bi
Q5 Ka
rsity
SA
Article history:
Received 26 August 2013
Received in revised form
12 April 2014
Accepted 17 April 2014
Available online xxx
Q1
and Loewy, 2008). In this way, the stoichiometric imbalances
e composition and
ite, 1993; Wardle,
viduals, but a spe-
realistic, although
community level
g and quantifying
et are necessary to
ual ramifies to the
et al., 2012).
Within a community, herbivores and predators often face
different nutritional challenges. This is evident even in the relative
ability to acquire and maintain a balance of micronutrients; het-
erotrophs, more so than autotrophs, represent nutritionally
balanced diets for their consumers (Galef, 1996). This pattern is
particularly apparent for the micronutrient sodium (Na). Plants
require and concentrate very little Na in general, but heterotrophs
need to concentrate Na well above plant levels to maintain neural
function, osmotic balance, reproduction and development (Ganguli
* Corresponding author. Department of Biology, University of Oklahoma, 730 Van
Vleet Oval, Norman, OK 73019, USA. Tel.: þ1 800 522 0772x4821; fax: þ1 405 325
6202.
E-mail addresses: naclay@ou.edu (N.A. Clay), Steve.Yanoviak@louisville.edu (S.
P. Yanoviak), mkaspari@ou.edu (M. Kaspari).
1 Present address: Department of Biology, 139 Life Sciences Building, University
Contents lists availab
Soil Biology &
journal homepage: www.els
Soil Biology & Biochemistry xxx (2014) 1e6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
SBB5844_proof ■ 3 May 2014 ■ 1/6of Louisville, Louisville, KY 40292, USA.1. Introduction
Nutritional ecology combines principles of physiology and
nutrient availability to predict how a given organism will respond
to changes in essential nutrients (Aumann and Emlen, 1965;
Raubenheimer et al., 2009). Low nutrient availability relative to
the requirement of the organism forces organisms to spend more
time acquiring these nutrients or suffer performance deficits (Arms
et al., 1974; Sterner and Elser, 2002; Simpson et al., 2006; Geerling
between consumers and their food can affect th
ecological functions provided by food webs (Wh
2002). Nutritional ecology often focuses on indi
cies rarely exists in solitude, and thus it is more
infrequent, to examine nutrient limitation at the
of organization (Simpson et al., 2010). Identifyin
the imbalances between consumers and their di
understand how nutrient limitation of an individ
community and ecosystem levels (e.g., HawlenaKeywords:
Nutritional ecology
Invertebrates
Tropical rainforest
Trophic ecology
Nutrient subsidies
Stable isotopeshttp://dx.doi.org/10.1016/j.soilbio.2014.04.021
0038-0717/
 2014 Published by Elsevier Ltd.
Please cite this article in press as: Clay, N.A
Biochemistry (2014), http://dx.doi.org/10.10Understanding individual nutritional requirements can generate good predictions for how communities
should be structured over gradients of nutrient availability. Sodium (Na) bioaccumulates from plants to
predators: it is relatively unimportant for plants, which concentrate very little Na in their tissues, but
critical for consumers, which concentrate Nawell above plant levels. Thus, plant consumers are likely Na-
limited but their predators, which consume salty prey, are likely not. From this framework, we can make
predictions about how an entire community should respond to Na subsidies in Na-poor environments.
We tested these predictions in an interior Amazonian brown (detrital) food web. Specifically, we
quantified the responses of microbi-detritivores and their predators to experimental pulses of Na by
adding 250 ml 0.5% NaCl solution that mimicked patchy urine deposition, or river water as controls, to 55
paired 0.25 m2 plots every other day. We regularly harvested plots over a 19 day period. Consistent with
the hypothesis of Na-limitation among plant consumers, the response (effect size) to Na addition by
microbi-detritivores like termites was >6 times that of predators. Moreover, consistent with the bio-
accumulation of Na, fewer predatory invertebrates increased on þNaCl plots, thus trophic position alone
was a good predictor of response. Our results support the Na bioaccumulation hypothesis and suggest
that patchy, short-term Na inputs (like urine) facilitate heterogeneity in these leaf litter communities.
Further, this study demonstrates that combining principles from nutritional and community ecology can
generate predictions about how communities should be structured over gradients of nutrient availability
with good accuracy.

 2014 Published by Elsevier Ltd.a r t i c l e i n f o a b s t r a c tShort-term sodium inputs attract micro
predators
Natalie A. Clay a,*, Stephen P. Yanoviak b,1, Michael
aDepartment of Biology, Graduate Program in Ecology and Evolutionary Biology, Unive
bDepartment of Biology, University of Arkansas at Little Rock, Little Rock, AR 72204, U
c Smithsonian Tropical Research Institute, Apartado 2072, Balboa, Panama., et al., Short-term sodium in
16/j.soilbio.2014.04.021-detritivores and their
spari a,c
of Oklahoma, Norman, OK 73019, USA
le at ScienceDirect
Biochemistry
evier .com/locate/soi lb io117
118
119
puts attract microbi-detritivores and their predators, Soil Biology &
2.2. Experimental design
We used a split-plot design (Lindquist, 1953) with Day sampled
as the wholeplot factor, Blocks as the subplot factor, and within
blocks were paired Treatment plots (Fig. 1). We used 55 paired
0.25 m2 (0.5  0.5 m) plots that consisted of a þNaCl and a Control
plot separated by 3 m and Blocks were separated by 5 m. Control
plots received 250 ml stream water (see Kaspari et al., 2009 for
stream description and water chemistry) and þNaCl plots received
250ml 0.5% NaCl solution byweight in streamwater from the same
source. Although we added sodium in the form of NaCl, previous
experiments separating the two elements demonstrate that so-
dium, and not chloride, is driving consumer response and decom-
position (see Kaspari et al., 2009). Plots were fertilized on day 0 and
every other day until the termination of the experiment on day 19.
