2011
DOI: 10.1111/j.1365-2656.2011.01806.x
|View full text |Cite
|
Sign up to set email alerts
|

Comparing isotopic niche widths among and within communities: SIBER - Stable Isotope Bayesian Ellipses in R

Abstract: Summary1. The use of stable isotope data to infer characteristics of community structure and niche width of community members has become increasingly common. Although these developments have provided ecologists with new perspectives, their full impact has been hampered by an inability to statistically compare individual communities using descriptive metrics. 2. We solve these issues by reformulating the metrics in a Bayesian framework. This reformulation takes account of uncertainty in the sampled data and nat… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

15
3,076
4
22

Year Published

2013
2013
2022
2022

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 2,595 publications
(3,117 citation statements)
references
References 39 publications
15
3,076
4
22
Order By: Relevance
“…Nitrogen and carbon isotope analyses were performed simultaneously using continuous-flow isotope ratio mass spectrometry under the same conditions and degree of accuracy as reference [30]. [36]. Statistical significance of differences in SEA C between sample sets was based on the proportional outcome of 10 6 repeat measures.…”
Section: Methodsmentioning
confidence: 99%
“…Nitrogen and carbon isotope analyses were performed simultaneously using continuous-flow isotope ratio mass spectrometry under the same conditions and degree of accuracy as reference [30]. [36]. Statistical significance of differences in SEA C between sample sets was based on the proportional outcome of 10 6 repeat measures.…”
Section: Methodsmentioning
confidence: 99%
“…First, outlier trials should be removed, for example using the two-stage method for time-series data that leads to the improvement in the number of time-points being normally distributed [9]. Second, the trial size should be at least 10, as through simulations this was found to provide a 'sufficiently' low bias in the ellipse area versus convex area [16]. Related to trial size, as recommended for standardisation [12], instead of using the bivariate F-distribution that varies with sample size [17] the ellipse was scaled using the Chi-square distribution that leads to a fixed scaling factor for all sample [2].…”
Section: Discussionmentioning
confidence: 99%
“…% finds overlap of all quadrilaterals from series1 with all quadrilaterals from series2 beta1=zeros(m,1); %Series 1; default zeros is no overlap beta2=zeros(m,1); %Series 2; default zeros is no overlap (beta1 and beta2 will be same if no timelag) Q=NaN (2,16); R=NaN (2,16);S=NaN (2,16);T=NaN (2,16);%16 to compare 4 vectors of quad1 to 2 for t=1:m % STAGE4a: create combinations of the 16 vectors; series1 vectors Q(1,:)=[repmat(Px1(1,t),1,4),repmat(Px1(2,t),1,4),repmat(Px1(3,t),1,4),repmat(Px1(4,t),1 ,4)]; Q(2,:)=[repmat(Py1(1,t),1,4),repmat(Py1(2,t),1,4),repmat(Py1(3,t),1,4),repmat(Py1(4,t),1 ,4)]; R(1,:)=[repmat(Px1(2,t),1,4),repmat(Px1(3,t),1,4),repmat(Px1(4,t),1,4),repmat(Px1(1,t),1 ,4)]; R(2,:)=[repmat(Py1(2,t),1,4),repmat(Py1(3,t),1,4),repmat(Py1(4,t),1,4),repmat(Py1 (1,t) if W<=0 %skips this section if already know there is an overlap % find which quadrilaterals has lowest y value, and subtract min y from all y's % order so only need check 2 is inside 1 (and no need to check reverse that 1 inside 2) Z=[Px1 ; else W=1; %i.e. no cross, so quad2ofZ within quad1ofZ, so quads overlap end end end beta1(t)=beta1(t)+W; % STAGE4c beta2(h)=beta2(h)+W; end end gamma=isfinite(1./(beta1+beta2)); % STAGE4d…”
mentioning
confidence: 99%
“…In addition, Stable Isotope Bayesian Ellipses in R (SIBER, Jackson et al 2011) was used to characterize intra-group variation using a Bayesian framework. Given small sample sizes, however, this attempt was restricted to visual depictions of standard ellipse area (the bivariate equivalent of standard deviation) rather than calculation of the various within-group metrics being used more frequently in ecological (e.g.. Layman et al 2007a, b;Syväranta et al 2013) and, more recently, archeological isotopic studies (Szpak et al 2014).…”
Section: Methodsmentioning
confidence: 99%