Journal of Biodiversity Management & ForestryISSN: 2327-4417

Reach Us +1 (850) 754 6199

Research Article, J Biodivers Manage Forestry Vol: 4 Issue: 3

Vegetation Cover Change in Yellowstone National Park Detected using Landsat Satellite Image Analysis

Christopher Potter*
NASA Ames Research Center, Mail Stop 232-21, Moffett Field, CA
Corresponding author : Dr. Christopher Potter
NASA Ames Research Center, Mail Stop 232-21, Moffett Field, CA
Tel: 650-604-6164
E-mail: [email protected]
Received: April 16, 2015 Accepted: June 08, 2015 Published: June 11, 2015
Citation: Potter C (2015) Vegetation Cover Change in Yellowstone National Park Detected using Landsat Satellite Image Analysis. J Biodivers Manage Forestry 4:3. doi:10.4172/2327-4417.1000143

Abstract

Vegetation Cover Change in Yellowstone National Park Detected using Landsat Satellite Image Analysis

The northern Rocky Mountains are a region where future climate warming has the potential to alter vegetation cover and surface water runoff. Results from Landsat satellite image analysis since 1987 in all unburned areas (since the 1880s) of Yellowstone National Park (YNP) showed that consistent decreases in the normalized difference vegetation index (NDVI) have been strongly dependent on periodic variations in peak annual snow water equivalents (SWE). The unprecedented decline in SWE over the past 20+ years has had significant impacts on green vegetation cover across unburned ecosystems of YNP, especially during the year 2001, during which the peak SWE levels declined to an historic low of -1.4 standard deviations of the long-term mean SWE. Landsat NDVI analysis did not detect significant upward elevation shifts in unburned vegetation canopy cover across the timberline (above 2900 m elevation) over the past 20+ years in most alpine zones of YNP. The observed dieback of whitebark pine from field observations remains a plausible explanation for lower NDVI detection at and above timberline between the years 1987 and 2010.

Keywords: Landsat; Yellowstone national park; Climate change; Forest vegetation

Keywords

Landsat; Yellowstone national park; Climate change; Forest vegetation

Introduction

Late-20th century snowpack declines have been unprecedented in magnitude across the northern Rocky Mountains region when placed within the context of reconstructions over the past millennium. Snowpack reductions and associated springtime temperature warming may be having important impacts on streamflow and water supplies across the western United States [1].
The northern Rocky Mountains are a semi-arid region, supporting conifer forests at relatively high elevations, where cool temperatures and snow-dominated precipitation can support tree growth [2]. Climate records collected at the northeast entrance of Yellowstone National Park (YNP) have indicated that the growing season (number of days from May through October with lowest temperature above freezing) has been starting earlier and lengthened from an average of 88 days between 1985 and 1996 and to as much as 130 days by 2010 [3].
Prior to the 1990s, the upper timberline in the YNP region was estimated at around 2900 m [4,5]. The lower timberline in the region has been estimated at about 1900 m [6] and is controlled mainly by summer drought stress [7]. Romme and Turner speculated that climate change could move the upper timberline in YNP to 3300 m elevation or higher with projected warming trends over the next century [6]. They further conjectured that increased evapotranspiration, without compensating precipitation inputs or higher water use efficiency by plants, would likely increase drought stress downstream in Yellowstone river drainages.
Projections of continued regional warming over the next 50- 100 years have been associated with predictions for YNP and other protected wilderness areas of the northern Rocky Mountain region of longer fire seasons, larger annual area burned at high severity [2], higher basin-wide evapotranspiration (ET) fluxes from vegetation, and decreased river water flows [1]. These prolonged changes in temperatures and wildfire frequency are not consistent with persistence of the suite of conifer species that have dominated the Yellowstone landscape in modern times, but rather are consistent with lower montane woodlands or non-forest vegetation cover [8]. Water supplies in YNP depend on snow- or glacier-fed surface runoff, such that increases in vegetation ET flux may lead to lower runoff downstream to rivers of the Greater Yellowstone Ecosystem [6].
The effectiveness of Landsat as a tool for monitoring landscapewide changes in the northern Rocky Mountain region has been demonstrated in several previous studies. For example, Jakubauskas and Price [9] used Landsat images to estimate conifer forest age, tree height, and biomass in YNP. Landenburger and Jewett et al. [10,11] used Landsat images for whitebark pine mapping. Geremia et al. [12] used Landsat for grassland forage production mapping on the Northern Range of YNP. Annual vegetation ET flux in montane and sub-alpine vegetation communities of the western U. S. has been closely correlated with the satellite normalized difference vegetation index (NDVI) [13].
The Landsat NDVI has been shown to be a reliable index to monitor large-scale change in green vegetation cover and forest productivity, especially following disturbance in remote mountain areas of the western U. S. [14-21]. Results from Landsat image studies have shown that canopy green leaf cover typically increases rapidly over the first five years following a stand-replacing disturbance, doubling in value by about 10 years after the disturbance, and then leveling off to approach pre-disturbance (mature) stand values by about 25-30 years after the disturbance event [22].
In the present study, Landsat satellite imagery at 30-m ground resolution was analyzed for unburned (since the year 1880) areas of YNP. The long-term Landsat image archive is an important and unique data set to determine if canopy green cover had changed significantly across the national park area since the mid-1980s. The study design controlled for variations in peak snow pack years and elevation gradients. The scope of the study covered YNP, which has been a protected wilderness area since 1872, and a natural laboratory for studies of moisture availability and NDVI trends for the northern Rocky Mountains region over the previous three decades.
The main questions addressed in this study of long-term changes in Landsat NDVI were:
• Has the unprecedented variation in SWE over the past 20+ years had significant impacts on green vegetation cover across unburned ecosystems of YNP?
• Has there been a detectable upward elevation shift in unburned vegetation canopy cover across the upper timberline (i.e., to above 2900 m elevation) over the past 20+ years in YNP?

