Ecosystem respiration in a Mediterranean Turkey oak forest: eddy covariance and chamber-based models comparison
Author(s)
Consalvo, Claudia
Date Issued
April 14, 2014
Type
Doctoral Thesis
Abstract
Eddy covariance and up-scaled chamber measurements were used to estimate ecosystem respiration (Reco) in a Quercus cerris L. coppice forest in Central Italy. The annual variation of Reco and its components, was analyzed in relation to seasonal changes in soil temperature and soil moisture. Chamber measurements on soil, stems and leaves were carried out between June 2010 and December 2012. Each plot was chosen to be always within the footprint of the eddy covariance tower installed in the forest compartment.
The ecosystem respiration and its main components shown a strong seasonal variability. Summer aridity played a crucial role on the observed reduction of respiration that was evident both in soil and stem CO2 efflux rates. Soil respiration (Rs) and stem respiration (Rstem) increased from winter to spring, decreased in summer and suddenly increased in autumn after rainfall events. A different trend was showed by leaf dark respiration (Rl) that increased from spring to summer and decreased in autumn.
Based on up-scaled chamber measurements, ecosystem respiration (Recocuv) showed a significant temporal variation over the year. Rs and Rstem followed the same trend of ecosystem respiration estimated by eddy covariance (Recoeddy) with a reduction during summer months consequently to summer drought. On the contrary, Rl peaked in August and decreased in autumn and was responsible for the decoupling of the two Reco estimates. The total annual Recoeddy chosen as reference was 9.11 t C ha-1 y-1, while the total annual contribution of Rs, Rstem and Rl was 6.79, 1.77 and 4.74 t C ha-1 y-1, respectively. Thus, the relative contribution of each component was 51% for the soil, 13% for the stem and 36% for leaves, confirming that whole ecosystem respiration is dominated by soil respiration. Recocuv was 32% larger than ecosystem respiration estimated by eddy covariance.
Through model validation techniques was possible to quantify the uncertainty. The results showed that method of estimation used, sample size, and biotic and abiotic factors influencing the respiration rates, can lead to both underestimation and overestimation higher than 50%.
Additional information
Dottorato di ricerca in Ecologia forestale
File(s)![Thumbnail Image]()
Name
cconsalvo_tesid.pdf
Size
2.35 MB
Format
Adobe PDF
Checksum (MD5)
555c0df5134b91861abbf241fbc68f53