We destructively harvested 5 random Blocks on Days 1, 2, 4, 7, 9,
14, 15, 16, 17, 18, and 19. This allowed us to examine both rapid
effects that mimic a migrating mammal excretion event (1 day),
and longer term subsidies that mimic latrines where mammals
such as howler monkeys (as well as other primates) common
throughout the Neotropics display excretion site-fidelity and de-
& Biochemistry xxx (2014) 1e6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
SBB5844_proof ■ 3 May 2014 ■ 2/6et al., 1969; Bursey and Watson, 1983; Blumwald et al., 2000;
Geerling and Loewy, 2008). Schowalter et al. (1981) measured
plant, herbivore, and predator Na concentrations at 0.02, 0.31,
0.66% Na respectively, which demonstrates the large stoichiometric
disparity between plants and herbivores. Herbivores and detri-
tivores, as plant consumers, are thus potentially the most vulner-
able to Na-limitation. Predators, in contrast, likely do not suffer
from Na-deficits as their diet (other heterotrophs) is already rich in
Na (Schowalter et al., 1981; Simpson et al., 2006; Kaspari et al.,
2008). In fact, predators are generally insensitive to Na in choice
tests (Bradshaw et al., 1996), but herbivores can taste and respond
to Na even at concentrations as low as 0.005 M (Bell and Kitchell,
1966). Thus, Na bioaccumulates from plants to predators, but we
know of no studies that compare Na-limitation among plant con-
sumers and their predators in a given community.
In general, Na is abundantly available near coastlines due to its
deposition in rainfall, but away from oceanic sources there is a Na
shortfall (Stallard and Edmond, 1981; Kaspari et al., 2009, National
Atmospheric Deposition Program, 2011). However, local historic,
geological, biotic, or anthropogenic factors may create important
sources of Na in otherwise Na-poor environments. These include
salt or clay licks (Tracy and McNaughton, 1995; Dormaar and
Walker, 1996; Powell et al., 2009), urine (McNaughton et al.,
1997), and road salts (Canadian Environmental Protection Act,
2001; Kaspari et al., 2010). Larger animals such as sheep are able
to travel to reliable salt licks (Poole and Bachmann, 2008), but
smaller organisms like consumers of the brown food web (BFW)
likely rely on patchy and ephemeral Na subsidies such as urine,
feces and corpses. BFWs are donor-controlled and comprised of
microbes, microbi-detritivores that consume plant detritus and
associated microbes, and their predators that interact and ulti-
mately decompose ca. 90% of the productivity (uneaten plant
biomass) in green food webs (Fittkau and Klinge, 1973; Coleman
et al., 1996; Wardle, 2002; Decaëns, 2010). Consequently, under-
standing Na-limitation and its effects on BFWs can facilitate our
understanding of broad geographic community and decomposition
patterns.
Here we evaluate the responses of an interior Amazonian BFW
to experimental patches of NaCl that simulate ephemeral Na sub-
sides. In Na-poor environments like interior Amazon rainforests
(Stallard and Edmond, 1981; Kaspari et al., 2008; Dudley et al.,
2012), microbi-detritivores, but not their predators, should seek
Na (Geerling and Loewy, 2008). We test the prediction that
microbi-detritivores will accumulate in response to NaCl, but
predators, which should be less Na-limited (via the bio-
accumulation of Na), should not. By comparing BFWs of paired
Control and þNaCl plots over a period of 19 days, we tested for a
difference in community composition between paired plots and we
further tested whether trophic position (i.e., microbi-detritivore, 1
predator, 2 predator) via the bioaccumulation hypothesis was a
good predictor of response. We document how short-term Na
subsidies create predictable changes in community composition.
2. Materials and methods
2.1. Study site
We ran experimental plots from December 18, 2008 to January
6, 2009 at the Amazon Conservatory of Tropical Studies (ACTS) field
station 67 km NE of Iquitos, in Loreto Province, Peru (3.25S,
72.91W). ACTS is a lowland tropical rainforest that receives ca.
3000 mm average rainfall per year and is characterized by variable
soils in terms of geologic origins and fluvial inputs, but mainly
consist of clay oxisols and ultisols (Madigosky and Vatnick, 2000;
N.A. Clay et al. / Soil Biology2Mäki et al., 2001).
Please cite this article in press as: Clay, N.A., et al., Short-term sodium in
Biochemistry (2014), http://dx.doi.org/10.1016/j.soilbio.2014.04.0212.3. Trophic position
We used stable isotope analysis to assign 32 focal invertebrate
taxa to a trophic level (see Fig. 2 for taxa list). These focal in-
vertebrates accounted for 93% of individuals collected and are
frequent subjects in other BFW studies (e.g., Coleman et al., 1996;
Chen and Wise, 1999; Wardle, 2002; Decaëns, 2010). Ants (Hyme-
noptera: Formicidae) accounted for 21% of the total arthropod
abundance and twelve of the common ant genera were included
Fig. 1. Schematic diagram of the split-plot experimental design. The between-block
effects at the wholeplot level are Day (n ¼ 11), which is the day when 5 random
blocks were destructively sampled, and Blocks w. Day (n ¼ 5), which are the 5 random
blocks sampled on each day. In a split-plot design, these effects are indistinguishable.