Study Area Description

YNP is located in Wyoming (96%), Montana (3%), and Idaho (1%) (NW corner coordinates: 45° 15’ N, 111° 12’ W; SE corner coordinates: 44° 5’ N, 109° 49’ W). The NP area of 8980 km2 extends from elevations of 1540 m to 3760 m (Figure 1). Nearby mountain ranges, include the Gallatin Range to the northwest, the Beartooth Mountains in the north, the Absaroka Range to the east, and the Teton Range and the Madison Range to the southwest and west. The Continental Divide of North America runs diagonally through the south section of YNP.
Figure 1: Map of the unburned (since the 1880s) Yellowstone National Park study area, with shaded elevation and Despain vegetation cover types [5]. Unburned land cover color legend: Dark Green – Mature forest, Lighter Green – Younger forest, Yellow – Non-forest, Blue – Water. SNOTEL station locations are marked for the Northeast Entrance (45.01° N, -110.01° W), Madison Plateau, (44.59° N, -111.12° W), and Two Oceans Plateau (44.15° N, -110.22° W).
YNP is drained by three major river watersheds, the Yellowstone (Upper Yellowstone, Lamar, Soda Butte, Slough, and Gardner), the Missouri (Gallatin and Madison) and the Snake River [23]. The Snake River basin lies on the west side of the continental divide where the waters flow to the Pacific Ocean, whereas the Yellowstone and the Madison River basins lie east of the continental divide where they flow to the Gulf of Mexico. The Yellowstone River basin is the largest within YNP and extends from central Wyoming north to include most of southeastern Montana and a small part of western North Dakota.
The climate in YNP is generally cool and dry with mean January and July temperatures of -11.4°C and 10.8°C, respectively, and mean annual precipitation of 56.3 cm [24]. Winters are long and cold, lasting from mid-November to mid-March. Summers are short and often dry, usually lasting from July through August. Average annual snow depth in YNP is 5.4 m [25].
Over the past 25 years of climate records, the relatively dry years of 1987, 1993, 2001, 2005, and 2010 have been identified as extremes by compilation of annual SWE levels across the northern Rocky Mountain region [26]. Measurements of peak SWE levels in late March to early April at U. S. Department of Agriculture Snow Telemetry (SNOTEL) stations [27] within YNP confirm that the years listed above were among those with the greatest negative departures from mean SWE levels recorded from 1987 to 2014 (Figure 2), while 1996 and 2011 showed the greatest positive departures from longterm mean SWE levels (of 60 cm for all three stations). The SNOTEL station at the NE Entrance records consistently lowers SWE (< 50%) each year, compared to other stations within YNP.
Figure 2: Peak yearly SWE at three SNOTEL station locations (shown in Figure 1) within YNP.
The montane forest zone in YNP is found between 1200 and 1800 m, and the subapline forest zone is located between 1800 and 2700 m, approaching timberline [28]. The forests of YNP consist of five main conifer species [29]: lodgepole pine (Pinus contorta), whitebark pine (Pinus albicaulis), Douglas fir (Pseudotsuga menziesii), Engelmann spruce (Picea engelmannii), and subalpine fir (Abies lasiocarpa). Elevation and soil fertility are considered to be the two most important abiotic gradients controlling forest vegetation on the subalpine plateaus [9,30]. Non-forest vegetation is composed of four major cover types: grassland, sagebrush steppe (shrubland), wet sedge and willow meadow, and alpine meadow. The wildfires of 1988 burned over 2500 km2 in YNP and surrounding lands, and created a mosaic of burn severity classes, including light surface burn, severe surface burn, and crown fire [31].
Most of YNP is presently managed as a wilderness area, such that changes in vegetation cover would have been little affected by direct human activities, mainly logging or residential developments. Climate change, wildfire, and insect outbreaks are the main remaining drivers of vegetation cover change in YNP, making it a logical choice for studies of natural moisture availability and NDVI trends in the northern Rocky Mountains region over the previous three decades. By excluding areas burned by wildfire over the previous century in YNP, this study design isolated climate and pathogen/insect effects on patterns of vegetation cover change.

Methods

Image processing
Imagery from the Landsat Thematic Mapper (TM) sensor was selected between the years 1987 to 2010 from the US Geological Survey Earth Explorer data portal (http://earthexplorer.usgs.gov/). TM image data from path/row 38/29 were consistently acquired for an anniversary window between July 17 and August 12 each year, around the peak of the snow-free growing season in YNP [32] to minimize variation caused by seasonal vegetation fluxes and sun angle differences.
All images used in this study were acquired by Landsat TM sensors, geometrically registered using terrain correction algorithms (Level 1T) applied by the U. S. Geological Survey EROS Data Center. The Landsat Surface Reflectance Climate Data Record (CDR) [33] applied corrections to all the images used for top of atmosphere (TOA) reflectance, brightness temperature, and generated masks for clouds, cloud shadows, adjacent clouds, and surface water bodies.
Five cloud-filtered Landsat NDVI data sets from the relatively dry, low SWE years of 1987, 1993, 2001, 2005, and 2010 were processed for time-series comparisons. NDVI (scaled from 0 to 1 for positive values) was computed for all Landsat images as the differential reflectance between the red and near-infrared (NIR) portions of the spectrum by the equation:
NDVI = (NIR – Red) / (NIR + Red)
where NIR is the reflectance of wavelengths from 0.76 to 0.9 μm and Red is the reflectance from 0.63 to 0.69 μm. Advantages of NDVI for vegetation monitoring have been cited in its mathematical simplicity and ease of comparability across numerous multi-spectral remote sensing platforms [34]. Low values of NDVI (near 0) indicate barren land cover whereas high values of NDVI (near 0.9) indicate dense canopy vegetation cover. Negative NDVI values generally indicate water bodies.
Spatial layers
Elevation at 1 arc-second resolution was derived from the United States Geological Survey (USGS) National Elevation Dataset (NED). Slope (in percent) was calculated as the maximum rate of change in elevation value from that cell to its neighbors to determine the steepest downhill descent from the cell. Aspect was calculated by fitting a plane to the z-values of a 3×3 cell neighborhood around each 30-m cell [35]. The direction that the plane faces was set as the aspect for the cell. Aspect was expressed in degrees, moving clockwise from 0 (due north) to 360 (again due north).
Vegetation cover types within YNP were determined from Despain [5] based on color aerial photographs taken from 1969 to 1971. Cover types are based on the dominant tree species and time since the last disturbance (e.g., LP for lodgepole pine, 0 for a young stand = LP0). In addition to LP0 - LP3 cover types, which are most common in YNP, spruce-fir (SF), white bark pine (WB), aspen (AP), Douglas-fir (DF) forest, and non-forest cover types have been delineated. Areas within YNP that have not burned in wildfires over the past century (Figure 1) were delineated from the Fire History Polygons for Northern Rockies 1889-2003 [36] and from the National Monitoring Trends in Burn Severity (MTBS) [37], a multi-agency project designed to consistently map the burn severity and perimeters of fires across all lands of the United States.
Statistical analysis
Tests of statistical significance between NDVI dates were carried out using the two-sample Kolmogorov-Smirnov (K-S) test, a nonparametric method that compares the cumulative distributions of two data sets [38]. The K-S test does not assume that data were sampled from Gaussian distributions (nor any other defined distributions), nor can its results be affected by changing data ranks or by numerical (e.g., logarithm) transformations. The K-S test reports the maximum difference between the two cumulative distributions, and calculates a p value from that difference and the group sample sizes. It tests the null hypothesis that both groups were sampled from populations with identical distributions according to different medians, variances, or outliers. If the K-S p value is small (i.e., <0.05), it can be concluded that the two groups were sampled from populations with significantly different distributions. Within unburned (since 1930) cover types of YNP, 2000 point locations were randomly selected to test the differences between cumulative distributions of NDVI as a function of elevation, slope, aspect, and vegetation cover types.

