Bray-Curtis
- Created by
- Renato Passos, Eng. de Software
- Reviewed by
- Renato Passos, Eng. de Software
Last updated: Apr 18, 2026
About this calculator
The Bray-Curtis index is a similarity measure commonly used in ecology to compare species composition between two ecosystems. It calculates the ratio between the sum of minimum abundances for each species and the total sum of all abundances. This helps assess how similar two environments are, considering both species presence and quantity.
The formula Σmin(xi,yi)/Σ(xi+yi) compares pairs of data (xi and yi) representing species abundance in both environments. The numerator sums the lower values of each pair, while the denominator sums all abundances. Values close to 1 indicate high similarity, while lower values suggest greater differences.
This calculator is useful for biodiversity studies, ecological change monitoring, or comparing degraded areas with preserved ones. Input data must be absolute counts (integers) rather than percentages. Interpretation should note that the index is sensitive to species with high abundances.
Caution: Do not use fractions or normalized data. The Bray-Curtis index ignores species absent in both samples. It is more suitable for count data than presence/absence data. Accuracy depends on the quality and consistency of field or lab-collected data.
Frequently asked questions
What is the Bray-Curtis index used for?
It compares similarity between two ecosystems based on species abundance, helping identify differences in biological communities.
How should I format the input data?
Provide absolute counts (integers) of each species across both environments, listed in parallel arrays.
What does a result close to 1 mean?
Indicates high similarity between the two datasets, meaning species abundances are very similar in both samples.
How does it differ from the Jaccard index?
Bray-Curtis considers species abundance, while Jaccard only accounts for presence/absence, making it less sensitive to quantity differences.
Can I use it for lab data?
Yes, as long as the data represents real species counts or components in two distinct sample sets.