per plot were collected as
149
20cm by 20cm rectangular volumes. Fine roots were hand-picked from O horizon soils and
150
separated into <1 mm and 1-2 mm pools. Two mineral soil cores were collected from each plot
151
with a diamond bit corer mounted on a power auger and separated into 0-10 cm, 10-20 cm, 20-30
152
cm, and 30-50 cm pools. In some cases, the deepest sample depth extended to only 40 cm, where
153
glacial till prevented deeper coring. Rocks (>2mm) were re moved using 2mm sieves; rock
154
volume was determined via water displacement and the remaining soil weighed for estimates of
155
soil mass per area. Due to the rockiness of the area (often >40% rock), a single bulk density of
156
the fine soil mass was calculated for each horizon, and the mean soil mass of each horizon across
157
the site was used to convert %C data to an areal basis. Samples were kept in airtight baggies at
158
4°C for transport to Oregon State University. Subsamples used for the year-long incubation
159
remained field moist at 4°C until measurements began. The remaining soil was air dried and
160
stored in airtight plastic bags until analysis.
161
7
162
163
Respiration Analysis
CO2 respiration rates were measured in a laboratory incubation of the mineral soils from
164
0-10 cm and 10-20 cm. 70g dry-weight equivalent of moist soil from each soil core of the
165
mineral horizons was placed in microlysimeters based on the design described in Nadelhoffer
166
(1990). Soils were saturated with 0.01 M CaCl2 and allowed to drain until no more water was
167
lost through seepage for 4 hours. This moisture level was defined as field capacity, and 60% of
168
this field capacity was maintained throughout the experiment by periodic additions of deionized
169
H2O. We measured respiration on days 1, 3, 7, 15, 25, 33, 55, 63, 96, and 242 with a LiCor LI-
170
6400 and a custom soil respiration attachment to fit our microcosms. Between measurements,
171
microcosms were stored in the dark at room temperature (approximately 20oC). The time needed
172
to reach 1% of original soil C respired was used as an index of labile C (Conant et al. 2008).
173
174
Density fractionation and acid hydrolysis
175
Mineral soils from 0-10 cm and 10-20 cm were density-fractionated using sodium
176
polytungstate (SPT) at three densities (1.85 g cm-3, 2.4 g cm-3, 2.8 g cm-3) following Sollins et al.
177
(2009). Soils were also analyzed for labile C following the acid hydrolysis method of Paul et al.
178
(1997), with minor adjustments; specifically, particulate organic matter was removed with
179
sodium polytungstate at a density of <1.85 g cm-3 before hydrolysis.
180
181
182
Enzyme Analyses
The activities of two key soil enzymes involved in litter breakdown and phosphorus
183
acquisition were measured as in Caldwell et al. (1999). Levels of β-glucosidase (β-GLC, EC
184
3.2.1.21) and phosphomonoesterase (PME, EC 3.1.3.2) were measured by modifying
8
185
conventional p-nitrophenyl-ester based assays in surface mineral soil only. Assays were run
186
without conventional buffers to measure enzyme activity under actual soil matrix conditions.
187
The ratio of phosphomonoesterase to β-glucosidase was used as an index of nutrient-yielding
188
potential to energy-yielding potential.
189
190
Computational Methods
191
Statistics were calculated using the native functions in Matlab 2013a (The Mathworks, 2013).
192
Normality was tested using the Shapiro-Wilks test. When data were normally distributed, means
193
were compared using an N-way ANOVA including interaction terms followed by Tukey’s least
194
significant difference post-hoc test to assess the magnitude and directionality of differences.
195
When data was not found to be normal, medians were compared using the Kruskal-Wallis non-
196
parametric test, with Wilcoxon rank-sum post-hoc tests.
197
198
Results
199
There were significant differences in total C and N concentrations (g g-1 soil) and contents (g
200
m-2) of both mineral and organic soils among the treatments (Table 2). In the O-horizon, Double
201
Litter C and N were significantly greater than Control (p = 0.03) and OA-less and No Roots (p <
202
0.001 for both), but the removal treatments did not differ from one another. There were also
203
significant differences in fine root mass between treatments; Double Litter and Control plots had
204
significantly more fine root mass than all other treatments (p< 0.001 for all comparisons between
205
Double Litter and other treatments; p < 0.03 for comparisons between Control and other
206
treatments). Although slight, there were still measurable fine roots in the No Roots plots.
207
In the 0-10 cm mineral soil, Control and Double Litter C contents were not significantly
208
different from one another, but they were greater than C contents of all litter removal treatments
9
209
(Table 2). The OA-less soil C was significantly lower than all other treatments. Litter removal
210
treatments were not significantly different from each other. Trends for treatment differences in C
211
concentration in 10-20 cm mineral soil followed those of 0-10 cm mineral soil. OA-less soil C
212
content was significantly lower than all other treatments and Double Litter C content was greater
213
than plots without roots. However, C content in the 10-20 cm depth was also only significantly
214
less in the OA-less treatment (p = 0.04), with no other statistical differences between treatments.
215
There were no significant differences among treatments for N content for either soil horizon.
216
Additionally, there were no significant differences in the C:N ratios for either horizon.
217
In laboratory incubations, there were significant differences among plots in the total
218
amount of C respired after 242 days (Figure 1; p< 0.001). The amount of C respired in Control
219
and Double Litter soils at both depths were not significantly different from each other on a per g
220
soil C basis. However, in the surface (0-10cm) soils, cumulative C respired was lower in
221
reduced input treatments than in Control and Double Litter. In 10-20 cm soils patterns were less
222
clear. The Control and Double Litter soils were not significantly different from one another, but
223
the OA-less treatment had significantly higher respiration rates per g soil than all other
224
treatments (p<0.001). Respiration rates expressed per g C soil, a measure of the decomposability
225
of C in soil, followed patterns of respiration expressed per g soil, but differences among
226
treatments were not as large. The amount of time in days to respire 1% of soil C followed
227
patterns of total C respired after 242 days (Table 3).
