Return to search

Fuel load characterisation and quantification for the development of fuel models for Pinus patula in South Africa

Thesis (MScBosb)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: The characteristics and total fuel load of the forest floor (FF) and harvest residue (HR) are
needed to develop tools that can be used for fuel load management, fire risk analysis and
fire behaviour prediction for P. patula grown in the summer rainfall area of South Africa (SA).
Forest floor depth, mass and ash-free mass were measured and there was generally a
greater range in depth under sawtimber (ST) stands than under pulpwood (PLP) stands.
Forest floor loads, prior to ashing, ranged from 21 - 168 t ha-1 and 27 - 72 t ha-1, for ST and
PLP stands, respectively and loads increased linearly with stand age. Sawtimber and PLP
stands were analysed together, which resulted in a significant correlation between depth and
mass (r2 = 0.78, n = 31). A loss on ignition procedure carried out on sub-samples of the FF
improved the relationship between the FF depth and the ash-free mass for the different
stands, and provided a more accurate model for the prediction of mass from depth. A
multiple regression analysis revealed that age, altitude and mean annual precipitation (MAP)
provided the best subset and accounted for 72% of the variation in the FF mass observed.
The effect of increasing FF load and increasing moisture content on the fireline intensity
(FLI) was examined using the fire behaviour prediction programme, BehavePlus.
Harvest residue was quantified and characterised in terms of fuel size classes, under current
silvicultural regimes for ST and PLP, for the development of fuel models for this species over
its planted range in SA. An investigation into the proportion (by mass) of the 1- (0.0 - 0.6
cm), 10- (0.6 – 2.5 cm), 100- (2.5 – 7.6 cm) and 1000-hr (> 7.6 cm diameter) fuel classes of
the total HR mass indicated that there was a significant difference between the mass of the
1-, 10- and 1000-hr fuel classes of the two silvicultural regimes, and no significant difference
for the 100-hr fuel class. Two fuel models for P. patula HR and two models for standing
timber were developed using the new model (NEWMDL) programme of BEHAVE and tested
in BehavePlus.
Nutrient concentrations were used with FF layer and HR size class load data to estimate the
quantities of nutrients held in the fuel and to describe nutrient distributions in the fuel
complex. Significant differences in the nutrient concentration of the FF layers and fuel
components were observed which has important implications for fuel management. The
concentration of N determined in this study, relative to that determined in other similar
studies on P. patula was low. Forest floor loads were predicted and nutrient pools calculated
for typical ST and PLP stands at both low and high altitude to provide insight into the nutrient
distributions within the fuel complex. / AFRIKAANSE OPSOMMING: Die karaktereienskappe en totale brandstoflading van die bosvloer (FF) en kaalkap oorskot
(HR) word benodig om instrumente te ontwikkel wat gebruik kan word vir brandstoflading
bestuur, brandgevaar ontleding en brandgedrag voorspelling vir P. patula, wat in die somer
reënvalgebied van Suid-Afrika groei. Die bosvloer diepte, massa en asvrye massa is gemeet
en daar was oor die algemeen ‘n groter variasie in diepte onder saaghout (ST) opstande as
onder pulphout (PLP) opstande. Die bosvloerladings, voor verassing, het varieer van 21 –
168 t ha-1 en 27 – 72 t ha-1 vir ST en PLP opstande respektiwelik. Ladings het linieêr
vermeerder met opstand ouderdom. Saaghout en PLP opstande is saam geanaliseer en het
tot ‘n betekenisvolle korrelasie gelei tussen diepte en massa (r2 = 0.78, n = 31). ‘n Verliestydens-
ontbranding prosedure is uitgevoer op die FF monsters en het die verhouding tussen
FF diepte en die asvrye massa van die verskillende opstande verbeter. Dit het ook gelei tot
akkurater model vir die voorspelling van massa vanaf diepte. ‘n Veelvoudige regressie
analise het aan die lig gebring dat ouderdom, hoogte en gemiddelde jaarlikse reënval (MAP)
die beste sub-groep verskaf, en het 72% van die variasie in die FF massa verklaar.
Ondersoek is ingestel op die effek van toenemende FF lading en toenemende voginhoud op
die brandlyn intensiteit (FLI) deur die brandgedrag program, BehavePlus, toe te pas.
Die kaalkap oorskot is gekwantifiseer en gekarakteriseer volgens brandstof grootteklasse,
onder die huidige boskultuurstelsels vir ST en PLP, vir die ontwikkeling van
brandstofmodelle vir hierdie spesie oor die betrokke groeistreek in SA. ‘n Ondersoek in die
verhouding (volgens massa) van die 1- (0.0 – 0.6 cm), 10- (0.6 – 2.5 cm), 100- (2.5 – 7.6
cm) en 1000-uur (> 7.6 cm deursnee) brandstofklasse van die totale HR massa het
aangedui dat daar ‘n betekenisvolle verskil is tussen die massas van die 1-, 10- en 1000-uur
brandstofklasse van die twee boskultuurstelsels, en geen betekenisvolle verskil vir die 100-
uur brandstofklas nie. Twee brandstofmodelle is ontwikkel vir P. patula HR en twee modelle
vir staande hout deur gebruik te maak van die nuwe model (NEWMDL) program van
BEHAVE en getoets in BehavePlus.
Voedingstof konsentrasies is gebruik, tesame met die FF laag en HR klasgrootte ladingdata,
om die voedingstof inhoud van die brandstof te skat en om die voedingstof verspreiding te
beskryf in die brandstofkompleks. Betekenisvolle verskille is waargeneem in die voedingstof
konsentrasies van die FF lae en brandstof komponente wat belangrike implikasies inhou vir
brandstofbestuur. Die konsentrasie wat vir N in hierdie studie bepaal is, was laag relatief tot
ander soortgelyke studies vir P. patula. Die bosvloer ladings is voorspel en voedingstofpoele
bereken vir tipiese ST en PLP opstande vir beide lae en hoë hoogtes om insig te verkry,
sodat insig verkry kon word in die voedingstof verspreidings binne die brandstofkompleks.

Identiferoai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/20904
Date12 1900
CreatorsRoss, Timothy Ian
ContributorsDu Toit, Ben, Theron, J.M., Stellenbosch University. Faculty of Agrisciences. Dept. of Forest and Wood Science.
PublisherStellenbosch : Stellenbosch University
Source SetsSouth African National ETD Portal
Languageen_ZA
Detected LanguageUnknown
TypeThesis
Formatxviii, 138 leaves : ill.
RightsStellenbosch University

Page generated in 0.0103 seconds