US-3: RM Lower Treeline

PIRE US-3a: “RM Lower Treeline” Project Plan-Colorado & Greater Yellowstone

Title:  Tree-ring fire history and forest dynamics in lower montane Rocky Mountain ecosystems

Investigators:  Veblen, Tepley

Students:  Naficy (PhD CU), Rother (PhD CU), Kemp (U Idaho)

Opportunities for intern participation:  none, unless tied to overseas internship

Project Description

The project will be implemented largely as Ph.D. dissertations currently being designed.  Based on draft research designs (below), fieldwork will be initiated in late May 2011. It is likely that some of the objectives and methods will need to be modified based on initial field experiences.   In addition to the two dissertations, Tepley will develop a research activity under this project which will be planned during May and implemented beginning in June 2011.

Draft Project Description:  Controls on spatio-temporal variation in fire severity and fire effects in mixed severity fire regime forests of the Northern Rockies (Grad student Cameron Naficy)

Increases in wildfire frequency and the annual area burned have been documented globally and within the western U.S., specifically, in recent decades (Pechony & Shindell 2010, Littell et al 2009, Westerling et al. 2006).  Trends in fire severity, however, are less clear but are critical to understanding the ecological and societal effects of wildfires.  The historical patterns of burn severity, their biophysical controls, and the ecological consequences of shifts in burn severity are poorly understood. This is particularly true for forests characterized by mixed-severity fire regimes where high spatio-temporal variability in burn severity patterns has hindered theoretical and empirical advances. As a result, they have been far less studied than low- and high-severity fire regime forests despite their broad distribution across the western U.S. (Hessburg et al. 2007, Sherriff & Veblen 2007, Lentile et al. 2005). This research encompasses three central objectives:

1. To quantify spatial and temporal variation in contemporary and historical burn severity patterns in mixed severity fire regime forests of the Northern Rockies, both at a landscape scale and within frequent-fire forest types specifically.

2. To evaluate the relative importance of climate variability versus local-scale topographic and vegetation characteristics in structuring heterogeneity in burn severity patterns of frequent-fire forests.

3. To evaluate the consequences of variable fire severities and frequencies for tree mortality and post-fire tree establishment that determine forest succession dynamics across a range of biophysical conditions in landscapes characterized by relatively frequent fire.

This research will combine remote sensing and dendroecological techniques to evaluate these objectives for forests of pure Pseudotsuga menziesii and mixed-conifer Pinus ponderosa/Larix occidentalis, two of the dominant lower-middle elevation forest types in the Northern Rockies. For Pseudotsuga menziesii forests, our primary study areas will be located in the Madison and Paradise Valleys located in the northwestern portion of the Greater Yellowstone Ecosystem. Daley Lake, which has been sampled by the Whitlock Lab (MSU), is the only sampled lake outside of YNP. For Pinus ponderosa/Larix occidentalis forests, our study area will potentially include the Middle and South Forks of the Flathead River, or the Swan Valley and Monture Creek areas.

Draft Project Description:  Spatial and temporal variablity of post-fire conifer regeneration in lower treeline forests of the U.S. Rocky Mountains: Implications of a warmer, drier climate (Graduate Student Monica Rother)

Preliminary observations in the Colorado Front Range at recently burned lower montane forests of  ponderosa pine and Douglas fir indicate highly variable patterns of conifer regeneration.  Post-fire conifer establishment is surprisingly minimal at many sites that burned in the late 1980s to early 2000s. The relatively high proportion of recently burned lower montane forests exhibiting no or little conifer establishment contrasts with evidence from retrospective studies documenting abundant post-fire conifer establishment following burning in the 19th century (Veblen and Lorenz 1986; Mast et al. 1998; Ehle and Baker 2003).  Previous research has documented the sensitivity of ponderosa pine establishment to interannual climatic variability in Arizona (Savage et al. 1996) and at grassland ecotones in the Front Range (League and Veblen 2006). These previous studies and preliminary observations have led to the general hypothesis that under the warmer climatic conditions of the Front Range of the past c. 25 years, conditions have been less suitable for post-fire tree establishment in comparison with the 19th century. 

Our research will examine post-fire conifer regeneration in lower montane forests of the Colorado Front Range and Greater Yellowstone Area in relation to biophysical settings, fire severity, and recent climate variability.  We will compare our findings on recent post-fire seedling/sapling densities to existing studies of historic post-fire conifer densities to determine whether current patterns deviate from past patterns. Additionally, we will assess climate-regeneration relationships by examining climatic conditions during the year of establishment (Savage et al. 1996; League and Veblen 2006). The first year of the dissertation research will include extensive sampling of recent post-fire conifer regeneration in lower treeline environments in the Front Range and in the GYA.  We will address three primary research objectives:

1.  Quantify post-fire conifer establishment and survival (i.e. current juvenile densities) across a range of lower treeline sites that have burned since the mid-1980s in the Front Range and in the GYA.

