B51A-0039
Plant phenology patterns at three sites on the H.J. Andrews Experimental Forest, Oregon, 1987 to 2007.
Plant phenology data has been collected at three sites on the H.J. Andrews Experimental Forest, Oregon since the late 1970's. The sites were visited once every three weeks year-round. Current efforts to clean and archive this data are on-going. Here we present on a 20 year data set from 1987 to 2007. The three sites are located at Watersheds (WS) 10, 8, and 7/6, at an elevation of 466 m, 993 m, and 905/950 m respectively. Forests were old growth (WS 8) and regenerating clearcuts (WS 6, 10), or shelterwood clearcut with overstory removed in 1984 (WS 7) dominated by Douglas-fir, western hemlock, true firs, and western redcedar. One tree (Douglas-fir), two evergreen shrubs, three deciduous shrubs, and four herbs were followed for vegetative and flowering phenology, including bud swell, bud bread, leaf expansion, leaf color change, leaf fall, flower bud swell, blooming, petal loss, fruit formation, and seed dispersal. Weather stations are located at each site and tied into a network of stations in this LTER site. Physical factors such as snow depth, snow coverage, freeze-thaw activity, as well as lichen condition were also noted. We are asking two key questions regarding plant phenology patterns. 1. Has the growing season lengthened in the mountainous watershed of the HJ Andrews, and is this similar for a low elevation site (WS 10) versus a mid elevation site (WS 8, 7/6)? 2. Do snow-pack dynamics influence plant phenology more so than temperature (degree days)?
B51A-0040
Phenology of Net Ecosystem Exchange: A Simple Estimation Method
Carbon sequestration is important to global carbon budget and ecosystem function and dynamics research. Direct measurement of Net Ecosystem Exchange (NEE), a measure of the carbon sequestration of an ecosystem, is instrument, labor, and fiscally intensive, thus there is value to establish a simple, robust estimation method. Six ecosystem types across the United States, ranging from deciduous and coniferous forests to desert shrub land and grasslands, are compared. Initial results suggest instrumentally measured NEE and this proxy method are promising, showing excellent temporal matches of the two methods for onset and termination of carbon sequestration in a sub-alpine forest for the study period, 1997-2006. Moreover, the similarity of climatic signatures in all six ecosystems of this study suggests this proxy estimation method may be widely applicable across diverse environmental zones This estimation method is simply the interpretation of annual accumulated daily precipitation plotted against the annual daily accumulated degree growing days above a zero degree C base. Applicability at sub-seasonal time scales will also be discussed in this presentation.
B51A-0041
Phenology Across the LTER Network: Initial Findings, Future Directions
Phenology is, in the words of Aldo Leopold, a "horizontal science" that cuts across and binds together multiple biological disciplines. It is a far-reaching but poorly understood aspect of the environmental sciences. Phenological research has been a component of the Long Term Ecological Research (LTER) Network at several sites over the years. However, it has not received the attention or resources to bring it to the forefront as an effective theme for interdisciplinary and cross-site synthesis. With the recent establishment of the USA National Phenology Network (USA-NPN), it is appropriate to assess the status of phenological knowledge across the LTER Network. A workshop funded by the LTER Network Office took place at the Sevilleta Field Station during February 26 to March 2, 2007. From the workshop three main products emerged: (1) an inventory of LTER phenology datasets, (2) establishment of a website to facilitate information interchange, and (3) a white paper recommending next steps for the LTER Network to engage the USA-NPN. This poster relates the findings and recommendations of the workshop, including a summary of phenologically explicit and phenologically implicit LTER datasets and illustrations of how the climatic envelopes described by simple weather variables can provide context for phenological comparisons within and across sites.
