Sequoia-Kings Canyon National Park Research Articles

Fire return intervals explain different vegetation cover responses to wildfire restoration in two Sierra Nevada basins

A simple dynamical model was used to explore the forest cover dynamics for two basins in the Sierra Nevada of California, Illilouette Creek Basin (ICB) in Yosemite National Park and Sugarloaf Creek Basin (SCB) in Sequoia-Kings Canyon National Park. Since the 1970s, fire management in these basins has attempted to restore a near-natural fire regime, after nearly a century of fire exclusion and suppression. The model describes two canopy layers, representing mixed conifer and shrub-dominated landcover types, and is calibrated using landcover maps and fire recurrence and severity data from the ICB. The calibrated model is used to explore several scenarios pertaining to increasing fire severity and return interval in the ICB, and to explore the differences between the ICB and SCB. The results indicate that (i) the ICB in 2012 had not yet reached steady state forest cover, (ii) potential future changes in fire severity and frequency will yield reductions in forest cover, and (iii) differences in forest cover change in response to fire regime restoration between basins can be explained by differences in fire histories.

Monitoring of Sierra Nevada Caves Reveals the Potential for Stalagmites to Archive Seasonal Variability

In the southwestern United States, California (CA) is one of the most climatically sensitive regions given its low (≤250 mm/year) seasonal precipitation and its inherently variable hydroclimate, subject to large magnitude modulation. To reconstruct past climate change in CA, cave calcite deposits (stalagmites) have been utilized as an archive for environmentally sensitive proxies, such as stable isotope compositions (δ18O, δ13C) and trace element concentrations (e.g., Mg, Ba, Sr). Monitoring the cave and associated surface environments, the chemical evolution of cave drip-water, the calcite precipitated from the drip-water, and the response of these systems to seasonal variability in precipitation and temperature is imperative for interpreting stalagmite proxies. Here we present monitored drip-water and physical parameters at Lilburn Cave, Sequoia Kings Canyon National Park (Southern Sierra Nevada), CA, and measured trace element concentrations (Mg, Sr, Ba, Cu, Fe, Mn) and stable isotopic compositions (δ18O, δ2H) of drip-water and for calcite (δ18O) precipitated on glass substrates over a two-year period (November 2018 to February 2021) to better understand how chemical variability at this site is influenced by local and regional precipitation and temperature variability. Despite large variability in surface temperatures and precipitation amount and source region (North Pacific vs. subtropical Pacific), Lilburn Cave exhibits a constant cave environment year-round. At two of the three sites within the cave, drip-water δ18O and δ2H are influenced seasonally by evaporative enrichment. At a third collection site in the cave, the drip-water δ18O responds solely to precipitation δ18O variability. The Mg/Ca, Ba/Ca, and Sr/Ca ratios are seasonally responsive to prior calcite precipitation at all sites but minimally to water-rock interaction. Lastly, we examine the potential of trace metals (e.g., Mn2+ and Cu2+as a geochemical proxy of recharge and find that variability in their concentrations has high potential to denote the onset of the rainy season in the study region. The drip-water composition is recorded in the calcite, demonstrating that stalagmites from Lilburn Cave, and potentially more regionally, could record seasonal variability in weather even during periods of substantially reduced rainfall.

Assessing trends and vulnerabilities in the mutualism between whitebark pine (Pinus albicaulis) and Clark's nutcracker (Nucifraga columbiana) in national parks of the Sierra-Cascade region.

