Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Refining fuel loads in LPJ-GUESS-SPITFIRE for wet-dry areas : with an emphasis on Kruger National Park in South Africa

Persson, Ylva LU (2013) In Student thesis series INES NGEM01 20122
Dept of Physical Geography and Ecosystem Science
Abstract (Swedish)
Eld är en av de viktigaste störningsprocesserna som påverkar den terrestra biosfären genom att forma vegetationens spridning, komposition, struktur, växtdiversitet och biokemiska cykler. Några av de mest påverkade ekosystemen är våt-torra områden, här klassificerade som savann eller Medelhavsområden. Denna påverkan kan studeras på olika sätt, antingen genom ett långvarigt och fortlöpande experiment, som i Kruger National Park i Sydafrika, eller genom att använda sig av modeller, som till exempel LPJ-GUESS-SPITFIRE. SPITFIRE är en processbaserad eldmodell som har kopplats samman med en dynamisk vegetationsmodell för att förutspå eldens spridning, intensitet och hur länge det kan brinna.
I LPJ-GUESS-SPITFIRE används en faktor som kallas... (More)
Eld är en av de viktigaste störningsprocesserna som påverkar den terrestra biosfären genom att forma vegetationens spridning, komposition, struktur, växtdiversitet och biokemiska cykler. Några av de mest påverkade ekosystemen är våt-torra områden, här klassificerade som savann eller Medelhavsområden. Denna påverkan kan studeras på olika sätt, antingen genom ett långvarigt och fortlöpande experiment, som i Kruger National Park i Sydafrika, eller genom att använda sig av modeller, som till exempel LPJ-GUESS-SPITFIRE. SPITFIRE är en processbaserad eldmodell som har kopplats samman med en dynamisk vegetationsmodell för att förutspå eldens spridning, intensitet och hur länge det kan brinna.
I LPJ-GUESS-SPITFIRE används en faktor som kallas omsättningshastigheten, vilket är ett halvtidsvärde för nedbrytning av biologiskt avfall vid en viss temperatur och fuktighet. Den används för att räkna ut hur mycket tillgängligt växtmaterial som finns i det modellerade området för att kunna bestämma eldintensiteten. Denna konstant är satt till 2.85 och är densamma i alla världens ekosystem. I verkligheten beror nedbrytningen på vilka de aktiva förhållandena är och det är därför orealistiskt att använda samma konstant. Data från experimentet i Kruger National Park användes för att förbättra parametrarna i DGVM LPJ_GUESS_SPITFIRE för att hitta mer representativa nedbrytningsvärden i våt-torra områden. Avfallet delades in i två pooler som är det huvudsakliga bränslet för markbränder; avfall i form av löv och trä och deras omsättningshastigheter studerades individuellt. En litteraturstudie visade att de nuvarande värdet för löv är för högt och för lågt satt för trä. Nya värden testades, avfallsmängder för olika eldintervall räknades ut och jämfördes med mätta mängder i Kruger National Park. En omsättningshastighet på 0.6 år visade sig vara bästa representanten för våt-torra områden för lövavfall utan träd. Ett värde för träavfall kunde inte hittas, då det inte fanns tillgång till data att jämföra värdena med. Däremot kunde effekten av omsättningshastigheten studeras för att få en bättre förståelse för vilken effekt det hade på dessa typer av ekosystem. Denna uppsats visar vikten av att förstå hur olika ekosystem fungerar och att det fortfarande kan göras förbättringar i den använda modellen. (Less)
Abstract
Fire is one of the most important disturbance processes affecting the terrestrial biosphere, altering the vegetation composition and distribution, structure, plant diversity and biogeochemical cycles. Some of the most influenced ecosystems are wet-dry areas, here classified as savannah or Mediterranean regions. The influence can be studied in different ways, either by long-term experiments like the burn plot trial in Kruger National Park in South Africa, or by the use of models, for example LPJ-GUESS-SPITFIRE. SPITFIRE is a process-based fire model which have been coupled to a dynamic global vegetation model in order to predict fire spread, intensity and residence time of fires.
