We report on the potential for developing long-term fire histories from chir pine (Pinus roxburghii Sarg.) forests in the Western Himalayan foothills based on a preliminary study from a stand located in the state of Uttarakhand in northern India. Rings from trees collected to develop a master skeleton plot chronology were generally complacent with false rings present during most years, but were crossdatable with only minor difficulty. The oldest tree confidently crossdated back to 1886, with good sample depth (5 trees) from 1911, which helped date the fire scars in cross-sections collected from three trees. Fire frequency as determined from fire-scar dates was high, with mean and median fire intervals of 3 years from 1938 to 2006. Fires were likely from human ignitions given the prevalence of human land use in the site. Fire scars were generally recorded at false-ring boundaries and likely represent burning during the hot, dry period in May or early June before the onset of monsoon rainfall beginning in mid-June. Although only three fire-scarred trees were sampled, this preliminary assessment shows there is a potential for additional samples from other stands to develop longer-term fire histories to better understand the role of fire in the ecology and management of chir pine throughout its range in the Himalaya region.

Forest fires occur frequently in the boreal forest of North America and greatly affect vegetation dynamics, biogeochemical cycles and resident human populations. Estimates of the frequency of boreal forest fires would be useful for understanding boreal ecosystems and managing the affects of fires on human populations. The objectives of this work were to investigate relations between fire frequency and climate change, vegetation)n type and waterbreaks in Wood Buffalo National Park (WBNP), located in northern Alberta. To address these objectives, four hypotheses were tested: 1) tree ring-width records from the boreal forest can provide a proxy climate record; 2) annual area burned in the boreal forest varies in response to climate changes; 3) boreal fores: fire frequency varies with differences in forest type and the proximity to waterbreaks and 4) fossil pollen and macroscopic charcoal records from massive lake sediments can provide meaningful estimates of local fire frequency. The first hypothesis was tested by constructing tree ring chronologies from 3 white spruce and two jack pine sites in WBNP. All five chronologies were significantly positively correlated with June precipitation in the growth year or the previous year, and were significantly negatively correlated with historical records of fire weather and annual area burned. The second hypothesis was tested by analyzing historical records of annual area burned and climate. and tree ring records of fire history and climate. Annual area burned was significantly negatively correlated with seasonal means of fire weather indices. The time since last fire was estimated using tree ring records from 166 sites located throughout WBNP. These records exhibited decadal and centennial scale variations in fire frequency. Comparisons with tree ring other proxy climate records suggest that the~e variations are related to climatic changes. The third hypothesis was tested using survival analysis of the time since last fire records, disaggregated by dominant vegetation and the mean distance to waterbreaks. Sites dominated by jack pine (Pinus banksiana) and aspen (Populus tremuloides) exhibited significantly higher fire frequencies than did sites dominated by black spruce (Picea mariana) or white spruce (Picea glauca). Fire frequency increased with increased mean distance to waterbreaks. The fourth hypothesis was tested by analyzing fossil pollen and charcoal records from two lakes at -5 year resolution for 600 years. I compared their fire history records with local tree ring records of fire, and their mean fire intervals with regional fire frequency estimates for sites with similar vegetation and mean distances to waterbreaks. One lake exhibited a meaningful fire frequency estimate and the other lake did not. The poor fire frequency estimate was related to high sediment mixing and the lack of homogenous vegetation around the lake. The results indicate that: 1) area burned and fire frequency in the boreal forest of northern Alberta varies temporally at the annual, decadal and centennial scales; 2) fire frequency varies spatially in relation to vegetation type and mean waterbreak distance; and 3) lakes with massive sediments can provide meaningful estimates of local fire frequency. / Thesis / Doctor of Philosophy (PhD)