B51A-0042
Exploring the Consequences of Changing Land Surface Phenology on Regional Hydrometeorology in the Eurasian Semi-Arid Grain Belt
Have the changes in land surface phenology following the 1991 collapse of the Soviet Union affected the regional hydrometeorology of the semi-arid grain belt that extends from northern Kazakhstan westward across southern Russia into eastern Ukraine? We modified MM5 running as a regional climate model with the NOAH land surface scheme to accept an updated fractional vegetation cover (FVC) layer every ten days. This new FVC time series was generated from the PAL NDVI dataset using the recent Scaled Vegetation Index algorithm (Jiang et al. 2006 RSE 101:366-378). The seasonal study period ran from March through September. We focused on years before and after the disintegration of the Soviet Union that were selected considering that the North Atlantic Oscillation (NAO) has been shown to affect conditions in at least part of the study area. We present results for 1985 (neutral NAO), 1989 (high NAO), 1996 (low NAO), and 1997 (neutral NAO). We focus on two response fields: monthly total precipitation and monthly specific humidity at 850 hPa. There are substantial spatially coherent differences between predictions from the default MM5 FVC series and the new FVC series as measured by RMSE and bias. Use of the new FVC image time series enables MM5/NOAH to respond to a more realistic representation of the tempo and spatial heterogeneity of land surface phenology.
B51A-0043
Phenology During Recent Droughts in the Southwestern U.S.A.
Droughts in the southwestern U.S.A. receive considerable examination of both the inherent climatic anomalies and subsequent effects. Here, we integrate spatial climate data and bioclimatic indices to further investigate climatic differences during the 1950s and 2000s droughts, and to present the concurrent phenologic differences. Higher temperatures and lower atmospheric moisture during the 2000s drought primarily occur in spring and summer, whereas similar conditions occur during the 1950s drought in fall. Phenologically these conditions indicate in general that the 2000s drought is less limiting in minimum temperatures and more limiting in water stress. In the variable topography of the southwest, however, differential responses of phenology appear along elevation gradients and between seasons. Considering the 2000s drought as a global-change-type drought, dry conditions under warmer temperatures that the data of this study support when compared to 1971-2000 climatologies, results suggest that phenologic responses to current and future climate change in the southwest may display spatiotemporal variability of both improved and worsened growing constraints.
B51A-0044
Landscape Phenological Characterization Using Time Series of Spectral Vegetation Indices: Arizona and its National Parks
The Arizona landscape is undergoing changes in vegetation growth patterns that are due to disturbances related to wildfire, extreme drought and precipitation events, and human interactions. Many vegetation communities are also being affected by invasive species, variation in species range, and patch biodiversity. One means of evaluating and monitoring these vital signs that is of interest to the National Park Monitoring Network is vegetation phenology. Time series of spectral vegetation indices are used to examine the vegetation growth trajectories in response to disturbance, climate and human interactions. Vegetation phenology of the Arizona landscape is characterized with a focus on National Parks and surrounding areas. Long term MODIS and AVHRR time series of vegetation index data (1989-current) are used to characterize phenological metrics that include: time of the start, peak and end of the growing season with corresponding vegetation index values and time integrated metrics related to seasonal biomass production. The spatio-temporal phenological characterization utilizing MODIS and AVHRR time series data show distinctive vegetation response patterns and trajectories that provide a means to monitor the natural resources of Arizona and the National Parks and provide a better understanding of landscape response to disturbance, climate and human activities that can better inform conservation and management practices.
B51A-0045
Predictive Phenologic Modeling Using MODIS: A Tool for Rangeland Management
Grazing is the predominant land use activity in the rangelands of the Inter-Mountain West. Vegetation phenology affects the impact of grazing; new plant growth is especially palatable to grazing animals. Over time, preferential grazing gives less palatable plants an advantage in rooting depth and may alter the composition of plant species and lead to soil erosion. The benefits of using remote spectral imagery to predict the onset and advancement of the phenologic phases for expansive and/or inaccessible areas is recognized, however, the practical application of this technology has been limited. A time series of Moderate Resolution Imaging Spectrometer (MODIS) vegetation indices was analyzed to identify the temporal profile of the growing season for surface vegetation in the Upper Colorado River Basin (UCRB). Drivers to which the progression of phenologic transition dates are most responsive were explored. Using phenological parameters defined from the MODIS time series and relationships of the phenological transition dates to the drivers, a predictive phenological model specific to the environmental parameters of the UCRB was developed. This model can be implemented at various spatial extents and temporal windows as part of a rangeland management strategy.
