Landscape Ecology

, Volume 20, Issue 8, pp 971–987

Regeneration Strategies, Disturbance and Plant Interactions as Organizers of Vegetation Spatial Patterns in a Pine Forest

  • Pu Mou
  • Robert  H. Jones
  • Dali Guo
  • Andrew Lister
Research Article

DOI: 10.1007/s10980-005-7007-0

Cite this article as:
Mou, P., Jones, R.., Guo, D. et al. Landscape Ecol (2005) 20: 971. doi:10.1007/s10980-005-7007-0


To determine how vegetation pattern in early successional forests may be related to plant traits and types of disturbance, we measured percent cover of individual taxa annually in a South Carolina Pinus elliottii forest, starting one year before, and ending four years after harvest and tree girdling disturbances were applied. The 17 most important taxa surveyed were grouped into four regeneration strategies chosen a priori, and the spatial patterns of these groups and of the soil were investigated using global variability, semivariograms and kriged maps. We also examined spatial correlations across years, across taxa, and between species and soil disturbance. Seed bank taxa represented by Dichanthelium spp. increased rapidly and formed large patches, and then quickly declined. Taxa that regenerate by newly dispersed seeds, represented by Rhus copallina and Rubus spp. occurred at first in a few patches, and became widespread later. Stump sprouters, represented by Quercus spp. and Myrica cerifera, had rapid increases in cover, but their spatial patterns were largely determined by their pre-disturbance patterns. Prunus serotina, which relies on both sprouting and dispersed seed, had moderate cover and a random distribution. Within-species temporal correlation of spatial pattern was lower in girdled than in harvested plots, and was not clearly related to regeneration strategy. Forest floor disturbance was patchy and affected the pattern of Dichanthelium spp. in the harvested plots. Negative correlations between herbs and woody plants in harvested plots reflected the role of biotic (i.e., successional) filters on vegetation pattern. Surprisingly, no spatial correlations were detected between the nitrogen fixer, Myrica cerifera and other taxa in this N-limited system. In comparing the spatial and temporal patterns, we found kriged maps more informative than analysis of semivariograms alone. The maps and correlation statistics demonstrated that regeneration traits, spatial patterns of soil disturbances, and interactions among taxa influence dynamics of the spatial patterns of the plants. We also demonstrated that disturbance types affected the importance and interactions among these three factors, and caused different spatial patterns of the plant taxa.


Dichanthelium Forest disturbance Geostatistics Myrica cerifera Pinus elliottii Plant regeneration Quercus Rhus copallina Spatial heterogeneity 

Copyright information

© Springer 2005

Authors and Affiliations

  • Pu Mou
    • 1
  • Robert  H. Jones
    • 2
  • Dali Guo
    • 3
  • Andrew Lister
    • 4
  1. 1.Department of BiologyUniversity of North Carolina at GreensboroGreensboroUSA
  2. 2.Department of Biological SciencesVirginia TechBlacksburgUSA
  3. 3.Department of EcologyPeking UniversityChina
  4. 4.USDA Forest Service Northeastern Experimental StationAndrew ListerNewtown SquareUSA

Personalised recommendations