2.  Examine spatial variability of post-fire conifer regeneration in relation to site factors including topographic variables (e.g. aspect, slope steepness, elevation), fire severity (% tree mortality, scorch heights), competing vegetation (% cover of herbs and woody species), and distance to seed source.

3.  Analyze relationships between annual climatic variability and post-fire conifer establishment and survival based on tree-ring dating of tree establishments.

Related Activities

Research will provide some tree-ring fire dates for comparison with charcoal records from lakes (i.e. PIRE US-2 Colo RM/GYE) and work in PIRE US-3b extends fire-climate linkages into Idaho.

Year 2 Update:

Controls on spatio-temporal variation in fire severity and fire effects in mixed severity fire regime forests of the Northern Rockies (Naficy dissertation research)

CU PhD student Naficy is studying mixed-severity fire regimes that exhibit highly variable patterns of fire effects that are influenced both by temporal and spatial drivers. The high spatiotemporal heterogeneity of fire effects in mixed-severity fire regimes has hindered theoretical and empirical advances of their fire ecology, despite increasing recognition of their ecological importance and widespread distribution. Currently, there is only limited understanding of the historical patterns of fire severity in mixed-severity fire regime forests, their biophysical controls, and the ecological consequences of shifts in fire severity that may have occurred in response to land management practices and climate change. This research encompasses two central objectives: (1) characterize the spatiotemporal variation of fire severity (~1800 AD to present) in mixed-severity fire regime forests of the northern U.S. Rockies; and (2) assess the influence of interannual to interdecadal climate variability, topography and vegetation on spatiotemporal variation of fire severity for both historical and modern wildfires.  This research employs satellite imagery analysis, aerial photograph-based reconstructions, and dendroecological techniques across Douglas-fir (Pseudotsuga menziesii) forests of the Greater Yellowstone Ecosystem (GYE) and western larch (Larix occidentalis) forests of the Northern Continental Divide Ecosystem (NCDE), two of the dominant lower-middle elevation forest types in the Northern Rockies believed to exhibit a mixed-severity fire regime.

During summer of 2011, dendroecological sampling was conducted in 19 Douglas-fir sites in two drainages of the northern GYE. From these 19 sites, approximately 1200 tree cores and 80 fire scars were collected. Sampling of similar intensity is scheduled for summer 2012 in western larch forests of the NCDE. Sample sites in the NCDE are being stratified by historical forest patch structure using the Hessburg dataset so as to construct high-resolution spatiotemporal accounts of historical fire effects and evaluate the potential for different cross-scale approaches in the assessment of historical fire effects.

Spatial and temporal variability of post-fire conifer regeneration in lower treeline forests of the U.S. Rocky Mountains: Implications of a warmer, drier climate (Rother dissertation research)

CU PhD student Rother and Veblen collected data in seven post-burn areas of lower treeline forests of the Colorado Front Range. Year of burn ranged from 1988 to 2003 and all burns were of mixed severity. Data collection targeted four research objectives: (1) quantify post-fire conifer establishment and survival by examining the density of conifer juveniles across a range of lower treeline sites; (2) examine the spatial variability of juvenile conifer densities in relation to site factors such as fire severity (as indicated by percent tree mortality), competition with herbaceous and woody species, distance to seed source, and topographic variables including elevation and slope aspect; (3) analyze relationships between post-fire conifer regeneration and annual climate variability to determine if certain climate restraints limit or favor conifer establishment and survival; and (4) experimentally manipulate microclimate to determine effects on conifer seedling survival and growth in a post-disturbance environment.

To address objectives 1 and 2, a stratified random sampling approach was used whereby the selection of sample sites were stratified by slope aspect and fire severity. Fire severity was determined through use of the Differenced Normalized Burn Ratio (dNBR) provided by the Monitoring Trends in Burn Severity (MTBS) Project and was verified in the field through assessment of percent mortality. Slope aspect consisted of two classes (north-facing and south-facing) while fire severity consisted of three classes (low, moderate, and high). In each of the six settings, the midpoints of belt transects were randomly located. Each transect measured 2 × 50 m and was divided into 10 quadrats of 2 × 5 m each. Ground area covered by life-form type (graminoid, forb, tree seedling, mature tree, shrub, and fern) was estimated in each quadrat. For present juvenile conifers, the species, height, and diameter at base height were recorded. In addition, the distance to nearest seed source and the areal coverage of bare mineral soil, litter, rock-covered surface, and dead wood was recorded. Thus far, data from an average of 15 transects (150 quadrats) per fire were collected, mostly in high-severity settings. In the summer of 2012, additional data will be collected, and findings will be compared with existing retrospective studies of conifer regeneration in lower treeline forests to assess whether or not observed seedling densities are within documented historic ranges.