B51A-0046
AVHRR to MODIS Transition for Characterizing Land Surface Phenology
Challenges in characterizing land surface phenology from satellite imagery include creating a suitable phenology interpretation for difficult environments (e.g. arid lands, evergreen forests), dealing with non-vegetation signals, and more directly connecting the results to biophysical processes. In addition, transitioning from one generation of satellite sensors (e.g. AVHRR) to the next (e.g. MODIS) is critical for developing long-term satellite-derived records of phenology. However, this transition presents challenges for land surface phenology research as the sensors have differing radiometry, geometry, calibration, time of overpass and processing strategies. This study addresses some issues in transitioning between sensors by utilizing concurrent 16-day vegetation index composite data sets derived from AVHRR and MODIS. Phenological metrics including start of season and end of season time, duration of growing season, and time-integrated vegetation index are derived and evaluated for each sensor. Results show a varying agreement between the derived metrics, ranging from 0.28 to 0.68 r2 for different land cover types. Forested and arid ecosystems have the poorest agreement, while agricultural lands have the highest. While research is in progress to identify and account for the sources of these discrepancies, it is clear that transition algorithms will need to be developed and refined to enable longer term satellite application records.
B51A-0047
Using Light Interception and in-Situ Surface Reflectance in Support of Moderate Resolution Remotely Sensed Phenology
Time series of vegetation indices derived from remote sensing are increasingly being used to monitor vegetation phenology. However, a variety of factors confound interpretation of remote sensing results. To more fully exploit the power of remote sensing and to improve its utility for studies of landscape-to-regional scale phenology, more and better in-situ information is required for both calibration and validation purposes. In this paper, we present results from analyses of field data collected over the 2006 and 2007 growing season at Harvard Forest that use light interception and surface reflectance measurements to monitor canopy development. As part of this analysis we compare these high temporal measurements against 8-day composites of the enhanced vegetation index derived from the Moderate Resolution Imaging Sepctroradiometer. The goal of this study is to provide a proof-of- concept for future studies that include more spatially extensive and temporally continuous measurements related to canopy development. In the near future we hope to extend this study to include three additional forested sites in New England, thereby providing a transect that includes a gradient in climate and forest types.
B51A-0048
Mechanistic model for light-controlled leaf phenology in the Amazon rainforests
Satellite-based vegetation observations in the Amazon rainforest indicate a flush of leaves during the dry season when solar radiation is high. This light-controlled phenology is further confirmed with ground-based observations at the Tapajos National Forest (TNF; 2.86S, 54.96W, Para, Brazil) near km 67 of the Santarem-Cuiaba highway from 2001 to 2006. Observed leaf litterfall and canopy photosynthesis (Gross Primary Productivity: GPP) lags a few months past the seasonal variation of solar radiation. In well-watered rainforests, rich light leads to flush of new leaves, which have a high photosynthetic efficiency, consequently increasing GPP during the following months. In this study, we incorporate these mechanistic processes into the Ecosystem Demography model (ED) in order to capture the seasonality of leaf phenology and GPP, including the dry season flush of leaves. We use leaf litterfall rates, GPP and evapotranspiration measured at the TNF to constrain the model parameterizations. The initial model underestimates litterfall rates in both magnitude and seasonal fluctuation compared to the observed ones, and predicts seasonality of GPP opposite to the observed pattern, presenting peaks during the sunny dry season. The constrained model significantly improves the simulated litterfall rates and GPP against the observed ones. The model simulates litterfall rates quite accurately, and captures some of the seasonal dynamics of GPP. We also show that this modification in phenology, together with other changes in the model sensitivity to environmental conditions, improves the predicted seasonality of Net Ecosystem Exchange (NEE).
