Data di Pubblicazione:
2016
Abstract:
Weaimto investigate the occurrence of forest humus forms (Moder, Amphi andMull) in relation to environmental
factors describing parent material, climate and tree species. Boosted regression trees (BRTs) were applied as
modeling tool to analyze data of 238 plots of the BioSoil database covering the whole Italian forest territory.
Though predictive ability was not very high, especially for the Amphi form, we could gain significant insight
into factors controlling humus form differentiation. In the BRT analysis, the diversity of tree species was the
most important predictor for Moder and Mull models and specific plant effects were evidenced. However, our
results showed that the geographic distribution of Italian forest species was influenced by soil and climate
conditions, partly explaining the high weight of tree species as factor. The importance of the soil nutritional
status, due to parent material properties, in driving humus form differentiation was stated, highlighting the
key role played by pH and calcium content, with the hitherto understated importance of phosphorus. This
study further clarified the functioning of the still poorly understood Amphi form. Reduced effective soil volume
(EfVol) combined with seasonality appeared to constrain pedofauna activity in otherwise favorable and nutrient
rich systems, favoring the evolution of Amphi instead of Mull form
Tipologia CRIS:
03A-Articolo su Rivista
Keywords:
Boosted regression trees; Forest soils; Humus forms; Soil Science
Elenco autori:
Andreetta, Anna; Cecchini, Guia; Bonifacio, Eleonora; Comolli, Roberto; Vingiani, Simona; Carnicelli, Stefano
Link alla scheda completa:
Link al Full Text:
Pubblicato in: