Darrell C. West

Holocene Fires in the Northern Amazon Basin

Abstract The ubiquitous occurrence of charcoal in the forest soils of the Upper Rio Negro region of Colombia and Venezuela indicates the presence of frequent and widespread fires in the Amazon Basin, possibly associated with extremely dry periods or human disturbances. Charcoal ranged from 3.12 to 24.76 mg/cm 3 in the upper 50 cm of soil and was more abundant in Oxisols and Ultisols than in other soil types. Charcoal dates range from 6260 yr B.P. to the present. Several dates coincide with dry phases recorded during the Holocene. Ceramic shards were found at several sites, and thermoluminescence analysis indicates that their ages range from 3750 to 460 yr B.P. The age of charcoal and shards confirms that this region has been subjected to fire and human disturbances during the past 6000 yr.

Simulating the Role of Climate Change and Species Immigration in Forest Succession

Forest succession in its traditional sense implies two important features that resist direct examination. First, classical definitions of forest succession generally connote directional changes in species composition and in community structure through time. Following a major disturbance, a directional succession of tree species in our temperate forests may involve 200 (Oosting 1942), 400 (McAndrews 1976), or even 1000 years (Franklin and Hemstrom Chapter 14). There is hardly the opportunity to observe the direction, let alone the species succession, that occurs within our lifetime.

Two High-Throughput Techniques for Determining Wood Properties as Part of a Molecular Genetics Analysis of Hybrid Poplar and Loblolly Pine

Two new high-through put techniques, computer tomography X-ray densitometry (CT scan) and pyrolysis molecular beam mass spectrometry (pyMBMS), coupled with quantitative trait loci (QTL) analysis, were tested as a means to overcome the time and cost associated with conventional characterization of biomass feedstock components. Applications of these two techniques were evaluated using hybrid poplar for the CT scan and loblolly pine for the pyMBMS. Segregating progeny from hybrid poplar varied in specific gravity, with individual mean estimates ranging from 0.21–0.41. Progeny from loblolly pine varied in lignin, α cellulose, and mannan contents, with individual mean estimates of lignin content ranging from 28.7–33.1%, α cellulose content from 28.8–43.5% and mannan content from 4.2–10.1%. QTL analysis of the loblolly pine data suggested that eleven QTLs were associated with individual feedstock characteristics and that two QTLs for several feedstock components were linked to the same position on the loblolly pine genetic map. Each QTL individually accounted for 7–13% of the total phenotypic variation in associated loblolly pine feedstock components.

Testing a simulation model for reconstruction of prehistoric forest-stand dynamics

Abstract Three characteristics of the output of a forest-stand simulation model were matched to pollen records of actual vegetation in central Tennessee. Temporal shifts of individual pollen taxon frequencies were compared to shifts of individual plant species frequencies in simulated biomass for the last 16,000 yr. Individual pollen profiles (temporally ordered species frequencies) were also compared to simulated biomass profiles during that period. Modern ratios of pollen to vegetation composition (R values) were compared with those calculated from simulated biomass percentages and fossil pollen percentages. The model output was similar to the comparable characteristics of the pollen record. The model output is therefore a plausible description of vegetation characteristics at the site of pollen deposition in central Tennessee. The model produced information unavailable from other sets of prehistoric data. This information describes the invasion and growth of the yellow-poplar which produces no windborne pollen, and of palynologically indistinguishable oak and pine species. These results suggest that many paleoecological questions can be answered through appropriate simulation modeling studies.