The within-block effects at the subplot level are the paired Treatment plots (n ¼ 2),posit daily both urine and feces (Feely, 2005; Dos Santos Neves
et al., 2010). Prior to each harvest, we measured litter depth 1 cm
inside the four plot corners of both treatment plots using a wire
gently inserted into the litter down to mineral soil. Invertebrates
were collected from each 0.25 m2 treatment plot by collecting all
leaf litter down to the topsoil and sifting it through a 1 cm2 metal
mesh. We then discarded the coarse leaf litter and extracted in-
vertebrates from the siftate (the fine leaf litter and inverts that fell
through the 1 cm2 mesh) over 24 h using Winklers (Bestelmeyer
et al., 2000).Blocks (n ¼ 55), which contain each paired Treatment plots, and the Treatment  Day
(Blocks w. Day) interaction (see Table 1). 130
puts attract microbi-detritivores and their predators, Soil Biology &
s (gr
1 p
ebrat
rese
& B
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
SBB5844_proof ■ 3 May 2014 ■ 3/6among the 32 focal invertebrates (see Fig. 2 for focal ant genera). N
isotope values are reported using delta notation (d15N) where:
d ¼ ((Rsample/Rstandard)  1)  1000. R ¼ ratio of heavy to light
isotope (15N/14N for nitrogen stable isotopes) of the sample and
standard. Delta N values are expressed in& (per mil notation). All
Fig. 2. Trophic position (black circles) and effect size (F) based on univariate Permanova
Vertical solid lines designate trophic position divides (autotrophs, microbi-detritivores,
the mean F for each trophic position excluding the 2 predators that had only one invert
in abundance on þNaCl plots determined from univariate Permanovas. Homoptera rep
N.A. Clay et al. / Soil Biologyinvertebrates were preserved in 95% ethanol prior to stable isotope
analysis; as such, we only used N stable isotope values that are
unaffected by ethanol (Fábián, 1998). A homogenate of multiple
individuals was used in each sample to best represent each in-
vertebrate’s respective trophic position because many of the in-
vertebrates are represented at the Class, Order, or Family level.
Samples were dried at 60 C and ca. 500 mg of each homogenate
was encapsulated in tin capsules.We included plants (grass, shrubs,
and tree leaves) haphazardly sampled from the experimental site to
provide the basal d15N values within the system. Nitrogen stable
isotopes were analyzed using a Costech Elemental Analyzer inter-
faced through a ConFlo III open split valve with a Thermo Finnigan
Delta V isotope ratio mass spectrometer (Costech Analytical Tech-
nologies Inc., Valencia, CA, USA). d15N values and precision were
determined using a laboratory standard referenced against the in-
ternational standard for N of atmospheric nitrogen (international
standards calibrated with USGS 40 and 41). Nitrogen stable isotope
precision was 0.128& (N ¼ 8).
Trophic positions were calculated as: Trophic Position ¼
lþ (d15Nsecondary consumer d15Nbase)/DN.We used the average plant
value (2.876  1.875&) as d15Nbase (Scheu and Falca, 2000; Ometto
et al., 2006; Hyodo et al., 2010), l was 1 (the trophic level of the
organisms used to estimate the base, i.e., autotrophs), d15Nsecondary
consumer was directly measured, and we used the standard 3.4&
enrichment per trophic level as our DN (Fry, 1988; Post, 2002).
Organisms occupying Trophic Position 2 are hereafter referred to as
microbi-detritivores, those in Trophic Position 3 as 1 predators and
those in Trophic Position 4 as 2 predators. The placement of these
organisms within each Trophic Position is consistent with other
studies (Coleman et al., 1996; Scheu and Falca, 2000). The only
exceptions were Staphylinids and Pselaphids (Coleoptera) which
Please cite this article in press as: Clay, N.A., et al., Short-term sodium in
Biochemistry (2014), http://dx.doi.org/10.1016/j.soilbio.2014.04.021are typically classified as predators. Since the purpose of using
stable isotopes was to eliminate subjectivity in the assignment of
organisms to Trophic Positions, we kept them as microbi-
detritivores. Thus, a significant result in the analyses of Trophic
Position and effect sizewould further indicate that the assumptions
ay diamonds) of the 32 invertebrates measured within the Amazonian brown food web.
redators, 2 predators, see Methods for how determined). Horizontal dashed lines are
e member. Stars in front of invertebrate labels represent significant (p  0.05) increases
nts: Hemiptera: Sternorrhyncha and Auchenorrhyncha.
iochemistry xxx (2014) 1e6 3of our hypotheses are robust despite the few taxa that may have
debatable Trophic Position placement.
