LD1-2: Seasonality of fire combined with legacies from past fire events control fire severity, which in turn regulates the site thermal/moisture regime (LD2-2), soil carbon storage (LD1-3, BG2-1), and post-fire recruitment and succession (FD1-3, FD3-1)

Previous LTER research at Bonanza Creek has focused on the role of state variables in controlling many important forest ecosystem processes (Van Cleve et al. 1993, 1996). With respect to variations in soil temperature and moisture and depth of organic soils, topographic controls on solar illumination were thought to have a dominant control on soil temperature, permafrost formation, soil drainage and rates of decomposition, which in turn influence, depth of the ground-layer organic matter (Van Cleve and Viereck 1981). In flat areas without topography, some have suggested that the drainage properties of the underlying mineral soil layer influences soil moisture and permafrost formation (Harden et al. 2000). Our recent LTER-sponsored research has shown that in black spruce forests located on sites with no topographic variation and little variation in the underlying mineral soil profile (and hence drainage), that soil temperature decreases as a function of the depth of the organic mat layer, while soil moisture increases (see Figure 1). Furthermore, measurements of soil temperature and moisture in a nine-year old burn show that depth of burning of organic soils in black spruce forests strongly controls post-fire patterns of soil temperature and moisture (Figure 2). We hypothesize that this relationship indicates that ecosystem processes, and not state variables, are interacting under some circumstances to control patterns of organic soil depth, in particular, that there is a positive feedback between fire severity (as measured by depth of remaining organic soil) and subsequent soil temperature and moisture conditions, which in turn controls subsequent fire severity.

The interactions between topography, soil drainage, and organic mat formation in black spruce forests in interior Alaska are undoubtedly complex, and more observations are required to understand the factors controlling variations in organic mat depth. The research associated with this hypothesis is designed to answer the following questions:

1. How does organic mat depth in black spruce forests vary as a function of site topography, elevation and soil drainage characteristics?

Our studies in the Delta Junction region has shown that mature (> 75 yrs since the last fire) stands of black spruce have a wide range in the depth of the organic mat layer, ranging between 10 and >34 cm. We ascribe these differences primarily to variations in organic depth associated with depth of burning during previous fires, with mineral soil drainage also playing a secondary role. We do not have enough data to derive conclusions for other sites primarily because data are lacking. We intend to collect additional data in order to understand the degree of variability that exists in a broad range of topographic settings. As a starting point, we will analyze data collected as part of the N/S and E/W transects collected by T. Nettleton and M. Walker during previous LTER studies. This research was focused on understanding how species composition and canopy tree characteristics varied as a function of topography. During this study, a limited number of measurements were made on organic and mineral soil characteristics of the different sites. As part of this study, we will collect additional organic mat depth measurements. We plan to collect organic soil depths along two 100 m long transects located 100 m apart, sampled every 5 m. For each sample point, we will extract a 20 by 20 cm core and measure the depth of the different organic mat layers. With this sample size, we can analyze the distribution of the depths of various sites. We will statistically analyze these data to determine if trends in organic soil depths as a function of topography and drainage exist. Additional sites will be sampled if it is determined that specific topography/soil drainage classes have been under-sampled.

2. How do soil temperature and moisture conditions in black spruce forests vary as a function of organic mat depth and local topographic and soil drainage characteristics?

Figure 1 shows a clear pattern between the temperature and moisture conditions of the top 20 to 30 cm of mineral soil and the depth of the organic mat at a site. While we expect similar trends to exist at other sites, we would expect these trends to be somewhat altered by local topography. To investigate this question, we will collect end of season soil temperature and moisture measurements from a number of sites. Ideally, we would monitor soil temperature and moisture throughout the growing season, but the cost of installing and maintaining the instrumentation required to do so is prohibitive. In stead, we will collect mineral soil temperature and moisture at the end of the growing season (e.g., end of August/early September) based on the assumption that these measures integrate the effects of season-long temperature and precipitation trends. At each site, we will dig three separate soil pits to the depth of permafrost or to the depth of gravel or bedrock, which ever comes first. In instances where gravel is encountered, we shall excavate the pit to a depth of 30 cm into the mineral soil layer. Using digital instrumentation, we shall measure soil moisture and temperature at 5 cm depth increments in the mineral soil layer. These data will allow us to statistically explore through regression analysis the relationship between soil temperature and moisture and site conditions. Note that this sampling approach is very efficient. We have found that two people can sample a single site within 30 minutes, which allows sampling of a large number of sites over a short time period.

3. How do variations in site conditions and timing of the burn during the growing season influence depth of burn into the organic mat layer?

We have already extensively sampled some 15 different fire events throughout the interior Alaska, with some sites having multiple sample sites. We need to extend this sample from several aspects. In some sites, we only sampled 1 or 2 sites within the burned area, and additional sites need to be sampled in order to capture the spatial variability in reduction of organic soil depth that naturally occurred within the event as well as features of the burn where burn severity is controlled by topography or timing of the burn during the growing season. In addition, we need to add more sites in order to capture variations associated with different factors in a statistically meaningful fashion. We have identified a number of fire events that recently have occurred and are road accessible which provide logical candidates for sampling. These include a number of fire events that occurred in the 1999 to 2003 time period. In addition, there are likely to be additional burn events that occur during the upcoming years that can be sampled. For each site selected, we will collect samples in the same fashion as identified above, e.g., we will establish two 100 m long transect lines, and measure organic mat depths every 5 m. For each burn event, we will collect organic soil depths from adjacent unburned stands in order to estimate the pre-burn conditions. We will collect tree cores from 5 canopy trees at each site in order to estimate the age of the stand.

4. How do variations in burn severity (organic mat consumption)?

Data will be collected within different burns to determine how fire severity effects soil temperature and moisture, and how this relationship varies as a function of topography and soil drainage. We will select a number of burns that had variation in terms of depth of organic soil remaining after a fire, and collect measurements in the same fashion as discussed in Question 2 above. If possible, we will collect data from burns in different topographic settings in order to understand how aspect and elevation influence soil temperature.

Figure 1. Average soil temperature in the top 30 cm (a) and soil moisture in the top 20 cm (b) from unburned black spruce forest stands adjacent to the HC94 fire. These data were collected between 17 and 20 August 2003. The error bars represent the range between high and low values collected from 3 soil pits in each site.

Figure 2. Average soil temperature in the top 30 cm (a) and soil moisture in the top 20 cm (b) from burned black spruce forest stands adjacent in the HC94 fire. These data were collected between 17 and 20 August 2003. Note that the unburned soil moisture in the black spruce sites without tussocks was lower than in the sites with tussocks (44 vs. 66%).