B51A-0049
Impacts of Pixel Deformation and Misregistration on Cross-calibration of Vegetation Index Data Records
The development of a long-term, seamless vegetation index (VI) data record requires data assemblage from multiple sensors and their cross-calibration. The latter could be performed by directly comparing data from an overlapping period of observations on a per-pixel or per-window (e.g., 5-by-5 window) basis. Due to differences in orbital, scanning, geolocation accuracy characteristics of sensors, however, pairs of observations to be cross- compared had different footprint sizes/scales and were acquired at slightly different locations, resulting in different coverage of surface areas. In this study, we characterized the effects of these footprint deformation and misregistration on cross-sensor VI comparisons. The objectives were to establish error bounds in cross- calibration results due to these effects and to develop recommendation for reducing the impacts of pixel deformation and misregistration on cross-calibration. Orbital, scanning, and geolocation error characteristics of three satellite sensors, Terra Moderate Resolution Imaging Spectroradiometer (MODIS), NOAA-14 Advanced Very High Resolution Radiometer (AVHRR), and SPOT-4 VEGETATION, were modeled and footprints of these three sensors were predicted for a 16-day compositing period (June 1998 and June 2002) over agricultural fields in Bondville, IL. An atmospherically-corrected Landsat Enhanced Thematic Mapper (ETM) image acquired within the compositing period was acquired and spatially aggregated to simulate normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) values for the modeled observation/footprint conditions. The results showed that root mean square errors (RMSE) of the NDVI and EVI values varied from day to day with the mean RMSE values of 0.02 for the NDVI and 0.015 for the EVI. After examining a series of spatial averaging with various window sizes, we found that taking 5-by-5 to 7-by-7 averages of neighboring pixels would effectively reduce RMSE to obtain reliable cross-calibration results. These error bounds and number of pixels to be averaged are expected to vary depending on surface heterogeneity and the same procedures need to be applied to other areas for obtaining general recommendation.
B51A-0050
A MODIS-Based Protocol for Monitoring Seasonal Processes in Southwest Alaska
The National Park Service's (NPS) Inventory and Monitoring (I&M) Program was established, in part, to describe long-term trends in ecosystem properties. The Southwest Alaska Network (SWAN) of parks, one of 32 networks in the I&M Program, is using Moderate Resolution Imaging Spectroradiometer (MODIS) data to measure the following metrics: 1) onset of ice formation and break up on large freshwater lakes, 2) timing and extent of snow cover, and 3) vegetation phenology. Remote sensing specialists with the U.S. Geological Survey (USGS), in cooperation with NPS-SWAN ecologists, have developed a set of protocols for acquiring, processing, and analyzing MODIS data. Season-dependent landscape processes are interpreted and analyzed using standard MODIS products including daily calibrated top-of-the-atmosphere radiance, snow cover extent, and vegetation index data. The timing and duration of seasonal landscape processes are reported for each of the metrics and are coupled with other data sources to interpret ecological patterns and changes across the park network. Although the I&M Program is primarily concerned with multiscale landscape processes that have ecological impacts within parks, they are also aware that park ecosystems can be significantly impacted by ecological conditions or events occurring outside of the parks. These MODIS-based protocols offer the ability to assess broad scale patterns in ecosystem performance throughout the region. While the MODIS record (2000 to present) is not yet sufficient for most statistical trend analyses, the approach taken in these protocols establishes a consistent data set that will be used to produce time-series analyses and to map regions of change for the SWAN. In this paper we present an overview of the methods employed and initial results for each of the seasonal metrics.
