Rates of litter decomposition over 6 years in Canadian forests: influence of litter quality and climate

The effects of litter quality and climate on decomposition rates of plant tissues were examined using percent mass remaining (MR) data of 10 foliar litter types and 1 wood type during 6 years exposure at 18 upland forest sites across Canada. Litter-quality variables used included initial nutrient co...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Canadian journal of forest research 2002-05, Vol.32 (5), p.789-804
Hauptverfasser: Trofymow, J.A, Moore, T.R, Titus, B, Prescott, C, Morrison, I, Siltanen, M, Smith, S, Fyles, J, Wein, R, Camire, C
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The effects of litter quality and climate on decomposition rates of plant tissues were examined using percent mass remaining (MR) data of 10 foliar litter types and 1 wood type during 6 years exposure at 18 upland forest sites across Canada. Litter-quality variables used included initial nutrient contents (N, P, S, K, Ca, Mg) and carbon fractions (determined by proximate analysis and (13)C nuclear magnetic resonance spectroscopy). Climate variables used included mean annual temperature; total, summer, and winter precipitation; and potential evapotranspiration. A single-exponential decay model with intercept was fit using the natural logarithm of 0- to 6-year percent MR data (LNMR) for all 198 type by site combinations. Model fit was good for most sites and types (r(2) = 0.64-0.98), although poorest for cold sites with low-quality materials. Multiple regression of model slope (Kf) and intercept (A) terms demonstrated the importance of temperature, summer precipitation, and the acid-unhydrolyzable residue to N ratio (AUR/N) (r(2) = 0.65) for Kf, and winter precipitation and several litter-quality variables including AUR/N for A (r(2) = 0.60). Comparison of observed versus predicted LNMR for the best overall combined models were good (r(2) = 0.75-0.80), although showed some bias, likely because of other site- and type-specific factors as predictions using 198 equations accounted for more variance (r(2) = 0.95) and showed no bias.
ISSN:0045-5067
1208-6037
DOI:10.1139/x01-117