Results

Attributes of non-burned areas
Areas within YNP that had not burned over the past century and were predominately classified as lodgepole pine successional (LP2), non-forested, and whitebark pine, which together comprised over 75% of all unburned areas within the study area (Table 1). Unburned lodgepole pine patches were estimated to have the largest area (both mean and maximum patch areas) among all the unburned cover classes, and were located at a mean elevation of 2414 m. Unburned non-forested patches were generally less than half the size of unburned lodgepole pine patches and were located at roughly the same mean elevation, but extended to alpine elevations at 3300 m. Unburned whitebark pine patches were generally less than half the size of unburned lodgepole pine patches, but were located at a higher mean elevation of 2756 m and on steeper mean slopes, compared to either unburned lodgepole pine or non-forested patches. Unburned aspen and Douglas fir patches were located at the lowest mean elevations, generally below 2200 m.
Table 1: Attributes of cover classes in YNP that had not burned over the past century [5].
NDVI of non-burned areas
Most cover classes within YNP that had not burned over the past century showed a decline in mean NDVI starting from the relatively dry year (according to SWE deviations in Figure 2) of 1987 to the driest year on record of 2001 (Figure 3). On average, lodgepole pine and Douglas fir classes did not respond as strongly to the increasingly lower SWE levels from 1987 to 1993 and 2001 as did the whitebark pine, aspen and, non-forest classes. All classes gradually recovered in mean NDVI between 2001 and 2005 (and 2010) to approximately 1987 mean NDVI levels.
Figure 3: Mean NDVI for unburned cover classes in YNP from 1987 to 2010, together with standard deviations in peak yearly SWE from the long-term mean 1987 to 2014.
The statistical comparison of the yearly NDVI values sampled from 2000 randomly selected locations within each of the unburned lodgepole pine, non-forested, and whitebark pine cover classes confirmed that NDVI declined significantly between 1987 and 2001 in all three cover types (Table 2). NDVI was significantly higher (p<0.01) in 2010 compared to 2001 for all three dominate cover types. NDVI for lodgepole pine and non-forest cover in 2010 was significantly higher than whitebark pine NDVI in 2010.
Table 2: Kolmogorov-Smirnov test results of maximal difference estimations for the cumulative distributions of NDVI between paired years in the three dominate unburned cover types sampled in YNP (N = 2000 points for each cover class). All paired yearly difference comparison in NDVI cumulative distributions were significant at p < 0.01.
Results for NDVI versus elevation for unburned non-forest cover (scatter plot in Figure 4) showed a consistent non-linear decrease in NDVI (R2>0.3) in all years sampled, whereas there were no significant correlations detected between NDVI and elevation for either unburned lodgepole pine or whitebark pine cover classes (R2<0.05 in all cases). Significant correlation results (at R2>0.3) were not detected between NDVI and percent slope or aspect for any of the unburned cover types sampled.
Figure 4: NDVI versus elevation for unburned non-forest cover (N = 2000, randomly sampled points) in YNP.
Timberline NDVI Change
The upper timberline boundary at around 2900 m [4,5] was delineated within YNP, and NDVI values from 1987 and 2010 were sampled to determine if vegetation green cover in the approximately 537 km2 of unburned area at and above timberline elevations had changed over the 23 years between these NDVI dates. The peak SWE levels from these two years were comparable at 19 cm and 18 cm (1987 and 2010, respectively) at the NE Entrance station (Figure 2).
A total of 541 separate area units located above timberline entirely within the YNP boundary, averaging 84 ha coverage, were delineated for statistical comparison of 20+ years of NDVI change. The majority of these area units were located along the extreme eastern boarder of YNP, and in the northwestern section of the Park at a distance less than 28 km south of Gardiner, Montana. Results showed that average change in NDVI between 1987 and 2010 was significantly different from zero (either positive or negative at p<0.05; Gelman and Hill, 2007) in 99 percent of all these area units However, only onethird of these unburned area units located above timberline showed a significant increase in average NDVI between 1987 and 2010. The overall mean NDVI in 1987 was 0.332 for all area units located above timberline within YNP, whereas the overall mean NDVI in 2010 was 0.314. The maximum negative change in mean NDVI (per unit) was -0.13 and the maximum positive change in mean NDVI was +0.08 above timberline between 1987 and 2010.