2.4. Statistical analysis
We collected invertebrates on þNaCl and Control plots to test
two predictions: 1) NaCl plots attract more invertebrates and 2) this
effect diminishes with higher trophic position. In BFWs, leaf litter
functions both as consumer habitat and food (Wardle, 2002). Litter
depth is patchy and as with aquatic systems, deeper litter (i.e., a
great volume of leaf litter) as habitat is positively correlated with
invertebrate abundance and richness (Angermeier and Schlosser,
1989; Ettema and Wardle, 2002; Kaspari and Yanoviak, 2009;
Shik and Kaspari, 2010). Average litter depth on plots varied from
0.28 to 13.37 cm, which is representative of the natural heteroge-
neity in tropical forest litter depth (Fauth et al., 1989; N.A. Clay,
unpublished data). However, NaCl addition enhances decomposi-
tion and litter depth decreased onþNaCl plots (average litter depth
of Control: 5.18  3.15 vs. þNaCl: 4.18  3.30 cm; Wilcoxon Signed
Ranks Test, d.f. ¼ 55, p ¼ 0.007, see Kaspari et al., 2009 for further
discussion). Consequently, litter depth on þNaCl plots accounted
for 24.1% of the variation in invertebrate abundance
(log10Abundance ¼ 0.606*log10LitterDepth þ 1.871, R2 ¼ 0.241, p ¼
0.0001), and 8.6% on Control plots (log10Abundance ¼
0.632*log10LitterDepth þ 1.568, R2 ¼ 0.086, p ¼ 0.03). To account
for this variation, control for habitat size-abundance relationships,
and reduce potential type II error from natural and non-random,
treatment-derived variation in litter depth, we thus used abun-
dance per habitat volume: the number of invertebrates m3 (plot
area litter depth) as is common in aquatic research (Downing and
Leibold, 2002; Bagge et al., 2004).
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
puts attract microbi-detritivores and their predators, Soil Biology &
add-on package (Anderson et al., 2008). Permanova is a mulit-
mally distributed (KolmogoroveSmirnov, d.f. ¼ 31, p ¼ 0.166).
Second, we tested whether F was a good predictor of group
focal invertebrates spanned 3 Trophic Positions: 15 microbi-
detritivores, 16 1 predators, and 1 2 predator (Fig. 2).
3.1. Effects of NaCl on BFW composition
The addition of NaCl (Treatment) changed community compo-
sition (Within-blocks, Treatment: p ¼ 0.0127; Table 1). At the
wholeplot level, the community composition also differed among
sample days (Between-blocks, Day: p ¼ 0.0001), but the effects of
NaCl addition on community composition were consistent
regardless of the sampling Day (Treatment  Day: p ¼ 0.5208;
Table 1). In other words, as is characteristic of a dynamic tropical
forest, there was daily variation in community composition for
both þNaCl and Control plots, but the magnitude of the Treatment
effect between plots remained constant over time (i.e., no
large bottom-up effects within the detrital community in response
Treatment 1 5619.1 5619.1 5.884 0.0127
& Biochemistry xxx (2014) 1e6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
SBB5844_proof ■ 3 May 2014 ■ 4/6membership and how well it correctly classified invertebrate taxa
to their respective trophic position using logistic regression (SPSS v.
10.0). We used F of the 31 invertebrates in microbi-detritivore and
1 predator groups as our predictor variable and our binomial
response was yes or no invertebrate is a microbi-detritivore.
We performed a second logistic regression where Collembola
was broken into families (Onychiuridae, Hypogastruridae, Smin-
thuridae, Entomobryidae, Isotomidae) and Predatory mites were
broken into Mesostigmatids and Prostigmatids to insure that dif-
ferences in the taxonomic resolution among invertebrates did not
introduce potential bias. These taxa were analyzed as groups in the
stable isotope analysis to achieve sufficient mass. We chose these
taxa because they were the most abundant in their respective
Trophic Positions representing 42% and 27% of microbi-detritivore
and predator abundances respectively. Because of their high
abundances, these taxa would be the most likely to introduce po-
tential bias when lower taxonomic ranks were grouped.
3. Results
We collected 16,367 invertebrates from the 55 pairedþNaCl and
Control plots. þNaCl plots had more invertebrates (Permanova,
p < 0.0001). Median total invertebrate were 26 (IQR ¼ 15e68) m3variate permutational analysis of variance designed for community
data (Anderson, 2001). We square-root transformed the abundance
data from the 32 invertebrate taxa prior to analysis to control for
rare taxa and used BrayeCurtis dissimilarities as the distance
measure. The between-blocks effects were Day (wholeplot: n ¼ 11)
and Block (n ¼ 5) and the within-blocks effects were Treatment
(subplot: n ¼ 2) and the Day  Treatment interaction (Kirk, 1982;
Gotelli and Ellison, 2004, Fig. 1). We used 9999 permutations to
test the null hypotheses of 1) no difference in BFW communities
among Days and Blocks with Day at the wholeplot level, 2) no
difference in BFW communities between treatments (pairs) at the
subplot level, and 3) no interaction between Treatment and Day
(Blocksw. Day) at the subplot level. A significant effect of Treatment
was followed by a posteriori univariate Permanovas on the 32 focal
invertebrates to determine which invertebrates within commu-
nities responded to Treatment effects and calculate their effect
sizes.
2.6. Testing for diminished responses among predators
We then testedwhether trophic positionwas a good predictor of
response with the prediction from the Na bioaccumulation hy-
pothesis that microbi-detritivores but not predators should recruit
to þNaCl addition. First, we determined whether microbi-
detritivores had a stronger response to þNaCl addition than pred-
ators using a t-test to test the null hypothesis of no difference in
effect size between microbi-detritivores and 1 predators. We only
compared the microbi-detritivores and 1 predators because only 1
invertebrate taxon occupied the 2 predators position (2 predator
Hemiptera: Heteroptera, Fig. 2). We used the pseudo-F value (see
Anderson, 2001) as effect size from the a posteriori univariate
Permanovas on the 31 invertebrate taxa. Effect size (F) was nor-2.5. Testing differences in community structure
To determine if communities on þNaCl plots differed from
Control plots we used Primer 6 (version 6.1.11) computer program
(Primer-E Ltd., Plymouth, United Kingdom) with the PERMANOVA
N.A. Clay et al. / Soil Biology4onþNaCl plots versus 15 (IQR¼ 7e30) m3 on Control plots. The 32
Please cite this article in press as: Clay, N.A., et al., Short-term sodium in
Biochemistry (2014), http://dx.doi.org/10.1016/j.soilbio.2014.04.021Blocks 54 108,780 2014.5 2.053 0.0001
Treatment  Blocks w. Day 4 3819.8 945.94 0.973 0.5208
Residual 50 49,064 981.28to direct consumer resource provisioning. Additionally, the one 2
Table 1
Results of the split-plot permutational manova in Primer-E with PERMANOVA add-
on based on BrayeCurtis dissimilarities for the abundances of 32 invertebrates
within an Amazonian brown food web (see Fig. 1 for experimental design). Inver-
tebrate abundance (number of individuals) was analyzed as per volume (m3 ¼ plot
area  litter depth).
