tidal hardwood swamps - Maryland Department of Natural Resources

vegetation community types in the USNVC using the Ecology Access Reporting Tool (Version 2.7;
NatureServe 2002) and regional classifications from various states.
Table 3. Compositional Summary Statistics (adapted from Fleming and Coulling 2001)
Compositional Statistic Definition
Frequency The number of samples in a group in which a species occurs
Mean Cover Back-transformed cover class value corresponding to mean percent cover
calculated from midpoint values of cover class ranges
Relative Cover The arithmetic difference between mean cover (for a given group of samples) and
total mean cover (for the entire dataset)(= Mean Cover – Total Mean Cover)
Constancy The proportion of samples in a group in which a species occurs (= frequency /
number of samples in a group x 100)
Fidelity The degree to which a species is restricted to a group, expressed as the proportion
of total frequency that frequency in a give group constitutes (= frequency / total
frequency x 100)
Indicator Value (IV) (= Constancy x Fidelity / 100)
Indicator Value Adjusted by Cover, Scale (Adj IV [scaled]) (= Indicator Value x Mean Cover / 9)
Indicator Value Adjusted by Cover, Unscaled (Adj IV [unscaled]) (= Indicator Value x 2 relative cover )
Mean Species Richness The average number of species present per plot (S); only species rooted inside
plot boundaries were included in this calculation
Homoteneity The mean constancy of the S most constant species, expressed as a fraction;
higher values for homoteneity indicate a greater uniformity in species composition
among plots.
Ordination techniques were used to identify the relationships of recognized vegetation types to one
another and the environmental gradients along which they are distributed (Gauch 1982; Jongman et al.
1995). These techniques were also used to validate the vegetation types determined with the
classification models. Ordination was performed using the Detrended Correspondence Analysis (Hill
1989a) and Non-metric Multidemsional Scaling (NMDS; Kruskal 1964) modules in PC-ORD (McCune and
Mefford 1995).
The objective algorithms of the analysis techniques within PC-ORD were the primary tool used to
determine the vegetation classification (McCune and Mefford 1995). But, these analysis techniques often
do not recognize compositional subtleties of similar communities. They often focus on presence or
absence of certain species, which can be due to seasonal and conditional biases rather than true
community shift. Therefore, a certain degree of subjective determination by highly trained project
ecologists, with the consultation of regional ecologists, was utilized to fine-tune the classification.
Detailed descriptions of each vegetation community type were prepared. They contain descriptions of
physiognomy and composition, the range of habitat conditions across which a type occurs, and spatial
distribution. They also include the features that distinguish a type from similar types, nomenclatural
synonymy, global and state conservation rank, lists of rare species, a discussion of characteristic species,
and conservation and management concerns. Also, a list of high quality reference sites was created.
These include detailed site descriptions and accurate digital maps created in Maptech® Terrain Navigator
Pro (Version 6.02) and ArcView 3.2a.
Data compilation and analysis occurred during the time period from December 2003 to March 2004.
10