Transcript (a) (b)

Who’s eating What, hoW and Where? Polychaete feeding behavior and geograPhical distribution in the gulf of Mexico.
Russell Carvalho, Anja Schulze
Texas A & M University at Galveston.
Polychaete annelids contribute greatly to macrofaunal species diversity in dynamic deep-sea ecosystems. The aim of our study was to assess the functional composition of deep-sea polychaete assemblages in the Gulf of Mexico. We examined various factors influencing the
diversity, density and distribution of this taxon. Polychaete assemblages are ubiquitous in the deep sea but abundance, diversity and species composition are sensitive to changing environmental conditions. We tested the hypothesis that polychaete feeding-guilds show a stronger
correlation with depth and environmental characteristics than their taxonomic groupings. Polychaetes were sampled from a total of 51 stations consisting of 7 transects and individual juxtaposed stations in the Gulf of Mexico. Additionally, we also carried out ecological niche
modeling (ENM) using the Maximum Entropy model (MaxEnt) to estimate the potential geographical distribution of the most dominant polychaete species found in our study.
The highest species diversity was present at the central and the upper slope of the Mississippi trough area. Polychaete density decreased exponentially from the shallow, central regions to the eastern and western regions. Diversity in polychaete feeding-guilds was high on the
Mississippi trough, upper and mid-slope regions and declined to a few guilds on the abyssal plain region. A significant correlation was observed between polychaete feeding-guilds and environmental variables. Using species presence-only data for ENM we generated potential
distribution maps of the four most dominant polychaete species found at a regional scale in the Gulf of Mexico. We conclude that that feeding-guild analysis and ecological niche modeling are powerful tools for polychaete diversity studies, as they provide a clear understanding
of the spatio-temporal patterns and trophic organization of deep-sea benthic assemblages.
Ecological Niche Models of polychaete species in the
Gulf of Mexico
4000
3500
3000
2500
2000
1500
1000
500
0
0
1000
2000
Depth (m)
3000
No of Species (S)
y = 0.0002x2 - 1.437x + 2598.4
R² = 0.6125
200
180
160
140
120
100
80
60
40
20
0
4000
y=
0
1000
(a)
2000
Depth (m)
1E-08x2
- 0.0252x + 101.71
R² = 0.4804
3000
Shannon-Wiener (Hloge)
Results on Diversity analysis
Density (Indiv m-2)
An important constituent of the deep-sea benthos; polychaetes have been
widely used as model taxa in biodiversity and descriptive ecology studies.
Providing precise clues of deep-water conditions, the community structure
of these soft sediment dwelling animals is used to assess the overall
functioning and health of the benthic environment. Recent literature 1-3, has
primarily focused on diversity and distribution trends of these animals. Their
results on the ecology and the geographic ranges of polychaete species in the
Gulf of Mexico are often ambiguous, with species identifications based on
outdated references and taxonomic keys, originally generated for other
geographic regions. Due to constraints in sampling deep-sea environments,
the accurate ranges of many polychaete species are not known. This critical
issue can be addressed by modeling the ecological niches and the
geographical ranges occupied by these taxa. Ecological Niche Modeling
(ENM) is a recent concept which makes use of species occurrence data and
associated environmental data to estimate the geographic range of a species
4-5. This method is currently being used in a range of marine studies 6-8. By
modeling the environmental limits of a species, predictions can be made
wherever its preferred set of environmental conditions occur within areas
studied 7. In addition, the model allows us to predict past as well as future
species distributions under changing climatic scenarios.
Our aim in this study was to –
1. Validate the Maxent model performance by dividing our dataset into
training (75%) and test-data (25%) 9.
2. Model species distributions based on species ‘presence-only’ data and
project these models on maps using a GIS-based software.
3. Conduct a feeding-guild analysis of all families exhibiting various feedingmodes10 and use this data for species niche modeling.
Our future goal is using ENM to assess other polychaete species exhibiting
different feeding-modes and to examine if these species show any preferences
to specific areas along depth ranges in the Gulf of Mexico.
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
y = -6E-08x2 - 8E-05x + 3.6513
R² = 0.4065
0
4000
1000
(b)
2000
Depth (m)
3000
(c)
(a)
Fig 1 (a-c). Observe density and diversity of polychaetes in the Gulf of Mexico
(b)
(c)
Fig 4. Prediction map for the species (a) Spiophanes berkeleyorum (b) Tharyx annulosus and (c) Prionospio ehlersi
Based on our results (Tables 1-3) the relationships observed between polychaete
species, their respective feeding-guilds and their environmental variables demonstrated
significant correlations, supporting our hypothesis that feeding-guilds of polychaetes
would be more significantly correlated with depth, than taxonomic groupings.