228
The light soil fraction (densities <1.85 g cm-3) was the pool most responsive to litter input
229
manipulation (Figure 2). Control and Double Litter treatments had significantly more light
230
fraction material than the removal treatments (p< 0.001); there were no significant differences
231
among the removal treatments. Other fractions did not show any trends with detrital treatment.
10
232
In deeper (10-20 cm) soils, variability in density pool recovery was very high, and only the No
233
Inputs treatment was significantly less than the Control (p = 0.01).
234
There was no significant treatment effect on % non-hydrolyzable C or % non-
235
hydrolyzable N in soils from either depth, and data were extremely variable. Values in the top
236
10 cm of soil ranged from 0.2 to 3.3 % non-hydrolyzable C and 0 to 0.4 % non-hydrolyzeable N.
237
There were significant differences in activities of both enzymes among treatments (Table
238
4). Activities of both β-glucosidase and phosphomonoesterase were 30% higher in Double Litter
239
plots than in Control plots (p = 0.01), and activities were significantly (p = 0.03 for β-
240
glucosidase; p = 0.04 for phosphomonoesterase) reduced in litter removal plots. For
241
phosphomonoesterase, No Input plots were significantly lower than the other two removal
242
treatments (p<0.03); for β-glucosidase, No Litter was significantly greater than No Root
243
(p=0.003), which was greater than No Input (p<0.001). In all cases, OA-less soils were
244
significantly lower than Controls. Declines in β-glucosidase with root removals (No Root, No
245
Input) were greater than declines in phosphomonoesterase, and thus ratios of
246
phosphomonoesterase to β-glucosidase, an index of nutrient-yielding potential to energy-yielding
247
potential, was significantly greater (p<0.001) in root removal treatments.
248
249
Discussion
250
Our objectives were to determine the effect of changing litter inputs on the quality and
251
quantity of SOM and on microbial activity in the DIRT plots at Harvard Forest after 20 years,
252
and to compare the effects of above and below ground inputs on SOM dynamics.
253
Although Townsend et al. (2013) found higher total soil C content in mineral soils from
254
DIRT plots with increased litter inputs after 50 years, other younger (< 20 years) sites have not
11
255
observed significant differences in C content with added litter (Sulzman et al. 2005, Bowden et
256
al. 2013, Fekete et al. 2013). There are multiple reasons that Double Litter plots might not have
257
greater C content in mineral soil than Control plots; high rates of litter decomposition, and
258
microbial respiration of solubilized products of decomposition, might make detection of
259
significant changes in C pools difficult in the first few decades after land management or NPP
260
changes that result in altered rates of litter inputs. However, it is worth noting that although
261
these DIRT sites found no differences in soil C with litter additions, means of Double Litter plots
262
all tended to be lower than Control plots, suggestive of priming. However, significant increases
263
in soil C content were found in a litter addition experiment in a moist tropical forest (Leff et al.
264
2012), suggesting that soil C dynamics in tropical soils, with very high rates of decomposition
265
and soil respiration, may not exhibit a priming effect from new C inputs. After 28 years in the
266
original Wisconsin forested DIRT sites, Boone (1994) saw disparate effects in the Double Litter
267
treatments, noting an 85% soil C increase in the Noe Woods site, but a 15% decrease in the
268
Wingra Woods site, suggesting that at 28 years of inputs, priming, considered to be a short-term
269
effect, might be balanced by incorporation and stabilization of new C inputs. By year 50,
270
Townsend et al. (in review) saw significant increases in soil C in both Double Litter treatments at
271
this site, suggesting either than priming had ceased or that stabilization of C inputs was greater
272
than priming.
273
The lack of a significant increase in mineral soil N with increasing litter additions was
274
surprising, as Aber et al. (1993) reported a nearly 100% recovery of added N in soils at the
275
Harvard Forest Chronic N Amendment plots after 3 years. With N additions, we did see
276
significant increases in N content and concentration in the O-horizon, but C:N ratios did not
277
differ between treatments in any of the mineral soil depths (0-10 cm, p = 0.29; 10-20 cm, p =
12
278
0.51). Nadelhoffer et al. (2004) and Magill et al. (2004) also found high recovery of added N in
279
soils in these plots, although most N retention was in the organic horizon of soil. However other
280
studies suggest that these forests are at an intermediate stage of N saturation, and McDowell et
281
al. (2004) found that within a decade of experimental treatment, inorganic N fluxes reached or
282
exceeded the level of fertilizer application in most plots. The lack of response of mineral soil to
283
N additions via litter (about 2 – 2.5 g N m-2 yr-1; estimated from Bowden et al. 1993 and
284
Davidson et al. 2002) observed here could be due to (a) a low amount of N added compared to
285
the background level (approximately 250 – 300g N m-2) of N in this N-saturated ecosystem, (b)
286
plant uptake of the added N, (c) leaching of litter N once mineralized. Over 20 years of litter
287
addition, we have added about 15% of the background N to the soil via litter, which should be
288
detectable only if all of the N were to accumulate in the soil. Plant uptake of N did not increase
289
in the Chronic N Amendment plots (Frey et al. 2013), this suggests that leaching of the added N
290
was very high, or that the availability of mineralized N from litter was higher than for high doses
291
of inorganic fertilizer.
292
Virtually all litter manipulation experiments have shown a relatively rapid decrease in
293
soil C with either root or aboveground litter removal (Leff et al. 2012, Fekete et al. in review).