Dispersal of whitebark pine (Pinus albicaulis Engelm.), a keystone species of many high-elevation ecosystems in western North America, depends on Clark's nutcracker (Nucifraga columbiana Wilson), a seed-caching bird with an affinity for whitebark seeds. To the extent that this dependence is mutual, declines in whitebark seed production could cause declines in nutcracker abundance. Whitebark pine is in decline across much of its range due to interacting stressors, including the non-native pathogen white pine blister rust (Cronartium ribicola J. C. Fisch.). We used avian point-count data and tree surveys from four national park units to investigate whether trends in whitebark pine can explain trends in Clark's nutcracker. Spatial trends were modeled using recent data from two parks, while temporal trends were modeled using longer time-series of nutcracker and whitebark data from two additional parks. To assess the potential dependence of nutcrackers on whitebark, we linked a model of nutcracker density (accounting for detection probability) with a model of whitebark trends, using a Bayesian framework to translate uncertainty in whitebark metrics to uncertainty in nutcracker density. In Mount Rainier National Park, temporal models showed dramatic declines in nutcracker density concurrent with significant increases in whitebark crown mortality and trees infected with white pine blister rust. However, nutcrackers did not trend with whitebark metrics in North Cascades National Park Service Complex. In spatial models of data from Yosemite National Park and Sequoia-Kings Canyon National Park, nutcracker density varied not only with local cover of whitebark but also with elevation and, in Sequoia-Kings Canyon, with cover of another species of white pine. Our results add support for the hypothesis that the mutualism between whitebark pine and Clark's nutcracker is vulnerable to disruption by blister rust, and our approach integrates data across monitoring programs to explore trends in species interactions.

Forest Vegetation Change and Its Impacts on Soil Water Following 47 Years of Managed Wildfire

Managed wildfire is an increasingly relevant management option to restore variability in vegetation structure within fire-suppressed montane forests in western North America. Managed wildfire often reduces tree cover and density, potentially leading to increases in soil moisture availability, water storage in soils and groundwater, and streamflow. However, the potential hydrologic impacts of managed wildfire in montane watersheds remain uncertain and are likely context dependent. Here, we characterize the response of vegetation and soil moisture to 47 years (1971–2018) of managed wildfire in Sugarloaf Creek Basin (SCB) in Sequoia-Kings Canyon National Park in the Sierra Nevada, California, USA, using repeat plot measurements, remote sensing of vegetation, and a combination of continuous in situ and episodic spatially distributed soil moisture measurements. We find that, by comparison to a nearby watershed with higher vegetation productivity and greater fire frequency, the managed wildfire regime at SCB caused relatively little change in dominant vegetation over the 47 year period and relatively little response of soil moisture. Fire occurrence was limited to drier mixed-conifer sites; fire-caused overstory tree mortality patches were generally less than 10 ha, and fires had little effect on removing mid- and lower strata trees. Few dense meadow areas were created by fire, with most forest conversion leading to sparse meadow and shrub areas, which had similar soil moisture profiles to nearby mixed-conifer vegetation. Future fires in SCB could be managed to encourage greater tree mortality adjacent to wetlands to increase soil moisture, although the potential hydrologic benefits of the program in drier basins such as this one may be limited.

Using environmental DNA for early detection of amphibian chytrid fungus Batrachochytrium dendrobatidis prior to a ranid die-off.

Amphibian chytridiomycosis caused by the fungus Batrachochytrium dendrobatidis (Bd) is an emerging infectious disease that has been associated with mass mortality and extinctions of amphibians worldwide. Environmental DNA (eDNA) techniques have been used to detect the presence of Bd in the environment, but not to detect Bd prior to an amphibian die-off. We collected eDNA using filtered water samples from 13 lakes across Sequoia Kings Canyon National Park. Seven of those sites had populations of mountain yellow-legged frogs, an amphibian highly susceptible to chytridiomycosis, and 3 of those populations experienced a Bd related die-off 1 mo post-eDNA sampling. We detected Bd in eDNA samples that were collected 1 mo prior to the observed Bd-caused die-off at all 3 sites affected by Bd, and we did not detect Bd at the other sites where no die-off was observed. Our study indicates the potential to use eDNA techniques for early detection of Bd in the environment.

Marginal Bayesian nonparametric model for time to disease arrival of threatened amphibian populations.