In LPJ-GUESS-SPITFIRE a variable called the turnover time,... (More)
Fire is one of the most important disturbance processes affecting the terrestrial biosphere, altering the vegetation composition and distribution, structure, plant diversity and biogeochemical cycles. Some of the most influenced ecosystems are wet-dry areas, here classified as savannah or Mediterranean regions. The influence can be studied in different ways, either by long-term experiments like the burn plot trial in Kruger National Park in South Africa, or by the use of models, for example LPJ-GUESS-SPITFIRE. SPITFIRE is a process-based fire model which have been coupled to a dynamic global vegetation model in order to predict fire spread, intensity and residence time of fires.
In LPJ-GUESS-SPITFIRE a variable called the turnover time, which is the logarithmic decomposition rate for litter at a defined temperature and moisture content is used to calculate how much available litter (or in other words fuel) is available in the modelled patch in order to determine the fire intensity. This constant is set to 2.85 and is used for all ecosystems around the world. In reality however, the turnover time varies depending on the existing circumstances, which makes it unrealistic to use the same constant. Data obtained from the burn plot trial in Kruger National Park was used for parameter refinements within the DGVM LPJ-GUESS-SPITFIRE in order to find more representative values. The litter pool was divided into two pools which are the main input for surface fires; leaf litter and wood litter and their turnover times were studied individually. A literature study showed that presently used values for the leaf turnover time are overestimated by the model and underestimated for wood. In order to find more suitable parameters, new values were tested, litter amounts for different fire return intervals were calculated and compared with measured litter amounts in the experiment in Kruger National Park. A turnover time of 0.6 years for leaf litter without trees was found to be the best representative for wet-dry areas. The turnover time for woody litter could not be adjusted, since there was no available data to compare them with. However, the effect of turnover times within the range found in the literature study was examined in order to better understand the effect wood litter had on these kinds of ecosystems. This thesis shows the importance to understand how different ecosystems work and that improvements still can be made in the used model. (Less)
Please use this url to cite or link to this publication:
author
Persson, Ylva LU
supervisor
organization
course
NGEM01 20122
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Kruger National Park, turnover time, fuel, Fire, wet-dry areas, LPJ-GUESS-SPITFIRE, geography, physical geography
publication/series
Student thesis series INES
report number
268
language
English
id
3559027
date added to LUP
2013-02-28 10:25:11
date last changed
2013-02-28 10:25:11
@misc{3559027,
  abstract     = {{Fire is one of the most important disturbance processes affecting the terrestrial biosphere, altering the vegetation composition and distribution, structure, plant diversity and biogeochemical cycles. Some of the most influenced ecosystems are wet-dry areas, here classified as savannah or Mediterranean regions. The influence can be studied in different ways, either by long-term experiments like the burn plot trial in Kruger National Park in South Africa, or by the use of models, for example LPJ-GUESS-SPITFIRE. SPITFIRE is a process-based fire model which have been coupled to a dynamic global vegetation model in order to predict fire spread, intensity and residence time of fires.
In LPJ-GUESS-SPITFIRE a variable called the turnover time, which is the logarithmic decomposition rate for litter at a defined temperature and moisture content is used to calculate how much available litter (or in other words fuel) is available in the modelled patch in order to determine the fire intensity. This constant is set to 2.85 and is used for all ecosystems around the world. In reality however, the turnover time varies depending on the existing circumstances, which makes it unrealistic to use the same constant. Data obtained from the burn plot trial in Kruger National Park was used for parameter refinements within the DGVM LPJ-GUESS-SPITFIRE in order to find more representative values. The litter pool was divided into two pools which are the main input for surface fires; leaf litter and wood litter and their turnover times were studied individually. A literature study showed that presently used values for the leaf turnover time are overestimated by the model and underestimated for wood. In order to find more suitable parameters, new values were tested, litter amounts for different fire return intervals were calculated and compared with measured litter amounts in the experiment in Kruger National Park. A turnover time of 0.6 years for leaf litter without trees was found to be the best representative for wet-dry areas. The turnover time for woody litter could not be adjusted, since there was no available data to compare them with. However, the effect of turnover times within the range found in the literature study was examined in order to better understand the effect wood litter had on these kinds of ecosystems. This thesis shows the importance to understand how different ecosystems work and that improvements still can be made in the used model.}},
  author       = {{Persson, Ylva}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{Student thesis series INES}},
  title        = {{Refining fuel loads in LPJ-GUESS-SPITFIRE for wet-dry areas : with an emphasis on Kruger National Park in South Africa}},
  year         = {{2013}},
}