Student Number : 9008538J - MSc Dissertation - School of Biology - Faculty of Science / O’Regan SP, 2005. The long-term effects of fire frequency and season on the woody vegetation in the Pretoriuskop sourveld of the Kruger National Park. MSc Dissertation, University of the Witwatersrand, Johannesburg. The role of fire in the management of conservation areas has historically been a contentious issue in which traditional agricultural principles and ever-changing conservation principles tend to collide. The Kruger National Park (KNP) in the early 1950s was no exception where the appropriate use of fire and its ecosystem consequences were hotly debated. The controversy surrounding the management of fire in the KNP highlighted the significant lack of understanding of fire and its role in the ecosystem and because of this controversy, the Experimental Burn Plot (EBP) experiment was established in 1954. The EBP experiment comprised 12 treatments, and a pseudo-randomised block design was used in which the 12 fire treatments were replicated four times each in four of the six major vegetation zones identified at the time. The EBP experiment originally comprised 192 experimental plots approximately 7 Ha in size each and covered approximately 12 km2 in the KNP. The twelve fire treatments were an annual burn in August, biennial and triennial burns in February, April, August, October, and December, and a control on which fire was excluded. Despite having been plagued with negative assessments from internal and external researchers from its inception, the EBP experiment was meticulously maintained, and it has now become a valuable research asset in the KNP. Four replicates of twelve plots each were located in the Pretoriuskop sourveld landscape of the KNP. These replicates were named Fayi, Kambeni, Numbi, and Shabeni after nearby landmarks. The Pretoriuskop region is a moist infertile mesic-savanna, which experiences on average 744mm of rain annually. The dominant tree species in Pretoriuskop are Dichrostachys cinerea and Terminalia sericea and the dominant grass species is Hyperthelia dissoluta. A baseline survey of the woody vegetation was done on all the Pretoriuskop plots in 1954 by HP Van Der Schijff. A second survey of the woody vegetation on all the Pretoriuskop plots was done in 1996 by SP O’Regan. This provided a 42-year period of treatment application over which the effects of fire frequency and season on the woody vegetation of the Pretoriuskop region were studied. The aim of this study was to investigate the long-term effects of the twelve fire treatments on the density, structure, and species composition of the woody vegetation in Pretoriuskop. The objectives of this study were: 1. To carry out a complete re-survey of the trees and shrubs on the Pretoriuskop EBPs using similar methods as those used in the baseline survey in 1954. 2. To capture into a digital format pertinent woody vegetation survey data from surveys that had been conducted on the Pretoriuskop EBPs between 1954 and 1996. 3. To compare the density, structure, and composition of the woody vegetation on the Pretoriuskop EBPs between 1954 and 1996, to determine the effects of fire on the woody vegetation of Pretoriuskop. 4. To investigate the history of the Kruger National Park Experimental Burn Plots experiment. The four replicates in the Pretoriuskop region were found generally to have very similar woody vegetation traits (density, species composition, and structural composition). However, the EBPs were established and surveyed in two distinct phases, the first phase comprised the control, August Annual, and the Biennial plots, and the second phase comprised the Triennial plots. The baseline structural composition of the plots established in the first phase was different from the structural composition of the plots in the second phase. Furthermore, the Pretoriuskop EBPs are located in two distinct vegetation types, namely the open and the closed Terminalia sericea \ Combretum woodlands of the Pretoriuskop region. The Numbi and Shabeni replicates are in the open Terminalia sericea \ Combretum woodlands, and the Kambeni and Fayi replicates are in the closed Terminalia sericea \ Combretum woodlands. It was found that the species composition of the plots was influenced by the location of the plots in the different vegetation types. The exclusion of fire in the Pretoriuskop sourveld results in an increase in the density of the overstorey and understorey woody vegetation, and an increase in the number of species, species diversity, and species evenness. This is because fire sensitive and fire intolerant woody species become more abundant as the period between fires increases. In Pretoriuskop, there is no evidence of relay floristic succession, because fire sensitive and fire intolerant woody species do not replace fire tolerant species. Instead, the floristic succession is accumulative and fire tolerant, fire sensitive, and fire intolerant woody species coexist as the period between fires increases. Woody species tolerant of frequent fires in Pretoriuskop are Albizia versicolor, Catunaregam spinosa, Lonchocarpus capassa, Pavetta schumanniana, Senna petersiana, Strychnos madagascariensis, and Turraea nilotica. Woody species that are sensitive or intolerant of fire in Pretoriuskop are Acacia swazica, Bauhinia galpinii, Combretum mossambicense, Commiphora neglecta, Croton gratissimus, Dalbergia melanoxylon, Diospyros lycioides, Diospyros whyteana, Euclea natalensis, Hyperacanthus amoenus, Kraussia floribunda, Ochna natalitia, Olea europaea, Psydrax locuples, Putterlickia