To address objective 3, a large dataset of tree establishment dates was developed by harvesting tree seedlings and saplings in the post-burn areas. In the summer of 2011, approximately 100 samples from juvenile conifers were collected in five of the seven burn areas of the Colorado Front Range. Areas of relatively abundant juvenile conifers were first subjectively identified for sampling. Then, within these areas, two small plots were randomly located. Plot size varied to include a minimum of 50 juvenile trees (<150 cm tall). A handsaw was used to remove a 10-cm-long section from the juvenile that was centered on the root collar. The samples are now being processed and analyzed. We are cutting each sample 2 cm long successive cross sections. These successive cross sections are then sanded and examined under a microscope to determine the location of the root-shoot boundary, where pith first appears. Once establishment dates for a given site are identified, we will analyze the data to determine whether common years of establishment correspond with certain climatic conditions.

To address objective 4, a field experiment was developed this spring in a lower treeline area of the Colorado Front Range. A 35 x 35 m macroplot was subdivided into 100 grid cells. The central 1.75 x 1.75 m area of each grid cell was identified as the treatment area.  All vegetation (primarily grasses) was manually removed until a bare mineral soil surface was achieved. This procedure partially simulates the effects of wildfire. Tree seedlings of both ponderosa pine (Pinus ponderosa) and Douglas-fir (Pseudotsuga menziesii) (7 of each) were then planted in each treatment area, for a total of 1400 seedlings. Beginning in early June, experimental manipulations of both temperature and moisture will be initiated. Open-top chambers will be used to increase soil and air temperature and watering treatments will be used to increase soil moisture. 

Spatial reconstruction of fire history and forest dynamics in mixed conifer forests of the San Juan Mountains, Colorado (Tepley post-doc research comparative study to NZ-3)

Mixed-conifer forests at mid elevations of the U.S. Rocky Mountains span a broad range of environmental conditions from nearly pure ponderosa pine forests at lower elevations to spruce-fir forests at higher elevations. Mixed-conifer forests have complex disturbance regimes and forest development patterns, yet much of the research aimed at understanding historical disturbance regimes has focused on the drier end of the continuum, where field methods and analysis techniques address frequently burned ponderosa pine forests. These relatively dry forests generally support trees with multiple fire scars that can easily be sampled for interpretations of fire frequency and fire-climate relationships.  In contrast, multiple-scarred trees typically are less abundant in the wetter mixed-conifer forests. The lower abundance of trees with multiple scars in the wetter mixed-conifer forests is due in part to changes in species composition from dry to wet mixed-conifer forests, with greater abundance of species susceptible to heartrot, which are unlikely to preserve a long record of multiple fires. Wet mixed-conifer forests also are characterized by greater representation of moderate- and high-severity fire effects that tend to kill rather than scar trees. As a result, the historical disturbance dynamics of wet mixed-conifer forests remains poorly understood, and current management in these forests is drawn from the more frequently burned, drier mixed-conifer forests.

In Year 2, field work was initiated to evaluate and reconstruct historical disturbance patterns and forest dynamics across the continuum from warm/dry to cool/moist mixed-conifer forests in the San Juan Mountains of southwestern Colorado. From mid-June to late August 2011, forest stand structure, tree ages, and fire scars were collected in a series of 33 contiguous stands spanning an area of approximately 1800 ha adjacent to the Williams Creek Reservoir, approximately 25 miles north of Pagosa Springs. The objectives of this research are to: (1) reconstruct specific fire events over the last 300 years including their spatial extent and variation in fire severity; and (2) evaluate how the influences of fire on forest dynamics vary along the continuum from warm/dry to cool/moist mixed-conifer forests.

A total of 1,776 tree cores (average, 54 per stand) and 30 wedges from fire-scarred trees were collected. To date, approximately 3/4 of the tree cores and 2/3 of the fire-scar samples have been processed and cross-dated. Fire-scarred trees were limited primarily to the driest stands, where ponderosa pine is one of the dominant tree species at the lowest elevations in the study area. No fire-scarred samples were found in many of the higher-elevation stands that are dominated by white fir, trembling aspen (Populus tremuloides), and Douglas-fir (Pseudotsuga menziesii). However, the tree-age distributions in these stands tend to be multi-modal, with two or more pulses of establishment by the relatively shade-intolerant aspen or Douglas-fir, which are most likely initiated following a disturbance that killed canopy trees. Where synchronous peaks of establishment are identified across multiple stands, the spatial distribution of these stands will be evaluated in relation to the fire-scar data to determine whether or not these stands may have been burned in the same event. If a series of contiguous stands show a synchronous establishment peak that is associated with fire scars in one or more of the stands, it will be possible to reconstruct spatial variation in the severity of the fire. 