Our results demonstrate 1. The importance of polychaete feeding guild in deep-sea biology. We encourage
other similar studies to include such a conceptual framework in favor of ecological
and environmental assessments.
2. Based on the findings of this study, we anticipate Ecological Niche Modeling to
greatly facilitate studies on other marine deep-sea taxa. This new tool should be
used in addition to morphological and molecular data when assessing speciesdistribution ranges.
3. The spatial distribution of marine taxa can be successively predicted at a regional
scale using a range of other predictive models. Hence it would be interesting to
note the predictive power of each of these machine learning models in order to
generate the best predictions of species distributions.
(a)
(b)
Fig 2. (a) Observed percentage of feeding guilds in all samples (b) Observed feeding guilds vs no of species.
Table 1. Polychaete feeding-guilds as described in Fauchald
and Jumars (1979).
Types of Motility
Major Feeding mode
discretely
motile
motile
sessile
Burrowers
jawed pharynx
unarmed pharynx
Carnivores
jawed pharynx
unarmed pharynx
Filter feeders
tentaculate
Herbivores
jawed pharynx
Omnivores
jawed pharynx
Surface deposit feeders
unarmed pharynx
tentaculate
BMJ
BMX
-
BSX
CMJ
CMX
CDJ
-
-
HMJ
FDT
-
FST
-
-
ODJ
-
SMX
SMT
SDJ
SDT
SST
The feeding-guild is a three-letter code , the first letter indicates the major feeding mode, the second
indicates the motility and the third indicates the morphological structure used in feeding; position 1S, surface deposit-feeder; B, subsurface deposit-feeder; C, carnivore; F, filter-feeder; H, herbivore;
position 2-M, motile; D, discretely motile; S, sessile; and position 3-P,pumping; J, jawed; T,
tentaculate; X, other structures like unarmed proboscides and eversible sac-like pharynges.
Table 2. One-way ANOVA showing correlations for depth v/s guild
types based on the total number of species for each group.
Group 1
Guild
Error
Total
Group 2
Guild
Error
Total
Group 4
Guild
Error
Total
Group 5
Guild
Error
Total
Group 6
Guild
Error
Total
Group 7
Guild
Error
Total
Guild
Error
Total
Group 12
Guild
Error
Total
df
SS
MS
F
P
5
70
75
1482596
8977782
10460378
296519
128254
2.31
0.053
6
80
86
3371546
4546604
7918151
561924
56833
9.89
<.0001
4
75
79
1699928
1387766
3087694
424982
18504
0
<.0001
5
24
29
7890654
0
7890654
1578131
0
0
<.0001
5
63
68
244508
350361
594869
48902
5561
8.79
<.0001
4
49
53
9
127
136
41582728
14712298
56295026
11136794
38530312
49667106
10395682
300251
34.62
<.0001
1237422
303388
4.08
<.0001
3
36
39
4811372
2622233
7433604
1603791
72840
22.02
<.0001
The degrees of freedom (DF), sums of squares (SS) and mean square (MS) for each variance component is shown. The variance ratio is
given by F (test statistic) and its significance is shown by the p value.
Results on Maxent analysis
Table 3. Maxent analysis of Variable Contributions to
all species.
Variable
• The deep water region of the northern Gulf of Mexico was sampled between 2000 and
2002 during three cruises of the Deep Gulf of Mexico Benthos Program (DGoMB).
•Samples were collected from 51 stations on the R/V Gyre. Sampled macrofaunal specimens
were collected using a 0.1725 m2 GOMEX box corer or Gray-O’Hara boxcore.
•Polychaete specimens were separated to their lowest taxonomic unit using specialized keys,
subsequently all species were separated based on their feeding-guilds according to Fauchald
and Jumars 10, with exclusive modifications for the Spionidae 11.
•Community structure data analysis was carried out using Primer V6 12 . One-way ANOVA’s
were carried out using SPSS vs 20.0.0 13
•Using species presence-only data and a total of 23 environmental variables 14, we ran the
Maximum Entropy model (75% as training and 25% as test-data) for species distributions in
Maxent vs 3.3.3k 15. Species predictions maps were produced in ESRI®ArcGISTM 9.3.