Source d.f. SS MS Pseudo-F p
Between-block effects (Wholeplot level)
Day 10 23,518 2351.8 3.590 0.0001
Blocks w. Day 4 4670.2 1167.2 1.782 0.0009
Residual 40 26,203 655.08
Within-block effects (Subplot level)interaction).
3.2. Do invertebrates lower in trophic position respond more
strongly to NaCl?
Microbi-detritivores had a stronger response to þNaCl than 1
predators (t31 ¼ 2.209, p ¼ 0.039). The mean effect sizes (F) for
microbi-detritivores and 1 predators were 6.8 and 0.4 respectively
(Fig. 2). Further, F was a good predictor of trophic position (logistic
regression, d.f. ¼ 1, c2 ¼ 5.153, p ¼ 0.023). The model correctly
classified 67.7% of invertebrates to their respective trophic position,
and predicts the odds of increasing on þNaCl plots is 1.2 times
higher for microbi-detritivores than predators. The second logistic
regression with Collembola and Predatory mites analyzed at finer
taxonomic resolution (36 taxa total) yielded similar results: the
model correctly classified 63.9% of invertebrates to their respective
trophic position, with the same odds ratio of increasing on þNaCl
plots (Exp(b) ¼ 1.2; d.f. ¼1, c2 ¼ 5.281, p ¼ 0.022). These results are
further supported by the results of the univariate Permanovas on
response of the 32 invertebrates (Fig. 2). As predicted, the majority
of invertebrates feeding low in the food web (Microbi-detritivores)
increased on þNaCl plots (p  0.044; Fig. 2), including each family
of Collembola separately (p  0.027). In contrast, most 1 predators
did not increase on NaCl addition plots. The exceptions to this
pattern were Predatory mites (Acari: Mesostigmata and Pros-
tigmata; but when analyzed separately only Prostigmata increased
on þNaCl plots), and the ant genera Strumigenys, Hylomyrma, and
Solenopsis (p  0.033, Fig. 2). The large responses of microbi-
detritivores and the small responses of 1 predators support our
predictions of how Na-limitation is structured within a food web:Total 109 167,290 130
puts attract microbi-detritivores and their predators, Soil Biology &
tera (Arms et al., 1974), and ants (Kaspari et al., 2008) among other
& B
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
SBB5844_proof ■ 3 May 2014 ■ 5/6predator, Heteroptera (Hemiptera), increased on þNaCl plots
(p ¼ 0.021, Fig. 2).
4. Discussion
Sodium has long been known to impact the behavior and per-
formance of individual organisms (Arms et al., 1974; Belovsky and
Jordan, 1981; Simpson et al., 2006; Kaspari et al., 2008), but its
role in structuring communities had remained unexplored. Here
we demonstrated that by using the principles of nutritional ecology
at the individual level, we could predict with good accuracy how an
entire community should respond to subsidies of a limiting
nutrient. Because Na bioaccumulates from plants to predators,
heterotrophs low in trophic position like microbi-detritivores
rapidly recruited to pulses of NaCl that mimicked urine deposi-
tion in a Na-poor forest, whereas predators in general did not
(Fig. 2). NaCl addition attracted >33% more invertebrates than
Control plots largely driven by microbi-detritivores, which had >6
times the average effect size of 1 predators (Fig. 2), and this pattern
remained consistent over time regardless of daily variation in this
dynamic tropical forest (Table 1). Further, Trophic Position alone
was a good predictor of invertebrate response (effect size: F). These
results document the strong effects of NaCl additions on a large
fraction of the BFW and demonstrate that we can largely predict
community composition changes in response to nutrient addition
using principles of nutritional ecology pertaining to individual
invertebrates.
Although from the Na bioaccumulation hypothesis, predators
are not predicted to increase on þNaCl plots from Na-limitation,
predictions from classic food web ecology provide an obvious
mechanism for predator increase: increases in prey. Our experi-
mental design and time scalewere not adequate to test this directly,
but three of the four predators that responded were ants, which
forage widely and rapidly, and strongly recruit to increased prey
availability (Lach et al., 2010). Increased prey may also explain the
increase in the only 2 predator, Heteroptera (Hemiptera, including
Reduviids, Aradids, and Mirids; Fig. 2). Future studies should
combine principles from nutritional and energy-based trophic
ecology to fully understand the multiple mechanisms that shape
community structure based on resource availability (Simpson et al.,
2010).
In contrast to predators, microbi-detritivores strongly recruited
to þNaCl plots (Fig. 2). Here we only examined invertebrate
response and did not explore changes in the microbial community.
