Occupancy Frequency Distribution in Macroecology
In the fields of macroecology and community ecology, the occupancy frequency distribution (OFD) serves as a critical tool for analyzing how species are spread across a landscape. Simply put, an OFD describes the distribution of the number of species that occupy different numbers of sampled areas. First reported in 1918 by Danish botanist Christen C. Raunkiær during his research on plant communities, this metric is also frequently referred to in scientific literature as the species-range size distribution.
Key Facts
- Definition: The distribution of species based on the number of areas they occupy.
- Raunkiaer's Law: The observation that species in a community tend to be either very rare or very common (bimodal distribution).
- Scale Dependence: The shape of the OFD changes based on the size of the sampling interval (grain).
- Common Shapes: While bimodality is famous, empirical data shows right-skewed unimodal (approx. 46%), bimodal (approx. 27%), and uniform (approx. 27%) distributions.
- Primary Drivers: Factors include sampling grain, habitat heterogeneity, dispersal ability, and extinction-colonization dynamics.
The Concept of Bimodality and Raunkiaer's Law
A hallmark of many OFDs is bimodality, a pattern where the distribution has two peaks. This phenomenon is known as Raunkiaer's law of distribution of frequencies. According to this law, when species are assigned to five occupancy classes (each 20% wide), homogenous plant formations typically show peaks in the first class (0-20% occupancy) and the last class (81-100% occupancy). This suggests that species within a community are typically either rare or common.
However, the validity of Raunkiaer's law as a strict index of homogeneity has been questioned. In 1929, Henry Gleason noted that the law is essentially an expression of the fact that most associations contain more species with few individuals than species with many. He argued that the apparent bimodality is often a result of choosing a quadrat size that is most "serviceable" to show frequency, and that the pattern disappears if the quadrats are too large or too small.
Modern research supports a variety of OFD shapes. Tokeshi reported that roughly 46% of observations are right-skewed unimodal, 27% are bimodal, and 27% are uniform. More recent studies of 289 real communities reaffirm that bimodal OFDs occur in about 24% of cases.
ไม่มีภาพประกอบ
Factors Influencing OFD Shapes
The shape of an OFD is not static; it is heavily influenced by the sampling grain (the size of the sampling interval). Research by McGeoch and Gaston (2002) demonstrates that as sampling grains increase, the number of "satellite" (rare) species decreases while the number of "core" (common) species increases. This shift moves the OFD from a bimodal shape toward a right-skewed unimodal distribution.
This occurs because species range—measured as occupancy—is strongly affected by spatial scale and aggregation structure, a concept known as the scaling pattern of occupancy. This scale dependence also impacts other ecological patterns, such as the relationship between occupancy and abundance.
Beyond scale, several other factors can influence the OFD shape:
- Degree of habitat heterogeneity
- Species specificity
- Landscape productivity
- Position within the geographic range
- Species dispersal ability
- Extinction–colonization dynamics
Mechanisms Explaining Bimodality
Ecologists have proposed three primary models to explain why bimodal distributions occur in species occupancy.
1. Sampling Results
Bimodality can emerge from the random sampling of individuals from lognormal or log-series rank abundance distributions. In this scenario, the probability of choosing an individual from a species is proportional to its frequency. However, this model is considered less informative because the underlying mechanisms that generate lognormal species abundance distributions remain a subject of intense debate.
2. Core-Satellite Hypothesis
This model suggests that bimodality is generated by metapopulation dynamics—specifically the balance of colonization and extinction—associated with a strong rescue effect (where immigration prevents a local population from going extinct). While useful for communities influenced by dispersal and local extinction, this model is sensitive to specific immigration and extinction parameters and fails to explain the scale dependence of OFDs.
3. Occupancy Probability Transition (OPT) Model
The OPT model is based on the scaling pattern of occupancy under a self-similar assumption of species distributions. It utilizes a bisection scheme and the recursion probability of occupancy across different scales. The OPT model successfully supports two key observations: the prevalence of bimodality in interspecific distributions and the increase of satellite species at finer scales.
Further refinements by Hui and McGeoch (2007) suggest that self-similarity breaks down according to a power relationship with spatial scales. By adopting a power-scaling assumption, researchers can better model species distributions, particularly at fine scales. This approach also helped resolve the Harte-Maddux debate by demonstrating that the probability of a species occurring at one scale is not independent of its probability at the next, highlighting the importance of species co-occurrence patterns.
Summary of OFD Models
| Model | Primary Mechanism | Strengths | Weaknesses |
|---|---|---|---|
| Sampling Results | Random sampling from abundance distributions | Simple mathematical derivation | Not sensitive to biological mechanisms |
| Core-Satellite | Colonization-extinction metapopulation dynamics | Explains dispersal and local extinction | Highly sensitive to parameters; ignores scale |
| OPT Model | Scaling patterns and power-scaling assumptions | Explains scale dependence and fine-scale patterns | Mathematically complex |
Frequently Asked Questions
What is the difference between a core species and a satellite species?
In the context of occupancy frequency distributions, core species are those that are common and occupy a large proportion of the sampled areas, while satellite species are rare and occupy only a few areas.
Why does the sampling grain matter in OFD analysis?
The sampling grain (the size of the area being sampled) affects the perceived range of a species. Larger grains tend to increase the number of core species and decrease satellite species, potentially shifting a bimodal distribution toward a unimodal one.
What is Raunkiaer's law?
Raunkiaer's law is the observation that in many plant communities, the occupancy frequency distribution is bimodal, meaning most species are either very rare (0-20% occupancy) or very common (81-100% occupancy).
How does the OPT model improve upon the Core-Satellite hypothesis?
Unlike the Core-Satellite hypothesis, the Occupancy Probability Transition (OPT) model accounts for scale dependence, explaining how the number of rare species increases as the sampling scale becomes finer.
What is the "rescue effect" in metapopulation dynamics?
The rescue effect occurs when the immigration of individuals from a neighboring population prevents a local population from going extinct, thereby maintaining the species' occupancy in that area.