The global emergence of Batrachochytrium dendrobatidis (Bd) has caused the extinction of hundreds of amphibian species worldwide. It has become increasingly important to be able to precisely predict time to Bd arrival in a population. The data analyzed herein present a unique challenge in terms of modeling because there is a strong spatial component to Bd arrival time and the traditional proportional hazards assumption is grossly violated. To address these concerns, we develop a novel marginal Bayesian nonparametric survival model for spatially correlated right-censored data. This class of models assumes that the logarithm of survival times marginally follow a mixture of normal densities with a linear-dependent Dirichlet process prior as the random mixing measure, and their joint distribution is induced by a Gaussian copula model with a spatial correlation structure. To invert high-dimensional spatial correlation matrices, we adopt a full-scale approximation that can capture both large- and small-scale spatial dependence. An efficient Markov chain Monte Carlo algorithm with delayed rejection is proposed for posterior computation, and an R package spBayesSurv is provided to fit the model. This approach is first evaluated through simulations, then applied to threatened frog populations in Sequoia-Kings Canyon National Park.

Pigment‐based Identification of Ozone‐Damaged Pine Needles as a Basis for Spectral Segregation of Needle Conditions

Air pollution affects large areas of forest, and field assessment of these effects is a costly, site-specific process. This paper establishes a biochemical basis for identifying ozone-damaged pine trees to facilitate efficient remote sensing assessment of air pollution damage. Several thousand live needles were collected from ponderosa pine (Pinus ponderosa) and Jeffrey pine (P. jeffreyi) trees at three sites in Plumas National Forest and Sequoia-Kings Canyon National Park. These needles were assembled into 504 samples (based on the abaxial surface) and grouped according to five dominant needle conditions (green, winter fleck, sucking insect damage, scale insect damage, and ozone damage) and a random mixture of needles. Pigment concentrations per unit needle area of chlorophyll a, chlorophyll b, and total carotenoids were measured. The following pigment concentration ratios were calculated for all samples: chlorophyll a/total carotenoids, chlorophyll b/total carotenoids, total chlorophyll/carotenoids, chlorophyll a/chlorophyll b. The group of ozone-damaged needles had significantly lower mean pigment concentrations (family-wise p < 0.01) and significantly lower mean chlorophyll a/total carotenoid and total chlorophyll/total carotenoid ratios (family-wise p < 0.01) than all other groups of needles. Ozone-damaged needles had a significantly lower mean chlorophyll a/chlorophyll b ratio than all other groups except one (family-wise p < 0.01). Linear discriminant analysis with three factors (chlorophyll a concentration, the chlorophyll a/carotenoid ratio, and the chlorophyll a/chlorophyll b ratio) and subsequent maximum likelihood classification of damaged and non-damaged needles gave an overall cross-validated accuracy of 96%. These ozone-damaged needles are biochemically unique in relation to other needle conditions in this study, and further research is needed to generalize these results.

Imprint of oaks on nitrogen availability and δ15N in California grassland-savanna: a case of enhanced N inputs?

Woody vegetation is distributed patchily in many arid and semi-arid ecosystems, where it is often associated with elevated nitrogen (N) pools and availability in islands of fertility. We measured N availability and δ15N in paired blue-oak versus annual grass dominated patches to characterize the causes and consequences of spatial variation in N dynamics of grassland-savanna in Sequoia-Kings Canyon National Park. We found significantly greater surface soil N pools (0–20 cm) in oak patches compared to adjacent grass areas across a 700 m elevation gradient from foothills to the savanna-forest boundary. N accumulation under oaks was associated with a 0.6‰ depletion in soil δ15N relative to grass patches. Results from a simple δ15N mass balance simulation model, constrained by surface soil N and δ15N measured in the field, suggest that the development of islands of N fertility under oaks can be traced primarily to enhanced N inputs. Net N mineralization and percent nitrification in laboratory incubations were consistently higher under oaks across a range of experimental soil moisture regimes, suggesting a scenario whereby greater N inputs to oak patches result in net N accumulation and enhanced N cycling, with a potential for greater nitrate loss as well. N concentrations of three common herbaceous annual plants were nearly 50% greater under oak than in adjacent grass patches, with community composition shifted towards more N-demanding species under oaks. We find that oaks imprint distinct N-rich islands of fertility that foster local feedback between soil N cycling, plant N uptake, and herbaceous community composition. Such patch-scale differences in N inputs and plant–soil interactions increase biogeochemical heterogeneity in grassland-savanna ecosystems and may shape watershed-level responses to chronic N deposition.

