Managing Douglas fir stands in Italy: calibration and validation of Forest Vegetation Simulator (FVS) and analysis of climate sensitivity for future calibration of Climate-FVS
Author(s)
Castaldi, Cristiano
Date Issued
June 16, 2016
Type
Doctoral Thesis
Abstract
In Italy, Douglas fir (Pseudotsuga menziesii (Mirb.) Franco) has a high potential in terms of wood
production and drought tolerance. Testing exotic tree species in Italy dates back to the early years of
the last century with the valuable work done by Aldo Pavari. Distinctively, Douglas fir has provided
the most satisfactory results in terms of growth and yield. However, a growth reference for mature
stands is lacking. We calibrated and validated the Pacific Northwest variant of FVS to Douglas fir
plantations in Italy and ran the calibrated model to test management alternatives. We calibrated the
height-diameter, crown width, crown ratio, and diameter increment submodels of FVS using
multipliers fitted against tree measurements (n=704) and increment cores (180) from 20 plots across
th Apennine range. Validation was carried out on tree-level variables sampled in 1996 and 2015 in
two independent permanent plots (275 trees). Multiplier calibration improved the error of crown
submodels by 7-19%; self-calibration of the diameter growth submodel produced scale factors of
1.0 – 5.2 for each site. Validation of 20-years simulations was more satisfactory for tree diameter (-
6% to +1% mean percent error) than for height (-10% – +8%). Calibration reduced the error of
predicted basal area and yield after 50 years with respect to yield tables. Simulated response to
thinning diverged depending on site index and competition intensity. FVS is a viable option to
model the yield of Douglas fir plantations in Italy, reflecting current understanding of forest
ecosystem dynamics and how they respond to management interventions. First large-scale
experimental plantations of Douglas fir were established between 1922-1938, with the surviving
stands now exceeding conventional rotation ages (50-60 years). These stands offer a great
opportunity to carry out research on sensitivity of tree growth with respect to climate by this nonnative
tree species for the purpose of adaptive forest management in the Mediterranean area. To
this end, we have carried out dendroclimatic analyses in two 80-90 years old Douglas fir stands: the
northern most (Tuscany) and the southern most (Calabria) ones among the oldest plantations. We
sampled twenty dominant trees per stand and built a standardized mean ring-width chronology for
each site. We tested bootstrap correlations between site chronologies and minimum temperature,
maximum temperature, precipitation and standardized precipitation index (SPI) from the database
ClimateEU. We used the global correlation function across the entire lifetime of the stands, and the
moving correlation function to analyze periodic growth trends. The two sites share a positive
correlation between tree growth and winter-spring temperatures, and a negative correlation with
summer minima and maxima. Precipitation and SPI of the previous autumn are negatively
correlated with tree growth at both sites. Spring-summer precipitation and water balance have a
positive effect on growth in the northernmost site only, although the southernmost site displayed a
summer dry period. Differences in correlation strength and significant months are likely due to the
different latitude of the two sites, continentality (distance from the sea), and adaptive physiologic
activity (e.g., stopping cambial activity during the summer dry period). A shift and increase in
summer temperature and precipitation sensitivity in the later period of analysis may be indicative of
the effect of climate warming. Douglas fir in Southern Europe has thus been proved to be sensitive
to winter frost and spring water balance, but can tolerate summer drought, and has potential to be
planted extensively as a supplementary timber resource under mountain Mediterranean climate.
Additional information
Dottorato di ricerca in Ecologia forestale
File(s)![Thumbnail Image]()
Name
ccastaldi_tesid.pdf
Size
1.67 MB
Format
Adobe PDF
Checksum (MD5)
35df517e56cfdfd12a47073d1bc81ab6