Community and ecosystem analysis of forests recovering from landslide disturbance: White Mountain National Forest, New Hampshire

<p>Sites recovering from landslide disturbance offer many opportunities to study ecosystem processes under extreme conditions. Landslides reset the topographic, microclimatic, parent material and vegetation state factors of ecosystem development. Water, and nutrient resources following landslides are more heterogenously distributed. Soil structure and development is disrupted. Solar insolation and diurnal temperature fluctuation exceed that of the surrounding forest. Consequently, plant colonization and community succession differ from less severely disturbed sites such as timber harvest or agricultural afforestation.</p><p>Two chapters of this dissertation address the unique conditions of landslide ecosystem development. Chapter 2 compares historical data collected in 1956 on a chronosequence of landslides in the White Mountain National Forest to data collected 40 years later. Detrended correspondence analysis is used to determine the major driving variables of succession during the two sample periods. The 1956 dataset suggests that site age was the primary driver of succession on this chronosequence. The 1996--98 data showed that elevation affected species composition more than site age following 40 years of succession. Therefore, caution should be used when extrapolating results beyond the range of chronosequence data.</p><p>Chapter 3 addresses the patterns of nutrient status and cycling on 4 landslide sites of varying age. Foliar and litterfall nutrient content, and the ratios of the macro nutrients, P, K, Ca and Mg with N are compared to observe potential nutrient imbalances and limitation. Litter decay and nutrient dynamics illustrate relative cycling rates of these elements across the chronosequence. Significant correlations of foliage and litterfall P concentration and N:P ratios with site age imply that P may limit ecosystem development on these sites.</p><p>Chapter 4 is a methods paper that compares the equivalence of forest stand data collected with the point-centered quarter and fixed-area plot methods. Data from the landslide sites and from a mature uneven aged forest provides a range of site conditions for comparison of the two methods. Stand density, basal area, species richness and community structure measurements using the point-centered quarter method did not provide equivalent results with fixed area plot estimates.</p>

Name thesis:degree_name

Doctor of Philosophy

Level thesis:degree_level

Dissertation

Year

2002

Author dc:creator

• Bryant, David M

Contributors dc:contributor

• John O Aber

• Biology
• Ecology
• Biostatistics
• Agriculture
• Forestry and Wildlife

Repository record dc:identifier

https://scholars.unh.edu/dissertation/60

OAI identifier oai:identifier

oai:scholars.unh.edu:dissertation-1059