294
After 20 years in our DIRT plots, there was approximately a 10% decrease in C content with leaf
295
litter removal and approximately a 5% decrease in C content with root exclusion, with the 0-10
296
cm depth decreasing much more than 10-20 cm, 20-30 cm, and 30+ cm. After 50 years of litter
297
exclusion in the Wisconsin site, Townsend et al. (2013) saw a greater than 50% reduction in C
298
content in surface soils; we do not know why C reduction was so much less in Harvard Forest on
299
a per year basis. Litter removal has been shown to have significant effects on both C and N
300
content of soils. Sayer (2006), in a review of litterfall manipulation experiments, noted that
13
301
long-term litter harvesting can severely deplete nutrient availability in forested soils, with
302
decreases in concentrations of P, Ca, Mg, and K often apparent within 5 years. Although we did
303
not measure cation availability in these soils, the lack of measurable response in N to litter
304
removal parallels the small reduction in C in these soils, suggesting a high resistance of our SOM
305
to decomposition.
306
Several studies have suggested that roots may contribute to the more stable C pools more
307
than aboveground residues (ie Oades 1988, Clemmensen et al. 2013, Rasse et al 2005), and that
308
rhizosphere microbes contribute significantly to soil C stabilization (Wilson et al. 2009).
309
However, root exudates and mycorrhizae can also have the opposite effect by stimulating SOM
310
decomposition (Cheng et al. 2012). The small response to litter exclusion after 20 years in this
311
study is surprising, as is the result that removal of aboveground litter had a greater effect on total
312
C pools than did root removal. However, the low overall response suggests that this hypothesis
313
cannot be evaluated until a larger fraction of SOM is respired.
314
The light fraction (< 1.85 g cm-3) of soil has generally been found to be the most reactive
315
fraction to management or disturbance (Bremer et al. 1994, Liao et al. 2006, Spielvogel et al.
316
2006, Throop et al. 2013) although other studies have suggested that C accumulation occurs
317
primarily in heavy-fraction, slow turnover pools (Grandy and Robertson 2007). As expected,
318
decreased litter inputs caused decreases in light fraction pools, showing that this pool is more
319
rapidly depleted than higher density fraction pools, composed of soil carbon stabilized in
320
aggregates (intermediate density fractions) or by strong organo-mineral associations (heaviest
321
fractions) (Hatton et al. 2012, Mayzelle et al. 2013), even though priming of all pools has been
322
observed (Guenet et al. 2012). However, litter additions did not result in parallel increases in
323
either total soil C or in any measurable density pool. Rapid respiration of litter directly from the
14
324
O horizon or else in deeper mineral soil (Spears and Lajtha 2004), in combination with priming
325
of soil C pools, likely prevented significant accumulation of soil C even with doubled litter input
326
rates.
327
We calculated respiration in laboratory incubations both on a per gram basis, representing
328
simply C production rates, as well as on a C respired per g C soil as a measure of carbon quality,
329
specifically carbon lability. As expected, C decomposability was greater in Control and Double
330
Litter plots compared to the litter removal plots, although differences were slight. Had any
331
significant amount of added litter accumulated in the mineral soil of the Double Litter plots, we
332
would have expected greater respiration per g C in these soils, as more of the total C would have
333
been derived from fresh litter. However, given the lack of response of total C, or even light
334
fraction C in mineral soil, to detrital additions, the lack of response in respiration rates was
335
expected. Crow et al. (2006, 2009b) found a decrease in respired C of soil from Double Litter
336
plots compared to Controls at both the Andrews and Bousson DIRT sites, suggestive of priming
337
and depletion of labile C in the mineral soil. Those DIRT sites were measured at six (Andrews)
338
and twelve (Bousson) years after establishment; priming is suggested to be a short-lived
339
phenomenon (Guenet et al. 2010), and thus it is possible that early, high rates of priming are not
340
detectable in the 20 year Harvard Forest soils. We also expected high rates of respiration on a
341
per g C of initial soil in the OA-less plots, as virtually all C in these plots is derived from litter
342
that is 20 years old or younger. The fact that these rates in OA-less plots were lower than rates
343
for Control and Double Litter plots suggests that the soil microbial community has not yet fully
344
established itself in these soils. This was unexpected, given that tree roots, even if they had been
345
disturbed during plot installation, have had 20 years to recover.
15
346
Because extracellular soil enzymes are directly responsible for the initial processing of
347
detrital carbon and organic-bound nutrients (Sollins et al. 1996), treatment impacts on soil
348
enzyme activities should indicate initial functional response(s) of the microbial community to
349
fresh litter supply. Many studies have suggested that soil enzyme activities are generally the
350
most sensitive indicators of changes in the belowground microbial community from agricultural
351
residue management (Dick 1992, Dick 1994, Gregorich et al. 1994, Jordan et al. 1995). In a field
352
experiment in Costa Rica, both β-glucosidase and phosphomonoesterase responded more
353
strongly to litter removal than did total soil organic carbon (Caldwell et al. 1999). The soil
354
enzymes examined in this study are closely associated with the microbial processing of detrital
355
carbon and phosphorus; β-glucosidase is a key enzyme in cellulolytic activity during the break-
356
down of plant litter, while phosphomonoesterase mineralize ortho-P from organic phosphate
357
esters and is commonly used as a general indicator of microbial activity. It was not surprising,
358
therefore, that the response by enzymes was greater than the response of total C, density
359
fractions, or even respiration to litter manipulation. The fact that plots without roots showed
360
sharper decreases in energy yielding potential compared to nutrient yielding potential
361
underscores the importance of roots, root biota, and root exudates to the energy supply in soils.
