Zhiwei Wu

Relative effects of climatic and local factors on fire occurrence in boreal forest landscapes of northeastern China

Fire significantly affects species composition, structure, and ecosystem processes in boreal forests. Our study objective was to identify the relative effects of climate, vegetation, topography, and human activity on fire occurrence in Chinese boreal forest landscapes. We used historical fire ignition for 1966-2005 and the statistical method of Kernel Density Estimation to derive fire-occurrence density (number of fires/km(2)). The Random Forest models were used to quantify the relative effects of climate, vegetation, topography, and human activity on fire-occurrence density. Our results showed that fire-occurrence density tended to be spatially clustered. Human-caused fire occurrence was highly clustered at the southern part of the region, where human population density is high (comprising about 75% of the areas population). In the north-central areas where elevations are the highest in the region and less densely populated, lightning-caused fires were clustered. Climate factors (e.g., fine fuel and duff moisture content) were important at both regional and landscape scales. Human activity factors (e.g., distance to nearest settlement and road) were secondary to climate as the primary fire occurrence factors. Predictions of fire regimes often assume a strong linkage between climate and fire but usually with less emphasis placed on the effects of local factors such as human activity. We therefore suggest that accurate forecasting of fire regime should include human influences such as those measured by forest proximity to roads and human settlements.

Comparing Effects of Climate Warming, Fire, and Timber Harvesting on a Boreal Forest Landscape in Northeastern China

Forest management under a changing climate requires assessing the effects of climate warming and disturbance on the composition, age structure, and spatial patterns of tree species. We investigated these effects on a boreal forest in northeastern China using a factorial experimental design and simulation modeling. We used a spatially explicit forest landscape model (LANDIS) to evaluate the effects of three independent variables: climate (current and expected future), fire regime (current and increased fire), and timber harvesting (no harvest and legal harvest). Simulations indicate that this forested landscape would be significantly impacted under a changing climate. Climate warming would significantly increase the abundance of most trees, especially broadleaf species (aspen, poplar, and willow). However, climate warming would have less impact on the abundance of conifers, diversity of forest age structure, and variation in spatial landscape structure than burning and harvesting. Burning was the predominant influence in the abundance of conifers except larch and the abundance of trees in mid-stage. Harvesting impacts were greatest for the abundance of larch and birch, and the abundance of trees during establishment stage (1–40 years), early stage (41–80 years) and old- growth stage (>180 years). Disturbance by timber harvesting and burning may significantly alter forest ecosystem dynamics by increasing forest fragmentation and decreasing forest diversity. Results from the simulations provide insight into the long term management of this boreal forest.

Defining fire environment zones in the boreal forests of northeastern China.

Fire activity in boreal forests will substantially increase with prolonged growing seasons under a warming climate. This trend poses challenges to managing fires in boreal forest landscapes. A fire environment zone map offers a basis for evaluating these fire-related problems and designing more effective fire management plans to improve the allocation of management resources across a landscape. Toward that goal, we identified three fire environment zones across boreal forest landscapes in northeastern China using analytical methods to identify spatial clustering of the environmental variables of climate, vegetation, topography, and human activity. The three fire environment zones were found to be in strong agreement with the spatial distributions of the historical fire data (occurrence, size, and frequency) for 1966-2005. This paper discusses how the resulting fire environment zone map can be used to guide forest fire management and fire regime prediction.

Thematic and Spatial Resolutions Affect Model-Based Predictions of Tree Species Distribution

Subjective decisions of thematic and spatial resolutions in characterizing environmental heterogeneity may affect the characterizations of spatial pattern and the simulation of occurrence and rate of ecological processes, and in turn, model-based tree species distribution. Thus, this study quantified the importance of thematic and spatial resolutions, and their interaction in predictions of tree species distribution (quantified by species abundance). We investigated how model-predicted species abundances changed and whether tree species with different ecological traits (e.g., seed dispersal distance, competitive capacity) had different responses to varying thematic and spatial resolutions. We used the LANDIS forest landscape model to predict tree species distribution at the landscape scale and designed a series of scenarios with different thematic (different numbers of land types) and spatial resolutions combinations, and then statistically examined the differences of species abundance among these scenarios. Results showed that both thematic and spatial resolutions affected model-based predictions of species distribution, but thematic resolution had a greater effect. Species ecological traits affected the predictions. For species with moderate dispersal distance and relatively abundant seed sources, predicted abundance increased as thematic resolution increased. However, for species with long seeding distance or high shade tolerance, thematic resolution had an inverse effect on predicted abundance. When seed sources and dispersal distance were not limiting, the predicted species abundance increased with spatial resolution and vice versa. Results from this study may provide insights into the choice of thematic and spatial resolutions for model-based predictions of tree species distribution.