Patterns and Dynamics of Forests: An Application of Simulation Models

Studies of forest succession characteristically rely heavily on inference to order the spatial pattern of forest stands (usually on a regularly disturbed landscape) into a temporal pattern. The inferential structure used to arrange these data is, in a sense, a “theory of succession,” and the relation between the artificially ordered successional pattern and the ordering theory are, of necessity, tautological. This circularity persists regardless of the particular theory of succession (monoclimax, polyclimax, multiple paths, etc.) to which an ecologist happens to subscribe. The fact that both the observation data sets and the theory may both be quite elaborate makes it difficult to test the theories of succession. In this chapter, we shall take an alternate approach to studying successional dynamics. Instead of a direct and long-term observation of vegetation plots or a synthetic ordering of several plots, we shall use a set of reasonably detailed computer models of forest dynamics to explore system-level responses of different ecosystems. This simulation technique does not rely upon inferences made regarding temporal and spatial scales as do empirical studies of vegetation dynamics.

Environmental gradients in a simulation model of a beech-yellow-poplar stand☆

Abstract A simulation model for growth and succession of a hypothetical American-beech–yellow-poplar forest has been developed to study changes in response of this simple community under conditions of a slowly varying climate. As the temperature, through a model parameter equal to the number of growing degree days (GDD), is increased, a sharp discontinuity in response of the model is noted at approximately 4500 GDD. A similar discontinuity is observed at about 4800 GDD as the temperature is slowly decreased. This hysteretic response with width of 300 GDD can be compared for consistency against the rather sharp boundaries which occur between forest communities along smooth environmental gradients. Although many environmental gradients are responsible for transitions is natural systems, the model calculations described indicate that changes in a single environmental variable, temperature, can account for transition zones by affecting competitive ability.

Forests of the Southeastern United States: Quantitative Maps for Abovegound Woody Biomass, Carbon, and Dominance of Major Tree Taxa

Summary maps presented in this paper depict the current distribution patterns of aboveground woody biomass and carbon content of commercial forests in the southeastern United States. Contour maps of dominance of 10 important southeastern tree taxa are included. These maps were compiled from growing-stock volume data available from the Continuous Forest Inventories (CFI) of the United States Department of Agriculture (USDA) Forest Service, Southern and Southeastern Forest Experiment Stations. The maps of woody biomass and carbon establish the current patterns for standing crop and carbon storage within forests of the southeastern United States, an area comprising 15% of the temperate forest zone of the earth. The contour maps of tree dominance quantify variation in composition within and between forest types previously mapped only qualitatively by the Forest Service. These dominance maps illustrate the individualistic nature of the distributions in response to gradients of environmental factors and effects of land management practices.

Spectral analysis of forest model time series

Abstract The use of spectral analysis to elucidate the cyclic behavior in time series generated by a forest stand growth simulation model is discussed. A stand-level simulator, FORET, for an Appalachian deciduous forest is described. An estimate of the power spectral density of the total biomass time series is calculated. The power spectral density estimate indicates a dominant cyclic behavior with a period of about 200 years. In addition the spectral density is approximately bandlimited. This characteristic makes possible the application of the sampling theorem for analysis of sampling rates.

Predicting Afforestation Success During Climatic Warming at the Northern Limit of Forests

Global temperature increases from greenhouse gases are expected during the 21st Century, possibly as early as the next decade. Warming is predicted to be greatest at highest latitudes. Initial attempts to document climate change will be hampered by the great inter-annual variability in weather at high latitudes and the scarcity of long-term weather records. Certain key properties of these environmental changes, however, can be defined, despite uncertainty concerning results from present climate models. A group of scientists meeting at Villach, Austria, in 1987 (Jaeger, 1988) agreed that 3 °C was a moderate estimate of average global warming during the next century, while at high latitudes (60 to 90° N.), 0.6 to 0.7 °C per decade in winter and 0.1 to 0.2 °C per decade in summer are more probable climate change expectations. Results from the most recent climate model results are not very different and range from 0.4 to 0.8 °C per decade (Mitchell et al., 1990). In addition, results from both the Villach group and the more recent compilations agree that winter precipitation and soil moisture could increase at high latitudes as more precipitation falls, and as more precipitation falls as rain instead of snow. These climate changes are quite intense and rapid. For the next few decades, they may result in a much wider range of temperature and precipitation extremes than has ever been recorded in cold regions.