362
After 20 years of eliminating new organic inputs, we expected the proportions of non-
363
hydrolyzable C and N to be higher in soils under litter exclusion, where there was less labile
364
material. During acid hydrolysis, compounds such as proteins, nucleic acids, and
365
polysaccharides are digested, leaving compounds that are resistant to digestion, including
366
aromatic components and wax-derived long-chain aliphatics (Paul et al. 1997). Although our
367
respiration measurements suggest that litter removal soils had less labile C, we did not see
368
significant differences among treatments. This might be a methodological issue; differences in
16
369
respiration measurements could have been due to differences in light fraction material among
370
plots, and we removed light fraction material before acid hydrolysis, and thus measured a
371
hydrolysable fraction only in mineral-associated SOM, which did not differ among treatments.
372
Plante et al. (2006) also found that C hydrolysability of grassland/agricultural soil was invariant
373
with management treatment that should have increased SOM recalcitrance with decreasing SOM
374
content, suggesting that “recalcitrance” may not be tightly coupled to the biochemistry of
375
preserved organic fractions.
376
Soils that had O and A horizons removed will likely take many more decades to recover a
377
soil organic matter profile similar to undisturbed sites, even with normal litter inputs and root
378
activity, and B-horizon soils that are likely highly undersaturated with respect to C. After 20
379
years, OA-less plots had only 37 % percent of the C and 18 % N of control. Schlesinger (1990)
380
also noted the generally slow rate of C accumulation in terrestrial soils. Our results also suggest
381
that the soil microbial function in these plots has not fully recovered either, as respiration of
382
fresh C in these plots was lower than respiration of all C in Control plots.
383
Taken together, our results suggest that soil C pools were remarkably resistant to either
384
stabilization or destabilization over two decades. We saw little mineral C loss even in plots
385
without roots and thus root-derived aggregation, and no increase in mineral soil C accumulation
386
with large increases in litter inputs. Similarly, there appeared to be slow recovery of soil C and
387
N pools after removal of O and A soil horizons but with normal litter inputs. Clearly soil C
388
pools are not tightly coupled over decadal time-frames to changes in productivity or management
389
that affect litter inputs.
390
391
392
17
393
Bibliography
394
Aber, J. D., A. Magill, R. Boone, J. M. Melillo, P. Steudler, and R. Bowden. 1993. Plant and soil
395
responses to chronic nitrogen additions at the Harvard Forest, Massachusetts. Ecological
396
Applications 3:156-166.
397
398
399
Baldock, J.A., and Skjemstad, J.O. 2000. Role of the soil matrix and minerals in protecting
natural organic materials against biological attack. Organic Geochemistry 31: 697-710.
Bowden, R. D., Nadelhoffer, K. J., Boone, R. D., Melillo, J. M., Garrison, J. B. 1993.
400
Contributions of aboveground litter, belowground litter, and root respiration to total soil
401
respiration in a temperate mixed hardwood forest. Canadian Journal of Forest Research
402
23: 1402-1407.
403
404
Brant, J.B., Myrold, D.D., Sulzman, E.W. & Ehleringer, J. 2006. Root controls on soil microbial
community structure in forest soils. Oecologia 148, 650-659
405
Bremer, E., H. H. Janzen, and A. M. Johnston. 1994. Sensitivity of total, light fraction and mineralizable
406
organic matter to management practices in a Lethbridge soil. Can. J. Soil Sci. 74:131-138.
407
408
409
Caldwell, B.A., Griffiths, R.P., and Sollins, P. 1999. Soil enzyme response to vegetation disturbance in
two lowland Costa Rican Soils. Soil Biol. Biochem. 31:1603-1608
Cheng, L., F. L. Booker, C. Tu, K. O. Burkey, L. Zhou, H. D. Shew, T. W. Rufty, and S. Hu. 2012.
410
Arbuscular Mycorrhizal Fungi Increase Organic Carbon Decomposition Under Elevated CO2.
411
Science 337:1084-1087.
412
Cheng, W., D. W. Johnson, and S. Fu. 2003. Rhizosphere effects on decomposition: controls of plant
413
species, phenology, and fertilization. Soil Science Society of America Journal 67:1418-1427.
414
415
Chung, H., K. Ngo, A.F. Plante, and J. Six. 2009. Evidence for Carbon Saturation in a Highly
Structured and Organic-Matter-Rich Soil. Soil Sci. Soc. Am. J. 74: 130–138.
18
416
Clemmensen, K. E., A. Bahr, O. Ovaskainen, A. Dahlberg, A. Ekblad, H. Wallander, J. Stenlid, R. D.
417
Finlay, D. A. Wardle, and B. D. Lindahl. 2013. Roots and Associated Fungi Drive Long-Term
418
Carbon Sequestration in Boreal Forest. Science 339:1615-1618.
419
Conant, R. T., R. A. Drijber, M. L. Haddix, W. J. Parton, E. A. Paul, A. F. Plante, J. Six, and J. M.
420
Steinweg. 2008. Sensitivity of organic matter decomposition to warming varies with its quality.
421
Global Change Biology 14:868-877.
422
Crow, S. E., E. W. Sulzman, W. D. Rugh, R. D. Bowden, and K. Lajtha. 2006. Isotopic analysis of
423
respired CO2 during decomposition of separated soil organic matter pools. Soil Biology and
424
Biochemistry 38:3279-3291.
425
Crow, S.E., Lajtha, K., Bowden, R.D., Yano, Y., Brant, J.B., Caldwell, B.A., Sulzman, E.W.
426
2009a. Increased coniferous needle inputs accelerate decomposition of soil carbon in an
427
old-growth forest. Forest Ecology and Management 258, 2224-2232.
428
Crow, S.E., Lajtha, K., Filley, T.R., Swanston, C.W., Bowden, R.D., Caldwell, B.A. 2009b.
429
Sources of plant-derived carbon and stability of organic matter in soils: implications for
430
global change. Global Change Biology 15: 2003-2019.