Increased microbial necromass or biomass from Na application
could be an indirect mechanism for increases in microbi-
detritivores. However, microbi-detritivores responded from day 1
and remained consistent over time (Table 1). As such, sustained
microbi-detritivore recruitment to microbial necromass seems
unlikely as the necromass is finite. While increased microbial
biomass may play a role in microbi-detritivore recruitment,
numerous studies have used Na-soaked substances and demon-
strated that Na itself is sufficient to attract herbivores (e.g., Arms
et al., 1974; Barrows, 1974; Kaspari et al., 2008, 2010), and
microbi-detritivores would still obtain Na by increasing con-
sumption of salty organisms (e.g., Simpson et al., 2006).
Although microbi-detritivores increased nearly 1.5 fold
on þNaCl over Controls plots, a handful of microbi-detritivores had
similar abundances between treatments: Psocoptera, Blattaria,
Diptera larvae, Lepidoptera larvae, Staphylinidae (Coleoptera), Iso-
poda, and Pselaphidae (Coleoptera). Together, these invertebrates
represented only 5% of the total microbi-detritivore abundance. A
suite of mechanisms could potentially cause this pattern such as
increased predation pressure, ontogenetic constraints, competitive
N.A. Clay et al. / Soil Biologyexclusion, physiological Na adaptations, or a cryptic predatory
Please cite this article in press as: Clay, N.A., et al., Short-term sodium in
Biochemistry (2014), http://dx.doi.org/10.1016/j.soilbio.2014.04.021invertebrates and vertebrates. Na is lost during anymetabolic water
loss process such as sweating or excretion and these processes
increase with temperature (Peters, 1983). Thus, in Na-poor envi-
ronments, particularly in the hot tropics, short-term Na subsidies
are likely vital to consumers maintaining a Na-balance (Dudley
et al., 2012). Further experiments investigating the role of urine
deposition in structuring BFWs and their ecological functions like
decomposition over a gradient of Na availability will contribute to a
multi-scale understanding of the patchiness of BFWs.
The increased microbi-detritivores like termites and Collembola
on þNaCl plots likely drove the pattern of decreased litter depth
with the addition of NaCl. Here we tested Na’s effects on food web
structure and invertebrate response and controlled for differences
in habitat space between treatments, but Kaspari et al. (2009)
demonstrated that the addition of Na increases both leaf litter
and experimental filter paper decomposition and documented a
similar response in termite recruitment to NaCl treated cellulose.
The inverse relationship between habitat space and decomposition
rates in BFWsdthe “more food-less habitat effect” (Shik and
Kaspari, 2010; Jacquemin et al., 2012)dsuggests a dynamic mix of
limiting factors for invertebrates as litter decomposes. Our results
and those of Kaspari et al. (2009) demonstrate that even ephemeral
patchy Na subsidies can have rapid and large effects on both BFW
structure and function in Na-poor environments.
Our study is the first to apply concepts of Na-limitation to entire
foodwebs and it demonstrates how nutritional ecology can provide
a predictive framework for community ecology. Over 80% of earth’s
terrestrial surface is potentially Na-poor as oceanic inputs (i.e.,
“salty rain”) exponentially decrease with distance inland (Stallard
and Edmond, 1981; Kaspari et al., 2009, National Atmospheric
Deposition Program, 2011). Na should be included as one of the
major nutritional resources foci in future research in community
ecology, biogeography and classic trophic theory. Lastly, future
research should include effects of both short-and long-term Na
subsidies on food webs at both macro-and micro-spatial grains to
fully understand how Na availability impacts individuals to
ecosystem level processes.
Acknowledgments
We thank the Amazon Conservatory for Tropical Studies, Pam
Bucur, and Steve Madigosky for facilitating fieldwork and the
Peruvian Instituto Nacional de Recursos Naturales for permits.
Special thanks to Kendra Tholt, Michael O’Brien, and Deana Flatt for
assistance in the lab. Thanks to Don Shepard and Zoology Ecology
and Evolutionary Biology Discussion Group for valuable discussions
of this manuscript. Additionally, special thanks to Rick Maynard
and the University of Oklahoma stable isotope facility, J. Kelly and E.
Marsh-Matthews for the use of equipment. This project was funded
by grants from the National Geographic Society and NSF DEB-behavior, among others. However, without testing each specifically,
it is impossible to discern the mechanism from our results. But,
these exceptions provide opportunities to delve deeper into indi-
vidual differences in Na physiology and species interactions to
better understand when we should and should not predict Na-
seeking behavior.
At the time scale of weeks, increases in invertebrates on þNaCl
plots most likely arose from immigration rather than population
increases. The rapid response of BFW invertebrates in this study
suggests that they have evolved to quickly capitalize on ephemeral
resources like NaCl pulses. Rapid recruitment of plant consumers to
NaCl has also been documented in bees (Barrows, 1974), Lepidop-
iochemistry xxx (2014) 1e6 5Q2,3
0948762 to MEK. Q4 130
puts attract microbi-detritivores and their predators, Soil Biology &
References
Anderson, M.J., 2001. A new method for non-parametric multivariate analysis of
variance. Austral Ecology 26, 32e46.
Anderson, M.J., Gorley, R.N., Clarke, K.R., 2008. PERMANOVA for PRIMER: Guide to
Software and Statistical Methods. PRIMER-E Ltd., Plymouth, United Kingdom.
Hawlena, D., Strickland, M.S., Bradford, M.A., Schmitz, O.J., 2012. Fear of predation
slows plant-litter decomposition. Science 336, 1434e1438.
Hyodo, F., Matsumoto, T., Takematsu, Y., Kamoi, T., Fukuda, D., Nakagawa, M.,
Itioka, T., 2010. The structure of a food web in a tropical rain forest in Malaysia
based on carbon and nitrogen stable isotope ratios. Journal of Tropical Ecology
26, 205e214.