PIRE US-3b: “RM Lower Treeline” Project Plan

Title: Disturbance-mediated effects of climate on regeneration at lower treeline in the US Northern Rockies

Location: Fieldwork in Idaho and Montana; University of Idaho (UIdaho)

Investigators: Higuera

Students: Kemp (PhD UIdaho)

Project Description

Lower treeline in the U.S. Northern Rockies is largely dominated by mixed ponderosa pine (Pinus ponderosa) - Douglas-fir (Pseudotsuga menziesii) forests. In recent decades, mean annual temperatures in these forests have increased ca. 0.8°C, with projections of an additional increase of 2-5°C by 2090 (Mote et al. 2010). Increased temperatures will likely impact tree distributions both directly and indirectly, through disturbance-mediated impacts on recruitment and mortality (Botkin et al. 2007, Littell et al. 2010). For example, bioclimatic modeling suggests that 21st century climate change alone could lead to the contraction of P. ponderosa and P. menziesii “climate space” by 64% and 46%, respectively (Reyfeldt et al. 2006). Widespread fires across the Rocky Mountain West are also predicted to increase with warming (Westerling et al. 2006). While mature trees can persist for decades under suboptimal climate, seedling establishment is more vulnerable to climatic-induced stress; thus, declines in recruitment may precede overstory mortality, especially in recently disturbed landscapes (Littell et al. 2008, Frelich & Reich 2010). 

 This research examines the variability in tree establishment after fire as a function of climate and burn severity, by sampling across climate gradients in space and time. A latitudinal gradient from the Frank Church River of No Return Wilderness (FCW) in central Idaho to Glacier National Park (GNP) in northwestern Montana spans a modern moisture gradient (dry to wet), with local topographically-driven gradients in both regions capturing finer-scale variability. Comparing patterns from the early 21st century to those from the 18th-20th centuries will capture variability in decadal-scale climate, important for both fire severity and post-fire environmental conditions. The research objectives are to: 

1. Quantify post-fire establishment and growth of ponderosa pine and Douglas-fir seedlings from fires occurring between 2000 and 2010 across a moisture gradient from the FCW to GNP.

2. Relate post-fire establishment rates and spatial and temporal variability to burn severity, topography, and microclimate to infer causes of the observed patterns.

3. Compare rates and density of post-fire recruitment after recent fires in ponderosa pine-Douglas-fir ecosystems to historic patterns of regeneration in nearby sites.

 Related Activities

This works is well linked to PIRE-US3a, focusing on similar questions Colorado and the GYE.  Support for this research comes largely from NSF IGERT grant (Univ. Idaho).

Year 2 Update:

This project examines the variability in tree establishment after fire as a function of climate and burn severity, by sampling across climate gradients in space and time. A latitudinal gradient from the Frank Church River of No Return Wilderness (FCW) in central Idaho to Glacier National Park (GNP) in northwestern Montana spans a modern moisture gradient (dry to wet), with local topographically-driven gradients in both regions capturing finer-scale variability. Comparing patterns from the early 21st century to those from the 18th-20th centuries will capture variability in decadal-scale climate, important for both fire severity and post-fire environmental conditions. The research objectives are to: (1) quantify post-fire establishment and growth of ponderosa pine (Pinus ponderosa) and Douglas-fir (Pseudotsuga menziesii) seedlings from fires occurring between 2000 and 2010 across a moisture gradient from the FCW to GNP; (2) relate post-fire establishment rates and spatial and temporal variability to burn severity, topography, and microclimate to infer causes of the observed patterns; and (3) compare rates and density of post-fire recruitment after recent fires in ponderosa pine-Douglas-fir ecosystems to historic patterns of regeneration in nearby sites.

PhD student Kerry Kemp was funded on an NSF IGERT fellowship in 2011-2012.  During this period, she initiated preliminary fieldwork in summer 2011. Over the 2011-2012 academic year she fully developed her field-sampling plan within her PhD proposal, and she started field sampling in late May 2012. From May through August 2012, she will be sampling seedling demography and understory vegetation data from ca. 90 plots spanning climate space occupied by lower treeline in the FCW in south-central Idaho and the Selway-Bitteroot Wilderness in central Idaho and adjacent Montana.  PIRE funds may be directed at this activity in Year 3.