Three area units were selected for detailed illustration of NDVI change at the timberline. The first was in the Dome Mountain area (44.836° N, -110.847° W) in the northwestern section of YNP (Figure 5a) and the second was around Mount Langford (44.387° N, -110.121° W) 8 km east of the southeast arm of Yellowstone Lake (Figure 5b). Both of these illustrations, representing the majority of areas units above timberline, showed that NDVI was slightly higher in 2010 than in 1987 at relatively lower elevations (less than 2600 m), but that NDVI did not increase at elevations near and above timberline in the 23 years between these two dates. In the case of the Dome Mountain area, NDVI declined markedly at elevations near and above timberline in 2010 compared to 1987.
Figure 5: Changes in NDVI at the upper timberline boundary (2900 m elevation) in selected unburned units of YNP from 1987 and 2010. Topographic maps show the transition boundaries from forest cover (green) to non-forest cover (yellow) from Depsain [5]. Cloud cover is masked as white.
The third area was selected for illustration of NDVI change above timberline to represent the minority of units wherein average NDVI did increase significantly (p<0.05) at elevations near and above timberline over the past 20+ years. This 15 ha whitebark pine dominated unit (Figure 5c) was located in the southeastern corner of YNP near Trident Plateau above the Escapement Creek drainage (at 44.157° N, -110.044° W). In the timberline transition zone to 50 m above 2900 m elevation, NDVI increased from levels of around 0.3 in 1987 to around 0.5 in 2010.
Specific areas of interest
Soda Butte Creek drainage: Soda Butte Creek is located near the northeast entrance of YNP. The Creek flows for a length of approximately 30 kilometers from its headwaters near Cooke City, Montana until it empties into the Lamar River. Forest cover in the drainage basin is predominantly late successional lodgepole pine mixed with stands of subalpine fir, Douglas-fir and Engelmann spruce. The basin area including Pebble Creek bordering the northern portion of the drainage (Figure 6a), centered at around 2310 m (between Barronette and Abiathar Peaks), was unburned over the past century. From 1987 to 2001, the creek side forest stands at around 2210 m elevation declined in NDVI most notably with lower SWE levels in 2001, whereas the forested slopes above Soda Butte Creek between 2300 and 2400 m elevation maintained relatively high NDVI (>0.5) in all years sampled from 1987 to 2010.
Figure 6: Changes in NDVI in unburned areas of YNP from 1987 to 2001 and 2010. NDVI color legend as shown in Figure 5. Cloud cover is masked as white.
Lamar river valley: The Lamar River is a tributary of the Yellowstone River, approximately 48 km long, located entirely within YNP. The main channel is joined by many tributary streams, including Soda Butte Creek and Slough Creek, and empties into the Yellowstone River near Tower Junction. Grassland and sagebrush vegetation types dominate the Valley on most of the northern winter range. The area of the Lamar Valley centered at around 2180 m elevation (south of Bison Peak around 44.92° N, -110.25° W) was unburned over the past century. Ridgelines above 2300 m elevation showed consistently lower NDVI in all years sampled from 1987 to 2010 (Figure 6b), compared to valley and creek bottoms at lower elevations, which were the only portions of the landscape that maintained relatively high NDVI (> 0.5), and presumably good forage production, in the relatively lower SWE year of 2001. Over most of the Lamar Valley grasslands, NDVI in 2001 was below 0.35, compared to many of the same areas in 2010 when NDVI was observed at nearly twice that level.
Gardiner basin: The Gardiner River Basin is located in northwestern section of YNP consists of a variety of sagebrush habitats including the Wyoming big sagebrush (Artemisia tridentata wyomingensis) and blue bunch wheatgrass (Agropyron spicatum) that is found primarily at lower elevations in the basin. Mountain big sagebrush (A. t. vaseyana) - Idaho fescue (Festuca idahoensis) habitat type dominates the majority of Gardiner Basin. The landscape has relatively mild winter conditions compared to rangelands further inside YNP, which affords reliable winter foraging for ungulates. The area just east of Quadrant Mountain (at 44.91° N, -110.80° W) along Panther Creek was unburned over the past century. Vegetation cover in lower elevation (2500 m) sagebrush shrublands declined most notably from 1987 to 2001 (Figure 6c), with NDVI changing from levels above 0.5 to below 0.35, and but recovered in 2010 to near 1987 NDVI levels.
Mallard lake near Old Faithful: The Mallard Lake area 5 km east of Old Faithful at around 2585 m elevation (44.45° N, -110.75° W) is an unburned stand of middle-successional lodgepole pine. The forest cover in this area had few stands with dense tree cover, indicated by the majority of NDVI levels lower than 0.5 in all years sampled (Figure 6d). The entire lodgepole pine forest declined to NDVI levels around 0.3 in 2001, but recovered in 2010 to 1987 NDVI levels, particularly in the wetter sections along Spring Creek and other small streams in the drainage.