Jacquemin, J., Maraun, M., Roisin, Y., Leponse, M., 2012. Differential response of ants
to nutrient addition in a tropical brown food web. Soil Biology and Biochem-
N.A. Clay et al. / Soil Biology & Biochemistry xxx (2014) 1e66
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
SBB5844_proof ■ 3 May 2014 ■ 6/6Angermeier, P.L., Schlosser, I.J., 1989. Species-area relationships for stream fishes.
Ecology 70, 1450e1462.
Arms, K., Feeny, P., Lederhouse, R.C., 1974. Sodium: stimulus for puddling behavior
by tiger swallowtail butterflies, Papilio glaucus. Science 185, 372e374.
Aumann, G.D., Emlen, J.T., 1965. Relation of population density to sodium selection
by microtine rodents. Nature 208, 198e199.
Bagge, A.M., Poulin, R., Valtonen, E.T., 2004. Fish population size, and not density, as
the determining factor of parasite infection: a case study. Parasitology 128,
305e313.
Barrows, E.M., 1974. Aggregation behavior and response to sodium chloride in fe-
males of a solitary bee, Augochlora pura (Hymenoptera: Halictidae). The Florida
Entomologist 57, 189e193.
Bell, F.R., Kitchell, R.L., 1966. Taste reception in the goat, sheep and calf. Journal of
Physiology 183, 145e151.
Belovsky, G.E., Jordan, P.A., 1981. Sodium dynamics and adaptations of a moose
population. Journal of Mammalogy 62, 613e621.
Bestelmeyer, B.T., Agosti, D., Alonso, L.E., Brandao, C.R.F., Brown, W.L., Delabie, J.A.C.,
Silvestre, R., 2000. Field techniques for the study of ground dwelling ants e an
overview, description, and evaluation. In: Agosti, D., Majer, J.D., Alonso, L.E.,
Schultz, T.R. (Eds.), Ants: Standard Methods for Measuring and Monitoring
Biodiversity. Smithsonian, Washington DC, pp. 122e144.
Blumwald, E., Aharon, G.S., Apse, M.P., 2000. Sodium transport in plant cells. Bio-
chimica et Biophysica Acta 1465, 140e151.
Bradshaw, J.W.S., Goodwin, D., Legrad-Defretin, V., Nott, H.M.R., 1996. Food selec-
tion by the domestic cat, an obligate carnivore. Comparative Biochemistry and
Physiology 114, 205e209.
Bursey, R.G., Watson, M.L., 1983. The effect of sodium restriction during gestation on
offspring brain development in rats. American Journal of Clinical Nutrition 37,
43e51.
Canadian Environmental Protection Act 1999, 2001. Priority Substances List
Assessment Report: Road Salts. Minister of Public Works and Government
Services.
Chen, B., Wise, D.H., 1999. Bottom-up limitation of predaceous arthropods in a
detritus-based terrestrial food web. Ecology 80, 761e772.
Coleman, D.C., Crossley Jr., D.A., Hendrix, P.F., 1996. Fundamentals of Soil Ecology.
Academic Press, San Diego, CA.
Decaëns, T., 2010. Macroecological patterns in soil communities. Global Ecology and
Biogeography 19, 287e302.
Dormaar, J.F., Walker, B.D., 1996. Elemental content of animal licks along the eastern
slopes of the Rocky Mountains in southern Alberta, Canada. Canadian Journal of
Soil Science 76, 509e512.
Dos Santos Neves, N., Feer, F., Salmon, S., Chateil, C., Ponge, J., 2010. The impact of
red howler monkey latrines on the distribution of main nutrients and on topsoil
profiles in a tropical rain forest. Austral Ecology 35, 549e559.
Downing, A.L., Leibold, M.A., 2002. Ecosystem consequences of species richness and
composition in pond food webs. Nature 416, 837e841.
Dudley, R., Kaspari, M., Yanoviak, S.P., 2012. Lust for salt in the western Amazon.
Biotropica 44, 6e9.
Ettema, C.H., Wardle, D.A., 2002. Spatial soil ecology. Trends in Ecology and Evo-
lution 17, 177e183.
Fábián, M., 1998. The effects of different methods of preservation on the 15N con-
centration in Folsomia candida (Collembola). Applied Soil Ecology 9, 101e104.
Fauth, J.E., Crother, B.I., Slowinski, J.B., 1989. Elevational patterns of species richness,
evenness, and abundance of the Costa Rican leaf-litter herpetofauna. Biotropica
21, 178e185.
Feely, K., 2005. The role of clumped defecation in the spatial distribution of soil
nutrients and the availability of nutrients for plant uptake. Journal of Tropical
Ecology 21, 99e102.
Fittkau, E.J., Klinge, H., 1973. On biomass and trophic structure of the Central
Amazonian rainforest ecosystem. Biotropica 5, 2e14.
Fry, B., 1988. Food web structure on Georges Bank from stable C, N, and S isotopic
compositions. Limnology and Oceanography 33, 118e1190.
Galef Jr., B.G., 1996. Food selection: problems in understanding how we choose
foods to eat. Neuroscience and Biobehavioral Reviews 20, 67e73.
Ganguli, M.C., Smith, I.D., Hanson, L.E., 1969. Sodium metabolism and its require-
ment during reproduction in female rats. Journal of Nutrition 99, 225e233.
Geerling, J.C., Loewy, A.D., 2008. Central regulation of sodium appetite. Experi-
mental Physiology 93, 178e209.