431
Davidson, E. A., K.Savage, P. Bolstad, D. A. Clark, P. S. Curtis, D. S. Ellsworth, P. J. Hanson, B. E.
432
Law, Y.Luo, K. S. Pregitzer, C. Randolph, and D.Zak. 2002. Belowground carbon allocation in
433
forests estimated from litterfall and IRGA-based soil respiration measurements. Agricultural and
434
Forest Meteorology 113:39-51.
435
436
Dick, R.P., 1992. A review: long-term effects of agricultural systems on soil biochemical and
microbial parameters. Agric. Ecosystems Environ. 40: 25–36.
19
437
Dick, R.P. 1994. Soil enzyme activities as indicators of soil quality. pp. 107–124 in J.W.
438
Doran, D.C. Coleman, D.F. Bezdicek, B.A. Stewart (Eds.), Defining Soil Quality for a
439
Sustainable Environment, Soil Science Society of America, Madison.
440
Fekete, I., Z. Kotroczó, C. Varga, P. T. Nagy, J. A. Tóth, R. D. Bowden, G. Várbíró, and K. Lajtha. The
441
effects of detrital inputs on soil carbon content and CO2 release in a Central-European deciduous
442
forest. Soil Science Society of America Journal, in review.
443
444
445
Fontaine, S., G. Bardoux, L. Abbadie, and A. Mariotti. 2004. Carbon input to soil may decrease
soil carbon content. Ecology Letters 7:314-320.
Fontaine, S., A. Mariotti, and L. Abbadie. 2003. The priming effect of soil organic matter: a
446
question of microbial competition? Soil Biology and Biochemistry 35:837-843.
447
Gottschalk, P., J. U. Smith, M. Wattenbach, J. Bellarby, E. Stehfest, N. Arnell, T. J. Osborn, C.
448
Jones, and P. Smith. 2012. How will organic carbon stocks in mineral soils evolve under
449
future climate? Global projections using RothC for a range of climate change scenarios.
450
Biogeosciences 9:3151-3171.
451
452
453
Grandy, A. S. and G. P. Robertson. 2007. Land-Use Intensity Effects on Soil Organic Carbon
Accumulation Rates and Mechanisms. Ecosystems 10:59-57.
Gregorich, E.G., M.R. Carter, D.A. Angers, C.M. Monreal, and B.H. Ellert. 1994. Towards a
454
minimum data set to assess soil organic matter quality in agricultural soils. Canadian
455
Journal of Soil Science, 74: 367–385.
456
457
458
459
Guenet, B., S. Juarez, G. Bardoux, L. Abbadie, and C. Chenu. 2012. Evidence that stable C is as
vulnerable to priming effect as is more labile C in soil. Soil Biology and Biochemistry 52:43-48.
Hamer, U. and B. Marschner. 2005. Priming effects in different soil types induced by fructose, alanine,
oxalic acid and catechol additions. Soil Biology and Biochemistry 37:445-454.
20
460
461
462
Hassink, J. 1997. The capacity of soils to preserve organic C and N by their association with clay
and silt particles. Plant Soil 191(1): 77–87.
Hatton, P.-J., M. Kleber, B. Zeller, C. Moni, A. F. Plante, K. Townsend, L. Gelhaye, K. Lajtha, and D.
463
Derrien. 2012. Transfer of litter-derived N to soil mineral-organic associations: evidence from
464
decadal 15N tracer experiments. Organic Geochemistry 42:1489–1501.
465
Hoosbeek, M. R. and G. E. Scarascia-Mugnozza. 2009. Increased Litter Build Up and Soil Organic
466
Matter Stabilization in a Poplar Plantation After 6 Years of Atmospheric CO2 Enrichment
467
(FACE): Final Results of POP-EuroFACE Compared to Other Forest FACE Experiments.
468
Ecosystems 12:220-239.
469
Holub, S.M., Lajtha, K., Spears, J. D. H., Toth, J. A., Crow, S. E., Caldwell, B. A., Papp, M.,
470
Nagy, P. T. Organic matter manipulations have little effect on gross and net nitrogen
471
transformations in two temperate forest mineral soils in the USA and central Europe.
472
Forest Ecology and Management 214, 320-330 (2005).
473
474
475
Johnson, D. W. and Curtis, P. S. 2001. Effects of forest management on soil C and N storage:
meta analysis. Forest Ecology and Management 140: 227-238.
Johnson, D. W., Knoepp, J. D., Swank, W. T., Shan, J., Morris, L. A., Van Lear, D. H.,
476
Kapeluck, P. R. 2002. Effects of forest management on soil carbon: results of some long
477
term sampling studies. Environmental Pollution 116: S201-S208.
478
Jordan, D., R.J. Kremer, W.A. Bergfield, K.Y. Kim, and V.N. Cacnio. 1995. Evaluation of
479
microbial methods as potential indicators of soil quality in historical agricultural fields.
480
Biology and Fertility of Soils, 19: 297–302.
21
481
Lajtha, K., Crow, S.E., Yano, Y., Kaushal, S.S., Sulzman, E., Sollins, P., Spears, J.D.H. 2005.
482
Detrital Controls on Soil Solution N and Dissolved Organic Matter in Soils: A Field
483
Experiment. Biogeochemistry 76: 261-281.
484
485
486
Lal, R. 2004. Soil carbon sequestration impacts on global climate change and food security.
Science 304: 1623-1626.
Langley, J. A., D. C. McKinley, A. A. Wolf, B. A. Hungate, B. G. Drake, and J. P. Megonigal. 2009.
487
Priming depletes soil carbon and releases nitrogen in a scrub-oak ecosystem exposed to elevated
488
CO2. Soil Biology and Biochemistry 41:54-60.