Discussion

The results of this study, derived from more than 20 years of Landsat image analysis, support the conclusion that the detectable changes in NDVI over time in YNP have been strongly dependent on periodic variations in peak annual SWE. The unprecedented decline in SWE over the past 25 years has had significant impacts on green vegetation cover across unburned ecosystems of YNP, especially during the year 2001, when the peak SWE levels declined to an historic low of -1.4 standard deviations of the long-term mean SWE across YNP.
Landsat NDVI analysis did not detect an upward elevation shift in unburned vegetation canopy cover across the timberline (above 2900 m elevation) over the past 20+ years in most alpine zones of YNP. One reason for this observation is that whitebark pine, which retains snow and reduces erosion at high elevations while producing seeds that are an important food source for grizzly bears and other wildlife, is experiencing unprecedented mortality throughout its natural range [3]. A primary cause of this mortality is blister rust (Cronartium ribicola), an introduced pathogen that increases whitebark pine vulnerability to infestation by mountain pine beetle (Dendroctonus ponderosae) [39]. Approximately 20 percent of nearly 4,800 live trees were infected by whitebark blister rust when first surveyed by the Greater Yellowstone Inventory and Monitoring Network from 2004 to 2007. When trees were resurveyed from 2008 to 2010, 16 percent had suffered mortality [3].
Declines in NDVI with dieback of whitebark pines are more likely to be detected during years of low moisture availably [40] than they would be under more typical SWE levels of the past across YNP [1]. Climate change models have predicted average annual surface temperatures in the northern Rocky Mountain region to increase by 3.5°C by 2100, with the greatest levels of warming in the winter months [41]. This level of warming would likely promote widespread mountain pine beetle outbreaks that could eliminate whitebark pine from the region, with serious implications for grizzly bear food sources [11,39].
Warmer winters, thinner snowpacks, and expanded non-forest areas at subalpine elevations could alter grazing ungulate populations in YNP [6] and further complicate wildlife management planning in the region. Large numbers of ungulates commonly migrate out of YNP every winter [12], where they are subjected to hunting and exposure to domestic livestock herds [42]. If drier conditions and lower SWE years frequently depress herbaceous forage production on the Northern Range of YNP, then degradation of lower elevation grassland habitats (e.g., in the Lamar Valley) may promote an expanded winter range upward in elevation for grazing ungulate populations.
Based on the close relationship measured between NDVI and ET fluxes by vegetation, Goulden and Bales [13] concluded that climate warming predicted for the year 2100 in would “thicken” vegetation cover in major river basins of the Sierra Nevada mountains of California and increase basin-wide ET fluxes by 28 percent and decrease water flow downstream by more than 25 percent. In contrast, the findings from this Landsat image analysis in YNP suggest that NDVI has not increased significantly throughout the unburned forests of Park over the past 20+ years, and conflict with any assumptions of higher annual ET fluxes affecting river flows downstream in this protected area [6]. Forest pathogen and insect outbreaks are well-documented controls on tree physiology in the region. The current dynamics of forest cover change over most of the YNP area do not appear to be driving ET increases region-wide. Instead, continued trends of low snowfall and early snowmelt could alter water flows downstream on an unprecedented scale across the northern Rocky Mountain region.