Gotelli, N.J., Ellison, A.M., 2004. A Primer of Ecological Statistics. Sinauer Associates,
Sunderland, Massachusetts, USA.Please cite this article in press as: Clay, N.A., et al., Short-term sodium in
Biochemistry (2014), http://dx.doi.org/10.1016/j.soilbio.2014.04.021istry 46, 10e17.
Kaspari, M., Yanoviak, S.P., Dudley, R., 2008. On the biogeography of salt limitation:
a study of ant communities. Proceedings of the National Academy of Science,
USA 105, 17848e17851.
Kaspari, M., Yanoviak, S.P., 2009. Biogeochemistry and structure of tropical brown
food webs. Ecology 90, 3342e3351.
Kaspari, M., Yanoviak, S.P., Dudley, R., Yuan, M., Clay, N.A., 2009. Sodium shortage as
a constraint on the carbon cycle in an inland tropical rainforest. Proceedings of
the National Academy of Science, USA 106, 19405e19409.
Kaspari, M., Chang, C., Weaver, J., 2010. Salted roads and sodium limitation in a
northern forest ant community. Ecological Entomology 35, 543e548.
Kirk, R.E., 1982. Experimental design: procedures for the behavioral sciences.
Brooks/Cole, Monterey, CA.
Lach, L., Parr, C.L., Abbott, K.L., 2010. Ant Ecology. Oxford University Press, New York.
Lindquist, E.F., 1953. Design and Analysis of Experiments in Psychology and Edu-
cation. Houghton Mifflin, Boston.
Madigosky, S.R., Vatnick, I., 2000. Microclimatic characteristics of a primary tropical
Amazonian rain forest, ACEER, Iquitos, Peru. Selbyana 21, 165e172.
Mäki, S., Kalliola, R., Vuorinen, K., 2001. Road construction in the Peruvian Amazon:
causes and consequences. Environmental Conservation 28, 199e214.
McNaughton, S.J., Banyikwa, F.F., McNaughton, M.M., 1997. Promotion of the cycling
of diet-enhancing nutrients by African grazers. Science 278, 1798e2000.
National Atmospheric Deposition Program, 2011. Sodium Ion Wet Deposition
(Champaign, IL).
Ometto, J.P.H.B., Ehleringer, J.R., Domingues, T.F., Berry, J.A., Ishida, F.O.Y., Mazzi, E.,
Higuchi, N., Flanagan, L.G., Nardoto, G.B., Martinelli, L.A., 2006. The stable car-
bon and isotopic composition of vegetation in tropical forests of the Amazon
Basin, Brazil. Biogeochemistry 79, 251e274.
Peters, R.H., 1983. The Ecological Implications of Body Size. Cambridge University
Press, Cambridge.
Poole, K.G., Bachmann, K.D., 2008. Mineral Lick Use by GPS Radio Collared Mountain
Goats in Southeastern British Columbia. Updated with Field Checks 2007.
Aurora Wildlife Research and Bachmann Consulting Services.
Post, D.M., 2002. Using stable isotopes to estimate trophic position: models,
methods and assumptions. Ecology 83, 703e718.
Powell, L.L., Powell, T.U., Powell, G.V.N., Brightsmith, D.J., 2009. Parrots take it with a
grain of salt: available sodium content may drive collpa (clay lick) selection in
Southeastern Peru. Biotropica 41, 279e282.
Raubenheimer, D., Simpson, S.J., Mayntz, D., 2009. Nutrition, ecology and nutritional
ecology: towards an integrated framework. Functional Ecology 23, 4e16.
Scheu, S., Falca, M., 2000. The soil food web of two beech forests (Fagus sylvatica) of
contrasting humus type: stable isotope analysis of a macro- and a mesofauna-
dominated community. Oecologia 123, 285e296.
Schowalter, T.D., Webb, J.W., Crossley Jr., D.A., 1981. Community structure and
nutrient content of canopy arthropods in clearcut and uncut forest ecosystems.
Ecology 62, 1010e1019.
Shik, J.Z., Kaspari, M., 2010. More food, less habitat: how necromass and leaf litter
decomposition combine to regulate a litter ant community. Ecological Ento-
mology 35, 158e165.
Simpson, S.J., Sword, G.A., Lorch, P.D., Couzin, I.D., 2006. Cannibal crickets on a
forced march for protein and salt. Proceedings of the National Academy of
Science, USA 103, 4152e5156.
Simpson, S.J., Raubenheimer, D., Charleston, M.A., Clissold, F.J., 2010. Modelling
nutritional interactions: from individuals to communities. Trends in Ecology
and Evolution 25, 53e60.
Sterner, R.W., Elser, J.J., 2002. Ecological Stoichiometry: the Biology of Elements
from Molecules to the Biosphere. Princeton University Press, Princeton, NJ.
Stallard, R.F., Edmond, J.M., 1981. Geochemistry of the Amazon 1 Precipitation
chemistry and the marine contribution to the dissolved load at the time of peak
discharge. Journal of Geophysical Research 86, 9844e9858.
Tracy, B.E., McNaughton, S.J., 1995. Elemental analysis of mineral lick soils from the
Serengeti National Park, the Konza Prairie and Yellowstone National Park.
Ecography 18, 91e94.
Wardle, D.A., 2002. Communities and Ecosystems: Linking the Aboveground and
Belowground Components. Princeton University Press.
White, T.C.R., 1993. The Inadequate Environment: Nitrogen and the Abundance of
Animals. Springer-Verlag, Berlin Heidelberg.puts attract microbi-detritivores and their predators, Soil Biology &