pyracantha, Tarenna supra-axillaris, and Zanthoxylum capense. Dichrostachys cinerea and Terminalia sericea were found to dominate in areas that had been burnt frequently as well as areas where fire has been excluded. The change in the density of the woody vegetation as the inter-fire period increases is not linear but rather J shaped with an initial decrease in the density as the inter-fire period increases from 1 year to 3 years. This initial decrease in density is the result of a loss of very short (<1m tall) woody individuals. In contrast, there is no initial decrease in the number of tree equivalents (phytomass) of the woody vegetation as the inter-fire period increases. After the initial decrease in the density of the woody vegetation, the density increases as the inter-fire period increases beyond 3 years. Generally in Pretoriuskop, post fire age of the vegetation was found to be an important factor affecting the structure of the woody vegetation, and as the inter-fire period increases the number of structural groups, the structural diversity, and the structural evenness of the woody vegetation increases. As the inter-fire period increases the number of single-stem individuals relative to the number of multi-stem individuals increases, and the average height of the woody vegetation increases. The findings regarding the effects of fire frequency on the Pretoriuskop EBPs were similar to the findings on other fire experiments in mesic African savannas. The finding on the Pretoriuskop EBPs differed from the findings in other fire trials that were in arid savannas in Africa. Generally, the exclusion of fire in moist savannas (> 600 mm of rain annually) results in the woody vegetation becoming denser, while the exclusion of fire in arid to semi-arid savannas (< 600mm of rain annually) does not result in the woody vegetation becoming denser. In Pretoriuskop, fires occurring in summer between December and February have a different impact on the density, species composition, and structure of the woody vegetation than fires occurring in winter between August and October. Furthermore, fires occurring in April have a different impact on the density, species composition, and structure of the woody vegetation in Pretoriuskop. Woody vegetation burnt by summer fires is denser than woody vegetation burnt by winter fires. The number of species and species diversity of the woody vegetation is also higher in vegetation burnt by summer fires in comparison with vegetation burnt by winter fires. The density and species composition of woody vegetation in areas that have been burnt in summer fires is more similar to areas where fire has been excluded than to areas that have been burnt in winter fires. The woody species associated with vegetation burnt in summer fires and where fire has been excluded are Euclea natalensis, Antidesma venosum, Diospyros lycioides, Phyllanthus reticulatus, Grewia flavescens, Grewia monticola, Ochna natalitia, Peltophorum africanum, Rhus pyroides, Diospyros mespiliformis, Rhus transvaalensis, Securinega virosa, Putterlickia pyracantha, Rhus pentheri, Commiphora neglecta, Heteropyxis natalensis, and Olea europaea. Structurally the average height of the woody vegetation is taller in areas burnt by winter fires than in areas burnt by summer fires. The woody vegetation in areas burnt in summer fires have more single-stem individuals relative to multi-stem individuals than in areas burnt in winter fires. The structural composition of areas burnt in summer fires is more similar to areas where fire has been excluded than with areas burnt in winter fires. The structure of the woody vegetation in areas burnt in winter fires is generally dominated by multi-stem individuals that are 0-1m tall or 3-5m tall. The structure of the woody vegetation in areas burnt in summer fires or where fire has been excluded is dominated by both single-stem and multi-stem individuals of all heights and basal diameters. Findings regarding the effect of early dry season fires (April) in comparison with late dry season fire (August) on the woody vegetation are consistent with the findings on other fire trails in Africa. However, a comparison of all the fire-timing treatments between the Pretoriuskop and Satara EBPs in the KNP reveals that the timing of fires affects the woody vegetation differently in different areas even when the affects at certain times appear similar. The data collected on the Pretoriuskop EBPs reveals that there have been significant changes in the woody vegetation in Pretoriuskop between 1954 and 1996. The density of the woody vegetation increased between 1954 and 1996 by almost 200%. The number of species and the species diversity of the woody vegetation also increased between 1954 and 1996. In 1954, there were approximately equal numbers of single-stem and multi-stem individuals, while in 1996 there were more multi-stem individuals than single-stem individuals. The increase in atmospheric CO2 levels between 1954 and 1996 is believed to have been a factor that has driven the changes in the woody vegetation of Pretoriuskop between 1954 and 1996.