References

  1. Pederson GT, Gray ST, Ault T, Marsh W, Fagre DB, et al. (2010) Climatic controls on the snowmelt hydrology of the northern Rocky Mountains. J Climate 24: 1666-1687.

  2. Westerling AL, Hidalgo HG, Cayan DR, Swetnam TW (2006) Warming and earlier spring increase western U.S. forest wildfire activity. Science 313: 940-943.

  3. Yellowstone Center for Resources (YCR) (2011) Natural Resource Vital Signs, 2011. Yellowstone Center for Resources, Yellowstone National Park, Mammoth Hot Springs, Wyoming, USA.

  4. Patten DT (1963) Vegetational pattern in relation to environments in the Madison Range, Montana. Ecological Monogr 33: 375-406.

  5. Despain D (1990) Yellowstone's vegetation: consequences of history and environment. Roberts Rinehart Inc. Publishing, Boulder, Colorado, USA.

  6. Romme WH, Turner MG (1991) Implications of global climate change for biodiversity in the Greater Yellowstone Ecosystem. Conserv Biol 5: 373-386.

  7. Krebs C (1985) Ecology: the experimental analysis of distribution and abundance. (3rd edtn) Harper & Row, New York, USA.

  8. Littell JS, McKenzie D, Peterson DL, Westerling AL (2009) Climate and wildfire area burned in western U.S. ecoprovinces, 1916-2003. Ecol Appl 19: 1003-1021.

  9. Jakubauskas ME, Price K (2000) Regression-based estimation of lodgepole pine forest age from Landsat Thematic Mapper data. Geocarto International 15: 21-26.

  10. Landenburger L (2008) Mapping regional distribution of a single tree species: Whitebark pine in the Greater Yellowstone Ecosystem. Sensors 8: 4983-4994.

  11. Jewett JT, Lawrence RL, Marshall LA, Gessler PE, Powell SL, et al. (2011) Spatiotemporal relationships between climate and whitebark pine mortality in the Greater Yellowstone Ecosystem. Forest Science 57: 320-335.

  12. Geremia CPI, White RL, Wallen FGR, Watson JJ, Treanor J, et al. (2011) Predicting bison migration out of Yellowstone National Park using Bayesian models. PLoSOne 6: e16848.

  13. Goulden ML, Bales RC (2014) Mountain runoff vulnerability to increased evapotranspiration with vegetation expansion, Proc Natl Acad Sci USA 111: 14071-14075.

  14. Collins JB, Woodcock CE (1996) An assessment of several linear change detection techniques for mapping forest mortality using multitemporal Landsat TM data. Remote Sens Environ 56: 66-77.

  15. Amiro BD, Chen JM, Liu J (2000) Net primary productivity following forest fire for Canadian ecoregions. Canadian Journal of Forest Research 30: 939-947.

  16. Rogan J, Franklin J (2001) Mapping wildfire burn severity in southern California forests and shrublands using Enhanced Thematic Mapper imagery. Geocarto International, 16: 89-99.

  17. Rogan J, Miller J, Stow DA, Franklin J, Levien L, et al. (2003) Land-cover change monitoring with classification trees Using Landsat TM and ancillary data. Photogramm Eng Rem S 69: 793-804.

  18. Fischer L, Rosenberg M, Mahon L, Liu Z, Maurizi B, et al. (2004) Monitoring land cover changes in California, a USFS and CDF cooperative program, Northern Sierra Project Area - Cycle II. State of California, Resources Agency, Department of Forestry and Fire Protection, Sacramento, CA, USA.

  19. Epting J, Verbyla DL (2005) Landscape level interactions of pre-fire vegetation, burn severity, and post-fire vegetation over a 16-year period in interior Alaska. Can J Forest Res 35: 1367-1377.

  20. Cuevas-Gonzalez M, Gerard F, Balzter H, Riano D (2009) Analysing forest recovery after wildfire disturbance in boreal Siberia using remotely sensed vegetation indices. Global Change Biology 15: 561-577.