SCALE-SPECIFIC INFERENCE USING WAVELETS

Understanding of spatial pattern and scale has been identified as a key issue in ecology, yet ecology has traditionally lacked necessary tools for making inference about relationships between scale-specific patterns. We introduce wavelet-coefficient regression, in which the dependent and independent variables are wavelet transformed prior to analysis, as a means to formalize scale-specific relationships in ecological data. We apply this method to data on vegetation and environmental factors related to water availability from Sequoia-Kings Canyon National Park (California, USA). We find that the wavelet transform and wavelet-coefficient regression efficiently characterize scale-specific pattern in these data. We also find that different environmental factors show up as good predictors of vegetation growth at different scales and that these differences in scale greatly facilitate interpretation of the mechanisms relating water availability to vegetation growth.

Response by trout populations in alpine lakes to an experimental halt to stocking

Trout are often stocked into alpine lakes based on the generally untested assumption that resident populations would go extinct without stocking. The objectives of our study were to estimate the proportion of currently or formerly stocked alpine lakes in the Sierra Nevada, California, containing naturally reproducing trout populations (Oncorhynchus mykiss, Oncorhynchus mykiss aguabonita, Oncorhynchus clarki henshawi), identify characteristics of lakes associated with successful reproduction, and quantify the effects of stocking termination on trout density and individual growth rates in reproducing populations. We surveyed trout populations in 95 lakes in the John Muir Wilderness before and after a 4- to 8-year stocking halt and in 84 lakes in Sequoia–Kings Canyon National Park after a ≥20-year stocking hiatus. Based on recruitment during the no-stocking period, 70% and 68% of study lakes in the John Muir Wilderness and Sequoia-Kings Canyon National Park, respectively, contained reproducing populations. Results indicated that lakes with &gt;2.1 m2 of spawning habitat and at elevations &lt;3520 m nearly always showed evidence of reproduction. For reproducing populations, stocking termination did not result in significant changes in population density, but may have increased individual growth rates. We conclude that most trout stocking in Sierra Nevada alpine lakes could be permanently halted with only minimal impact on the recreational fishery.

Extending community ecology to landscapes

A goal of landscape ecology is to infer processes or constraints that generate spatial pattern in communities and ecosystems. The rich tradition of plant community ecology is now being extended to address spatial pattern in vegetation over large spatial extents. The challenge in this is that vegetation pattern on landscapes is fine-grained, which presents sampling problems for large study areas. Further, spatial autocorrelation in ecological data, coupled with strong patterns of correlation among environmental factors (such as the gradient complexes governed by elevation) make it difficult to make clear inferences about the agents patterning landscape-scale vegetation. Here we review the methods of plant community ecology as extended to landscapes and illustrate the challenges with a case study from Sequoia-Kings Canyon National Park in California’s southern Sierra Nevada. We outline an iterative approach to such studies, with three stages. The first stage is a pilot study to characterize the spatial scaling of environmental factors presumed to be important to vegetation; this stage can often be conducted virtually, using digital terrain data. The second stage is iterative and consists of building a preliminary explanatory model using a combination of ordination, classification, and Mantel tests: all analyses based on the same ecological distance or dissimilarity matrices. This preliminary model is then attacked to find its uncertain or sensitive parts, and these parametric conditions are mapped into geographic space to identify candidate sites for follow-up field studies in the third stage. This approach ensures that the most uncertain aspects of the preliminary model are refined in an efficient manner. As the approach proceeds toward a richer understanding of species-environment relationships and vegetation pattern, a need emerges for new kinds of field studies and novel extensions to existing statistical analyses. We discuss possible extensions of these as a natural consequence of this iterative process of model construction and revision.