489
Leff, J. W., W. R. Wieder, P. G. Taylor, A. R. Townsend, D. R. Nemergut, A. S. Grandy, and C. C.
490
Cleveland. 2012. Experimental litterfall manipulation drives large and rapid changes in soil
491
carbon cycling in a wet tropical forest. Global Change Biology 18:2969-2979.
492
Liao, J. D., T. W. Boutton, and J. D. Jastrow. 2006. Storage and dynamics of carbon and nitrogen in soil
493
physical fractions following woody plant invasion of grassland. Soil Biology and Biochemistry
494
38:3184-3196.
495
Magill, A. H., J. D. Aber, W. S. Currie, K. J. Nadelhoffer, M. E. Martin, W. H. McDowell, J. M.
496
Melillo, and P. Steudler. 2004. Ecosystem response to 15 years of chronic nitrogen additions at
497
the Harvard Forest LTER, Massachusetts, USA. Forest Ecology and Management 196:7-28.
498
Marschner, B., S. Brodowski, A. Dreves, G. Gleixner, A. Gude, P. M. Grootes, U. Hamer, A. Heim, G.
499
Jandl, R. Ji, K. Kaiser, K. Kalbitz, C. Kramer, P. Leinweber, J. Rethemeyer, A. Schäffer, M. W.
500
I. Schmidt, L. Schwark, and G. L. B. Wiesenberg. 2008. How relevant is recalcitrance for the
501
stabilization of organic matter in soils? J. Plant Nutr. Soil Sci. 171:91-110.
502
503
Mayzelle, M. M., M. L. Krusor, R. D. Bowden, K. Lajtha, and J. W. Six. in review. The Effects of
Decreasing Carbon Saturation Deficit on Temperate Forest Soil Carbon Cycling. Soil Science
22
504
505
Society of America Journal.
McDowell, W. H., A. H. Magill, J. A. Aitkenhead-Peterson, J. D. Aber, J. L. Merriam, and S. S.
506
Kaushal. 2004. Effects of chronic nitrogen amendment on dissolved organic matter and inorganic
507
nitrogen in soil solution. Forest Ecology and Management 196:29-41.
508
Nadelhoffer, K. J., R.D. Boone, R. D. Bowden, J. D. Canary, J. Kaye, P. Micks, A. Ricca, J. A.
509
Aitkenhead, K. Lajtha, and W.H. McDowell. 2004. The DIRT experiment: litter and root
510
influences on forest soil organic matter stocks and function. Chapter 15 in: D. Foster and J. Aber
511
(eds.), Forests in Time: The Environmental Consequences of 1000 Years of Change in New
512
England. Yale University Press, pp. 300-315
513
514
515
Nadelhoffer, K. J., M. R. Downs, and B. Fry. 1999. Sinks for 15N-enriched additions to an oak forest and
a red pine plantation. Ecological Applications 72:72-86.
Nadelhoffer, K. J. 1990. Microlysimeter for measuring nitrogen mineralization and microbial
516
respiration in aerobic soil incubations. Soil Science Society of America Journal 54:411-
517
415.
518
Nadelhoffer, K. J., B. P. Colman, W. S. Currie, A. Magill, and J. D. Aber. 2004. Decadal-scale fates of
519
N-15 tracers added to oak and pine stands under ambient and elevated N inputs at the Harvard
520
Forest (USA). Forest Ecology and Management 196:89-107.
521
Nottingham, A., B. Turner, P. Chamberlain, A. Stott, and E. J. Tanner. 2012. Priming and microbial
522
nutrient limitation in lowland tropical forest soils of contrasting fertility. Biogeochemistry
523
111:219-237.
524
Oades, J.M. 1988. The retention of organic matter in soils. Biogeochemistry 5:37-70
23
525
Olsson, B. A., Staaf, H., Lundkvist, H., Bengtsson, J., Rosen, K. 1996. Carbon and nitrogen in
526
coniferous forest soils after clear-felling and harvests of different intensity. Forest
527
Ecology and Management 82: 19-32.
528
Paul, E. A., S. J. Morris, R. T. Conant, and A. F. Plante. 2006. Does the Acid Hydrolysis-
529
Incubation Method Measure Meaningful Soil Organic Carbon Pools? . Soil Sci. Soc. Am.
530
J. 70:1023-1035.
531
Phillips, R. P., I. C. Meier, E. S. Bernhardt, A. S. Grandy, K. Wickings, and A. C. Finzi. 2012. Roots
532
and fungi accelerate carbon and nitrogen cycling in forests exposed to elevated CO2. Ecology
533
Letters 15:1042-1049.
534
Plante, A. F., R. T. Conant, E. A. Paul, K. Paustian, and J. Six. 2006. Acid hydrolysis of easily
535
dispersed and microaggregate-derived silt- and clay-sized fractions to isolate resistant soil
536
organic matter. European Journal of Soil Science 57:456-467.
537
538
539
540
541
542
543
544
545
546
Rasse, D P., C. Rumpel, and M.-F. Dignac. 2005. Is soil carbon mostly root carbon? Mechanisms for a
specific stabilisation. Plant and Soil 269:341–356.
Sayer, E. J. 2006. Using experimental manipulation to assess the roles of leaf litter in the functioning of
forest ecosystems. Biol. Rev. 81:1-31.
Sayer, E. J., J. S. Powers, and E. V. J. Tanner. 2007. Increased Litterfall in Tropical Forests Boosts the
Transfer of Soil CO2 to the Atmosphere. PLoS ONE 2 (12):e1299.
Schlesinger, W. H. 1990. Evidence from chronosequence studies for a low carbon-storage potential for
soils. Nature 348:232-234.