  21. Casady GM, Marsh SE (2010) Broad-scale environmental conditions responsible for post-fire vegetation dynamics. Remote Sens 2: 2643-2664.

  22. Potter C (2014) Ten years of forest cover change in the Sierra Nevada detected using Landsat satellite image analysis. Int J Remote Sens 35: 7136-7153.

  23. Yellowstone National Park (2012) Yellowstone Resources and Issues Handbook 2012, U.S. Department of the Interior, National Park Service, Yellowstone National Park, Wyoming, USA.

  24. Dirks RA, Martner BE (1982) The climate of Yellowstone and Grand Teton National Parks. Occasional Paper No. 6, U. S. National Park Service, Washington DC, USA.

  25. Huang S, Rich P, Crabtree R, Potter C, Fu P (2008) Modelling near-surface air temperature from solar radiation and lapse rate: application over complex terrain in Yellowstone National Park, USA. Physical Geography 29: 158-178.

  26. Pederson GT, Gray ST, Woodhouse CA, Betancourt JL, Fagre DB, et al. (2011) The unusual nature of recent snowpack declines in the North American Cordillera. Science 333: 332-335.

  27. Schaefer GL, Werner J (1996) SNOTEL into the Year 2000. American Meteorological Society Meeting, 12th Conference on Biometeorology and Aerobilogy, Atlanta, Georgia.

  28. Habeck JR (1987) Present-day vegetation in the northern Rocky Mountains. Annals of the Missouri Botanical Garden. 74: 804-840.

  29. Kokaly RF, Despain DG, Clark RN, Livo KE (2003) Mapping vegetation in Yellowstone National Park using spectral feature analysis of AVIRIS data. Remote Sens Environ 84: 437-456.

  30. Christensen NL, Agee JK, Brussard PF, Hughes J, Knight DH, et al. (1989) Interpreting the Yellowstone fires of 1988. BioScience 39: 678-85.

  31. Crabtree R, Potter C, Mullen R, Sheldon J, Huang J, Harmsen A, et al. (2009) A modeling and spatio-temporal analysis framework for monitoring environmental change using NPP as an ecosystem indicator. Remote Sens Environ 113: 1486-1496.

  32. Potter CS, Klooster S, Crabtree R, Huang S, Gross P, et al. (2011) Carbon fluxes in ecosystems of Yellowstone National Park predicted from remote sensing data and simulation modeling. Carbon Balance and Management 6: 3.

  33. Masek JG, Vermote EF, Saleous N, Wolfe R, Hall FG, et al. (2006) A Landsat surface reflectance data set for North America, 1990-2000, Geoscience and Remote Sensing Letters 3: 68-72.

  34. Lentile L, Holden A, Smith A, Falkowski M, Hudak A, et al. (2006) Remote sensing techniques to assess active fire characteristics and post-fire effects. Int J Wildland Fire 15: 319-345.

  35. Burrough PA, McDonell RA (1998) Principles of Geographical Information Systems. Oxford University Press, New York, USA.

  36. Gibson CE (2006) A northern Rocky Mountain polygon fire history: accuracy, limitations, strengths and recommended protocol of digital fire perimeter data. Thesis. University of Idaho, Moscow, Idaho, USA.

  37. Eidenshenk J, Schwind B, Brewer K, Zhu Z, Quayle B, et al. (2007) A project for monitoring trends in burn severity. Fire Ecology 3: 3-21.

  38. Lehmann E (2006) Nonparametrics: Statistical Methods Based on Ranks. Springer, New York, USA.

  39. Schrag AM, Bunn AG, Graumlich LJ (2007) Influence of bioclimatic variables on treeline conifer distribution in the Greater Yellowstone Ecosystem: implications for species of conservation concern. J Biogeogr 35: 698-710.

  40. Christensen JH, Hewitson B, Busuioc A, Chen X, Gao I, et al. (2007) Regional climate projections. Climate change 2007: The physical science basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK.

  41. Logan JA, Macfarlane WW, Willcox L (2010) Whitebark pine vulnerability to climate-driven mountain pine beetle disturbance in the Greater Yellowstone Ecosystem. Ecol Appl 20: 895-902.

  42. Berger J (1991) Greater Yellowstone's native ungulates: myths and realities. Conservation Biology 5: 353-363.

Track Your Manuscript

Share This Page