Fire is the primary natural disturbance vital to the ecological integrity of Quetico Provincial Park, Ontario, Canada. A new provincial park planning process (i.e., Class Environmental Assessment) has required the review of Quetico’s Fire Management Plan. To support this review, large and severe (stand-replacing) Quetico fires were studied using 1966 Ontario Ministry of Natural Resources (OMNR) forest resource inventory (FRI) mapping. A Geographic Information Systems (GIS) database of the FRI was created and updated with the OMNR digital fire atlas. This database was used as a time-since-fire and fire interval dataset to estimate fire frequency. It also served to archive the 1966 FRI for the largest protected area in the transition between the Boreal and Great Lakes-St. Lawrence forest regions. Non-parametric (Kaplan-Meier) survival analysis was used to estimate survival functions and mean fire intervals (i.e., the expected time between two consecutive stand-replacing fires for any location within the Park). Previous studies that have used Kaplan-Meier survival analysis methods have based fire frequency estimates solely on time-since-fire data. However, time-since-fire data cannot be equated with fire interval data when using non-parametric methods. At least one fire interval is required to obtain reliable results. The mean fire interval for the entire 475,782 ha Park between the years 1668 and 2007 was 230 years. Performing the analysis on various geographic and temporal partitions revealed fire frequency spatial and temporal variability. A constant (independent of time-since-fire) probability of burning was not observed for Quetico which is contrary to accepted conjecture for northwestern Ontario boreal/mixed-wood forests. A current fire cycle was also estimated for the Park (342 years) using the digital fire atlas. The results suggested that use of historical static fire frequency estimates as fire management prescriptions may not be justified given considerable fire frequency temporal variability. The observed fire frequency spatial variability suggests that studies should be undertaken at coarser scales than is the norm to characterise the regions fire regime in support of landscape level fire management planning.

Fire is the primary natural disturbance vital to the ecological integrity of Quetico Provincial Park, Ontario, Canada. A new provincial park planning process (i.e., Class Environmental Assessment) has required the review of Quetico’s Fire Management Plan. To support this review, large and severe (stand-replacing) Quetico fires were studied using 1966 Ontario Ministry of Natural Resources (OMNR) forest resource inventory (FRI) mapping. A Geographic Information Systems (GIS) database of the FRI was created and updated with the OMNR digital fire atlas. This database was used as a time-since-fire and fire interval dataset to estimate fire frequency. It also served to archive the 1966 FRI for the largest protected area in the transition between the Boreal and Great Lakes-St. Lawrence forest regions. Non-parametric (Kaplan-Meier) survival analysis was used to estimate survival functions and mean fire intervals (i.e., the expected time between two consecutive stand-replacing fires for any location within the Park). Previous studies that have used Kaplan-Meier survival analysis methods have based fire frequency estimates solely on time-since-fire data. However, time-since-fire data cannot be equated with fire interval data when using non-parametric methods. At least one fire interval is required to obtain reliable results. The mean fire interval for the entire 475,782 ha Park between the years 1668 and 2007 was 230 years. Performing the analysis on various geographic and temporal partitions revealed fire frequency spatial and temporal variability. A constant (independent of time-since-fire) probability of burning was not observed for Quetico which is contrary to accepted conjecture for northwestern Ontario boreal/mixed-wood forests. A current fire cycle was also estimated for the Park (342 years) using the digital fire atlas. The results suggested that use of historical static fire frequency estimates as fire management prescriptions may not be justified given considerable fire frequency temporal variability. The observed fire frequency spatial variability suggests that studies should be undertaken at coarser scales than is the norm to characterise the regions fire regime in support of landscape level fire management planning.

The savannas of southern Africa are a highly variable and globally-important biome supporting rapidly-expanding human populations, along with one of the greatest concentrations of wildlife on the continent. Savannas occupy a fifth of the earth's land surface, yet despite their ecological and economic significance, understanding of the complex couplings and feedbacks that drive spatiotemporal patterns of change are lacking. In Chapter 1 of my dissertation, I discuss some of the different theoretical frameworks used to understand complex and dynamic changes in savanna structure and composition. In Chapter 2, I evaluate spatial drivers of water quality declines in the Chobe River using spatiotemporal and geostatistical modeling of time series data collected along a transect spanning a mosaic of protected, urban, and developing urban land use. Chapter 3 explores the complex couplings and feedbacks that drive spatiotemporal patterns of land cover (LC) change across the Chobe District, with a particular focus on climate, fire, herbivory, and anthropogenic disturbance. In Chapter 4, I evaluated the utility of Distance sampling methods to: 1) derive seasonal fecal loading estimates in national park and unprotected land; 2) provide a simple, standardized method to estimate riparian fecal loading for use in distributed hydrological water quality models; 3) answer questions about complex drivers and patterns of water quality variability in a semi-arid southern African river system. Together, these findings have important implications to land use planning and water conservation in southern Africa's dryland savanna ecosystems. / Ph. D.