Comparing fuel reduction treatments for reducing wildfire size and intensity in a boreal forest landscape of northeastern China

Fuel load is often used to prioritize stands for fuel reduction treatments. However, wildfire size and intensity are not only related to fuel loads but also to a wide range of other spatially related factors such as topography, weather and human activity. In prioritizing fuel reduction treatments, we propose using burn probability to account for the effects of spatially related factors that can affect wildfire size and intensity. Our burn probability incorporated fuel load, ignition probability, and spread probability (spatial controls to wildfire) at a particular location across a landscape. Our goal was to assess differences in reducing wildfire size and intensity using fuel-load and burn-probability based treatment prioritization approaches. Our study was conducted in a boreal forest in northeastern China. We derived a fuel load map from a stand map and a burn probability map based on historical fire records and potential wildfire spread pattern. The burn probability map was validated using historical records of burned patches. We then simulated 100 ignitions and six fuel reduction treatments to compare fire size and intensity under two approaches of fuel treatment prioritization. We calibrated and validated simulated wildfires against historical wildfire data. Our results showed that fuel reduction treatments based on burn probability were more effective at reducing simulated wildfire size, mean and maximum rate of spread, and mean fire intensity, but less effective at reducing maximum fire intensity across the burned landscape than treatments based on fuel load. Thus, contributions from both fuels and spatially related factors should be considered for each fuel reduction treatment.

A coupled modeling framework for predicting ecosystem carbon dynamics in boreal forests

Abstract Carbon stocks in boreal forests play an important role in global carbon balance but are sensitive to climate change and disturbances. Ecological models offer valuable insights into the effects of climate change and disturbances on boreal forests carbon stocks. However, the current pixel-based model coupling approaches are challenging to apply over large spatial extents because high computational loads and model parameterizations. Therefore, we developed a new framework for coupling a forest ecosystem and a landscape model to predict aboveground and soil organic carbon stocks at the ecoregion level. Our results indicated that the new model-coupling framework has some advantages on computation efficiency and model validation. The model results showed that carbon stocks and its spatial distribution were significantly influenced by fire, harvest, and their interactions. Simulation results showed that boreal forests carbon stocks are vulnerable to loss because of future potential disturbances, complicating efforts to offset greenhouse gas emissions through forest management.

Scale effects of vegetation and topography on burn severity under prevailing fire weather conditions in boreal forest landscapes of Northeastern China

Understanding the controlling factors of burn severity requires consideration of the scale at which these factors work. This investigation explored how well topography and vegetation factors can explain variation of burn severity in a boreal forest landscape of northern China under prevailing fire weather conditions. Eight grain sizes were examined that ranged from 30 to 2500 m. A burn severity map was derived from calculating the difference between pre- and post-fire Normalized Difference Vegetation Index of two Landsat Thematic Mapper images. Results indicate that (1) burn severity in the boreal forest landscape of northern China was mainly controlled by vegetation at grain sizes smaller than 500 m. At grain sizes larger than 1000 m, topography accounted for more variation in burn severity; (2) the relative importance of topography factors was stable with increasing grain sizes and generally ranked in order of aspect, slope, and elevation; (3) stand age appeared to be more important where canopy cover and understory cover substantially fluctuated with increasing grain sizes; and (4) the linear relationships between burn severity and specific factors of topography and vegetation decreased with increasing grain sizes. Our study can help managers to design fire management plans according to vegetation characteristics that are found important in controlling burn severity and prioritize management locations based on the relative importance of vegetation and topography.