Declining Amphibian Populations

W T tORLDWIDE DECLINES IN AMPHIBIAN POPULATIONS ATtracted wide attention following a workshop sponsored by the National Research Council (NRC) (1, 2). The declines seem to be general, but some regions and many taxa are apparently unaffected (1). Are the declines due to normal climatic fluctuations, which produce droughts and frosts that can severely reduce population size locally, or are there other, more insidious causes? The report of census data for four amphibians (three salamanders and a frog) gathered during a 12-year period of continuous (daily) monitoring at the Savannah River Site in South Carolina contributes importantly to this question (3). Substantial variation was found in breeding population sizes and juvenile recruitment. Population declines were found in some years, but these were coupled with increases; there is a complex dynamic associated with many factors, especially local moisture regimen. The new data, the most extensive available, raise the possibility that declines are the coincidental effect of population fluctuations. If so, the consequences will vary greatly from place to place. The southeastern United States is a veritable carpet of amphibians, from the abundant salamanders of the highest Appalachian peaks to dense lowland populations of both frogs and salamanders. Should populations decline to local extinction in this area, chances of recovery are high. However, local extinctions have more profound implications in other parts of the world, where species are specialists for habitats that are localized or badly fragmented; under such circumstances, opportunities for recolonization are low to nonexistent (4). It is significant that participants at the NRC workshop, who heard a brief summary of the new data, considered it unlikely that global declines were coincidental, because of the number and widespread distribution of the reports, but population fluctuations clearly must be taken into account. Concern about declines arises because amphibians are abundant, integral components ofmany diverse ecosystems. They are local top carnivores that are major consumers of invertebrates, especially insects. General declines would have widespread consequences and might indicate more general environmental problems (5). Some declines have been dramatic. Ranid frogs have all but disappeared from southern California (6). By the late 1980s a montane frog (Rana muscosa) had disappeared from 98% of the ponds in which it had been studied in the mid-1970s in SequoiaKings Canyon National Park (1, 4). In Oregon, populations ofRana cascadae, monitored since the mid-1970s, have suffered about an 80% disappearance (1, 7). The golden toad (Bufo periglenes), endemic to the Monteverde Cloud Forest Preserve in Costa Rica, has not bred in its traditional breeding sites since 1987; many other frog taxa in the region have experienced declines and local disappearance during this period (1, 8). The gastric-brooding frog (Rheobatrachus silus), which lived in rivers in relatively undisturbed regions of Queensland, Australia, has not been seen since 1979, and several other sympatric species offrogs also are thought to be extinct (9). Sometime after 1981, 8 of 13 species of frogs that had been present in Reserva Atlantica, Brazil, disappeared, including unusual diurnal hylodine frogs, whose birdlike calls were distinctive features

Effects of Oxidant Air-pollution on Needle Health and Annual-ring Width in a Ponderosa Pine Forest

More than 485,000 hectares of the Ponderosa Pine forests in the southern Sierra Nevada of California are diseased as a result of air pollution, while approximately 25% of the area of Sequoia-Kings Canyon National Park receives enough oxidant air-pollution to elicit symptoms of smog injury on Ponderosa Pine, the dominant mid-elevation tree.In 1983 we re-examined four permanent study plots that had been established in 1974–75. Smog symptoms on current needles in 1983 occurred on 23.8% of the trees compared with 14.5% only in 1975. Second-year needles had 60.7% showing symptoms compared with 44.2% in 1975. Less than 0.5% of the trees retained 6-years-old needles, and 21% had only 2 years of needle retention. Mortality attributed to air pollution was observed.