Data collection for plant communities

Desirable qualities of community samples:

1. Appropriate:
   • to the character of the community

     
     

   • to investigator's research purposes

     
     

   • to plans for subsequent data analysis

     
     

   • in intensity and breadth to support research conclusions

     
     

2. Homogeneous in structure and composition
   • scale is problematic, because plants are rarely distributed randomly--they are patterned on several scales

     
     

   • units of community ecology (i.e., communities) are not naturally delimited--subjective decisions must be made about what area to include or exclude

     
     

   • sites vary in degree of homogeneity and in type and severity of disturbances

     
     

   It has been suggested that the problem of obtaining reasonable sample homogeneity can be resolved along two lines:

   1. subjective procedures are generally adequate (judgments of experienced ecologists tend to coincide)

   2. if classifying communities into "types" is the goal, then the critical matter is homogeneity within sample sites relative to overall variation in the data set; if considerable variability exists in the data set relative to "within-community", then clear patterns will emerge

3. Objective and standardized

   there are many different sampling procedures, so selection of one is subjective

   once selected, sampling procedure should be applicable in an objective, standardized way; it should also be unambiguous and operational

   allows comparisons across treatments, years (data sets?)--long-term data collection often needed

4. Efficient

   maximum amount of information per unit of time and effort

   
   

Standard community sampling procedures

Selection of sampling procedures should consider at least the following:

1. kinds of communities sampled

2. kinds of environmental and historical data needed to complement (corroborate?) vegetation data

3. scope, accuracy, and purposes of the study

4. requirements to allow comparison w/ other studies

5. requirements for valid application of anticipated data analysis methods

6. practical limitations