Six, J., Conant, R.T., Paul, E.A., Paustain, K. 2002. Stabilization mechanisms of soil organic
matter: implications for C-saturation of soils. Plant and Soil 241:155-176.
24
547
548
549
Sollins, P., Homann, P., Caldwell, B.A. 1996. Stabilization and destabilization of soil organic
matter: mechanisms and controls, Geoderma 74: 65-105.
Sollins, P., Kramer, M., Swanston,, Lajtha, K., Filley, T., Aufenkampe, A., Wagai, R., Bowdeb,
550
R. 2009. Sequential density fractionation across soils of contrasting mineralogy: evidence
551
for both microbial- and mineral-controlled soil organic matter stabilization.
552
Biogeochemistry 96:209-231.
553
554
555
556
557
Spears, J. D. H. and K. Lajtha. 2004. The imprint of coarse woody debris on soil chemistry in the
western Oregon Cascades. Biogeochemistry 71:163-175.
Spielvogel, S., J. Prietzel, and I. Kögel-Knabner. 2006. Soil Organic Matter Changes in a Spruce
Ecosystem 25 Years after Disturbance. Soil Science Society of America Journal 70:2130-2145.
Stewart, C.E., K. Paustian, R.T. Conant, A.F. Plante, and J.H. Six. 2009. Soil carbon saturation:
558
Implications for measurable carbon pool dynamics in long-term incubations. Soil Biol.
559
Biochem. 41: 357–366.
560
Sulzman, E.W., Brant, J.B., Bowden, R.D. and K. Lajtha. 2005. Contribution of aboveground
561
litter, belowground litter, and rhizosphere respiration to total soil CO2 efflux in an old
562
growth coniferous forest. Biogeochemistry 73, 231-256.
563
564
565
Vagen, T.G., Lal, R., Singh, B.R. 2005. Soil carbon sequestration in Sub-Saharan Africa: a
review. Land Degradation and Development 16: 53-71.
von Lutzow, M., Kogel-Knabner, I., Ekschmitt, K., Matzner, E., Guggenberger, G., Marschner,
566
B., Flessa, H. 2006. Stabilization of organic matter in temperate soils: mechanisms and
567
their relevance under different soil conditions: a review. European Journal of Soil
568
Science 57: 426-455.
569
von Lützow, M., I. Kögel-Knabner, B. Ludwig, E. Matzner, H. Flessa, K. Ekschmitt, G. Guggenberger,
25
570
B. Marschner, and K. Kalbitz. 2008. Stabilization mechanisms of organic matter in four
571
temperate soils: Development and application of a conceptual model. Journal of Plant Nutrition
572
and Soil Science 171:111-124.
573
Wilson, G. W. T., C. W. Rice, M. C. Rillig, A. Springer, and D. C. Hartnett. 2009. Soil aggregation and
574
carbon sequestration are tightly correlated with the abundance of arbuscular mycorrhizal fungi:
575
results from long-term field experiments. Ecology Letters 12:452-461.
576
577
Yanai, R. D., Currie, W. S., and Goodale, C. L. 2003. Soil carbon dynamics after forest harvest:
and ecosystem paradigm reconsidered. Ecosystems 6: 197-212.
578
26
579
TREATMENT
DESCRIPTION
Control
Natural above and belowground litter inputs are allowed.
Double Litter
Above ground inputs are doubled by adding litter removed annually and proportionately
allocated from the No Litter plots.
Aboveground inputs are removed annually from plots.
No Litter
No Roots
No Inputs
OA-less
580
581
Roots are excluded with trenching that extends from the soil surface to below 30 cm
depth. Aboveground annual growth (herbaceous vegetation) is removed with herbicide.
Aboveground inputs are excluded as in No Litter plots; belowground inputs are
prevented as in No Roots plots.
Top 30 cm of soil was replaced with mineral soil in 1990.
Table 1. Description of treatments at the Harvard Forest DIRT Plots.
582
27
583
CONTROL
DOUBLE LITTER
NO INPUTS
NO LITTER
NO ROOTS
OA-LESS
308.0±43.9 a
244.8±28.5 c
mg C g -1 soil
Organic C
concentration
O-horizon
324.6±11.0 a
427.3±9.5 b
0-10 cm
72.4±6.7 a
68.5±1.9 a,d,e
61.6±5.0 b
58.0±7.5 b
67.5±7.6 b,d,e
26.9±4.0 c
10-20 cm
35.1±3.0 a
35.7±3.0 a,d,e
33.0±2.0 b
33.2±4.7 b
33.4±4.5 b,d,e
15.4±0.4 c
20-30 cm
26.2±1.2
23.55±3.0
21.6±2.8
20.4±1.2
24.2±5.7
16.7±2.0
30 + cm
19.8±0.1
17.7±0.1
14.7±0.3
13.7±0.03
14.9±0.2
14.6±0.03
Organic N
concentration
O-horizon
mg C g -1 soil
14.4±0.4 a
17.1±0.4 b
14.6±1.9 a
10.8±1.2 a
0-10 cm
2.1±0.4
2.2±0.2
1.8±0.3
1.6±0.1
1.9±0.4
0.5±0.4
10-20 cm
0.6±0.3
0.6±0.2
0.2±0.1
0.4±0.3
0.6±0.5
†
C content
g C m -2
O-horizon
1686.7±236.6 a
2735.6±608.2 b
317.8±17.8 c
410.6±21.9 c
0-10 cm
4638.9±427.1 a
4389.7±121.8 a
3948.0±319.6 a
3717.8±479.7 a
4325.3±489.9 a
1724.3±253.6 b
10-20 cm
2286.6±195.7 a
2326.7±195.7 a
2149.8±127.5 a
2161.8±308.3 a
2339.5±292.2 a
1005.8±27.6 b
20-30 cm
1956.3±91.4
1660.8±253.7
1655.2±189.2
1581.0±30.6
1775.8±422.6
1221.4±148.4
30+ cm
1447.3±61.5
1300.6±103.8
1077.1±225.4
1005.1±20.1
1097.6±176.7
1071.5±24.0
N content
g C m -2
O- horizon
74.7±4.1 a
109.8±8.7 b
15.1±0.3 c
21.9±5.0 c
0-10 cm
137.2±27.8 a
140.8±11.2 a
114.8±18.3 a
104.9±44.5 a,b
149.6±27.5 a
25.9±26.7 b
10-20 cm
38.8±17.1
36.6±13.0
14.7±8.7
23.2±17.3
39.4±16.7
†
O-horizon
22.6
24.9
21.1
29.8
0-10 cm
35.9
31.4
36.3
35.2
29.8
†
Root mass
g m-2
Roots (< 1 mm)
128.6±23.4 a
236.4 ± 33.1 b
50.4±5.1 d
34.9±6.6 d
13.1±6.7 c
33.4±22.0 d
Roots (1-2 mm)
33.0±30.7 a
20.5±8.3 a
4.4±2.1 b
7.3±4.5 b
0.4±0.4 c
1.5±1.5 b
C:N ratio
584
585
586
587
588
589
590
591
Table 2. C and N concentration (mg C g-1 soil) and content (g m -2), C:N ration, and fine root
mass of Harvard Forest DIRT plot soils. At depths of 20 cm and greater, N content was too near
detection limits to report values with certainty. “†” represents values that could not be expressed
because N at depth was not within detection limits. Letters denote significant differences among
means according to Tukey’s HSD ( = 0.05).