4000
(b)
(a)
Fig 3. (a) Prediction map for Aphaelochaeta
marioni (b) Displays test omission rate and
predicted area as a function of the cumulative
threshold, averaged over 15 replicate runs.
(c)The receiver operating characteristic (ROC)
curve averaged over 15 replicate runs. The
specificity is defined using predicted area. The
average test for area under the curve (AUC) of
replicate runs is 0.997, and the standard
deviation is 0.001.
(c)
Sea surface temp
(MAX)
Photosynthetically
available radiation
(MEAN)
Percent
Permutation
Contribution
Importance
35.2
52.6
15.4
12.6
pH (MEAN)
10.2
0.3
Calcite
Sea surface temp
(RANGE)
Sea surface temp
(MIN)
9.8
0
8.3
0
6.8
0
Phosphate
3.3
2.3
Silicate
1
30
Salinity
0
0.4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Mahon, R. A., Mahon, H. K., Dauer, M. D. & Halanych, K. M. Discrete genetic boundaries of three Streblospio (Spionidae, Annelida) species and the status of S. shrubsolii. 2009 5, 172-178 (2009).
Pérez-Mendoza, A. Y., Hernández-Alcántara, P. & Solís-Weiss, V. Bathymetric distribution and diversity of deep water polychaetous annelids in the Sigsbee Basin, northwestern Gulf of Mexico.
Hydrobiologia
496,361-370, doi:10.1023/a:1026133907343 (2003).
Rowe, G. T. & Kennicutt II, M. C. Introduction to the Deep Gulf of Mexico Benthos Program. Deep Sea Research Part II 55, 2536-2540 (2008).
Robinson, L. M. et al. Pushing the limits in marine species distribution modelling: lessons from the land present challenges and opportunities Global Ecology and Biogeography. (Manuscript accepted for
publication) (2010).
Raxworthy, C. J., Ingram, C. M., Rabibisoa, N. & Pearson, R. G. Applications of Ecological Niche Modeling for Species Delimitation: A Review and Empirical Evaluation Using Day Geckos
(Phelsuma) from Madagascar. Syst. Biol. 56, 907-923 (2007).
Ba, J., Hou, Z., Platvoet, D., Zhu, L. & Li, S. Is Gammarus tigrinus (Crustacea, Amphipoda) becoming cosmopolitan through shipping? Predicting its potential invasive range using ecological niche
modeling. Hydrobiologia 649, 183-194, doi:10.1007/s10750-010-0244-5 (2010).
Walls, B. J. & Stigall, A. L. Analyzing niche stability and biogeography of Late Ordovician brachiopod species using ecological niche modeling. Palaeogeogr., Palaeoclimatol., Palaeoecol. 299, 15-29 (2011).
Huang, Z., Brooke, B. & Li, J. Performance of predictive models in marine benthic environments based on predictions of sponge distribution on the Australian continental shelf. Ecological informatics 112 (2011).
Pearson, R. G. Species’ Distribution Modeling for Conservation Educators and Practitioners. Synthesis. American Museum of Natural History. Available at http://ncep.amnh.org. (2007).
Fauchald, K. & Jumars, P. A. The diet of worms: a study of polychaete feeding guilds. Oceanogr. Mar. Biol. Annu. Rev. 17 (1979).
Foster, N. M. Spionidae (Polychaeta) of the Gulf of Mexico and Caribbean Sea. . Studies on the Fauna of Curac¸ao and other Caribbean Islands 36, 1-183 (1971).
Clarke, K. R. & Warwick, R. M. In:Change in marine communities. An approach to statistical analysis and interpretation seconded.PRIMER-E, Plymouth, UK. (2001).
SPSS. SPSS for Windows, Rel. 20.0.0. 2011. Chicago: SPSS Inc. (2011).
Tyberghein, L. et al. Bio-ORACLE: a global environmental dataset for marine species distribution modelling. Global Ecol. Biogeogr. 21, 272-281, doi:10.1111/j.1466-8238.2011.00656.x (2012).
Phillips, S. J., Anderson, R. P. & Schapire, R. E. Maximum entropy modelling of species geographic distributions. Ecol. Model. 190, 231-259 (2006).
)
We gratefully acknowledge the support of the Rowe lab at Texas A&M University at Galveston. We would also
like to thank all members of the Schulze lab for their overall help and support.