28
592
593
DAYS TO 1%
RESPIRED
0-10 cm
[days]
10-20 cm
594
595
596
597
18.5
15.6
26.3
27.5
34.7
16.6
16.5
29.7
16.7
20.4
9.6
8.5
NONHYDROLYZABLE C
O-horizon
[% ]
24.0 ± 12.0
23.9 ± 14.7
--
--
17.1 ± 7.5
15.7 ± 11.5
0-10cm
7.9 ± 3.7
6.7 ± 4.8
5.1 ± 2.8
5.6 ± 2.1
10.9 ± 6.6
6.2± 7.7
10-20cm
6.0 ± 5.8
3.6 ± 1.6
3.0 ± 0.8
7.5 ± 6.3
5.0 ± 4.1
1.1 ± 0.3
NONHYDROLYZABLE
N
O-horizon
[% ]
1.2 ± 0.5
1.1 ± 0.6
--
--
0.9±0.4
0.8 ± 0.5
0-10 cm
0.5 ± 0.3
0.4 ± 0.2
0.3 ± 0.1
0.3 ± 0.1
0.6±0.4
0.3 ± 0.4
10-20 cm
0.3 ± 0.3
0.2 ± 0.1
0.2 ± 0.1
0.4 ± 0.2
0.3±0.2
0.1 ± 0.02
Table 3. Number of days before 1% of soil C was respired, and % non-hydrolyzable carbon and
% non-hydrolyzable nitrogen for soils in Harvard Forest DIRT plots.
29
TREATMENT
Phosphomonoesterase
[µmol pNP g-1 h-1]
CONTROL
DOUBLE LIT
NO LITTER
NO ROOT
NO INPUT
OA-LESS
18.7 ± 1.2a
24.5 ± 1.9b
10.8 ± 1.5c
15.8 ± 2.2
13.8 ± 1.0d
15.0 ± 2.3d
3.5± 0.4a
4.5± 0.3b
1.1± 0.2c
2.8± 0.3d
1.5± 0.1e
2.0± 0.5f
5.4
5.4
5.6
9.1
9.7
7.7
--
30.9
-15.5
-26.0
-42.5
-19.6
--
30.8
-19.0
-56.2
-68.1
-43.9
ß-glucosidase
[µmol pNP g-1 h-1]
Phosphomonoesterase:ßglucosidase
% change in
Phosphomonoesterase
compared to Control
% change in ßglucosidase compared to
Control
598
599
600
601
Table 4. Enzyme activities, ratio of nutrient-yielding potential to energy-yielding potential and
percent change in enzyme activities compared to control in surface soils (0-10 cm).
30
602
603
604
605
Figure 1. Cumulative respiration for 0-10 cm and 10-20 cm Harvard Forest DIRT soils in
laboratory incubations. Values expressed as g C * g-1 C in initial soil samples.
31
606
607
608
609
610
Figure 2. Soil carbon concentrations in density fractions in 0-10cm and 10-20 cm soils at the Harvard
Forest DIRT plots.
32
611
612
613
List of Table and Figure Captions:
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
Table 2. C and N concentration (mg C g-1 soil) and content (g m -2), C:N ration, and fine root
mass of Harvard Forest DIRT plot soils. At depths of 20 cm and greater, N content was too near
detection limits to report values with certainty. “†” represents values that could not be expressed
because N at depth was not within detection limits. Letters denote significant differences among
Table 1. Description of treatments at the Harvard Forest DIRT Plots.
Table 3. Number of days before 1% of soil C was respired, and % non-hydrolyzable carbon and
% non-hydrolyzable nitrogen for soils in Harvard Forest DIRT plots.
Table 4. Enzyme activities, ratio of nutrient-yielding potential to energy-yielding potential and
percent change in enzyme activities compared to control in surface soils (0-10 cm).
Figure 1. Cumulative respiration for 0-10 cm and 10-20 cm Harvard Forest DIRT soils in
laboratory incubations. Values expressed as g C * g-1 C in initial soil samples.
Figure 2. Soil carbon concentrations in density fractions in 0-10cm and 10-20 cm soils at the
Harvard Forest DIRT plots.
33