B51A-0051
Evaluating Modeled Vegetation Phenology Over North American Continent With Satellite and Ground-based Observations
The two-way interaction between vegetation and atmosphere plays an important role in global energy, water, and carbon cycles. The strong seasonal and interannual variability of leaf area index (LAI) and its vegetation- dependent spatial heterogeneity influences the short-term weather forecast and seasonal climate prediction. Therefore prognostic simulation of land-atmosphere interaction with respect to climate variability and change is crucial, and requires realistic representation of transpiring leaves in response to diurnal, seasonal, interannual, and longer-term changes in weather and climate. Three existing prognostic phenology models are selected to simulate vegetation growth: the Community Land Model version 3 CLM3-CN, CLM3-DGVM, and the Simple Biosphere Model Version 2.5 with the Growth Season Index (SiB2.5-GSI). Off-line model simulations over North America are carried out at 32-km grid spacing by use of the North American Regional Reanalysis (NARR) product. The modeled carbon, water and energy fluxes are validated with measurements from the AmeriFlux network. Both, ground-based LAI measurements (at points) and satellite remote sensing NDVI and FPAR (spatially resolved) products will be used to evaluate the modeled prognostic vegetation phenology in response to seasonal to interannual climate forcings. Our ultimate goal is to build a globally-applicable, multi-scale vegetation modeling system with prognostic vegetation phenology that can address the strong spatial heterogeneity, the seasonal and interannual variability of vegetation distribution and its associated biophysical parameters within the terrestrial water and carbon cycle.
B51A-0052
Evaluation of Multi-Sensor Semi-Arid Crop Season Parameters Based on NDVI and Rainfall
In semi-arid regions in Africa, the amount of rainfall, its distribution throughout the season and the length of the growing season are important sources of interannual variability in food production. Staple cereal crops in West Africa such as millet and sorghum are often photo-period sensitive and thus a sowing delay in the Sahel is expected to translate into yield reduction. Metrics that estimate the beginning of the growing season in semi-arid monsoonal ecosystems have become a central part of crop models and early warning indicators of possible future reductions in yield due to growing season length. Land surface phenology measures based on satellite observed vegetation data can provide a unique source of information about the start of the growing season and improve estimation of variations in food production. Here we present a new land surface phenology model which is tuned to the semi-arid, monsoonal ecosystem of the West African Sahel. We implement this model on vegetation index data from AVHRR, SPOT-Vegetation and MODIS data, and evaluate the results along with a rainfall-based start of season estimate for the region. We focus on determining which of four different NDVI datasets are best able to capture ground observations compared to a standard rainfall-based SOS using observed emergence of crops in West Africa as a benchmark. The results reveal good agreement between the satellite-derived estimate of the start of season and the sowing dates reported. The best agreement is reached for the MODIS data with a spatial resolution of 8km and a temporal resolution of 16 days. The RMSE for all datasets varies between 12 and 26 days, with higher errors for the SPOT data and lower errors for the MODIS data. Shorter compositing periods only provide small improvements in precision.
B51A-0053
Analyzing Vegetation Phenology in the Context of Climate Variability in Burkina Faso: Preliminary Results
Vegetation cover in semi-arid Burkina Faso is characterized by high annual and interannual variability, which can be attributed to rainfall patterns and is also impacted by human activities such as extraction of woody vegetation, cultivation and grazing. The year-to-year timing, intensity, and duration of the growing season have important implications for the economy, which is dominated by the agricultural sector. The choice of the study site was inspired by results from an earlier study of vegetation dynamics and climate variability in the Sahel. It includes the Central Plateau of Burkina Faso, which stands out due to a satellite- observed greening trend and evidence of increased vegetation density on the ground, as well as its surroundings, which do not show such a trend. The aim of this study is to better characterize vegetation types and their seasonal variability that might have contributed to the observed trends in vegetation greenness. Normalized Difference Vegetation Index (NDVI) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) for the years 2000 to 2006 were analyzed using TIMESAT, an algorithm that can extract phenological metrics from time series of remotely sensed vegetation index and other data types. The resulting metrics were used to (1) classify the study region into different vegetation types and to (2) assess short term trends and variation in different aspects of the phenological cycle. For the years 2003 - 2006, the observed spatial patterns of vegetation phenology were analyzed in conjunction with Climate Prediction Center Morphing Technique (CMORPH) satellite precipitation estimates, and seasonal metrics extracted from them. Preliminary results show that the inclusion of phenological metrics provides a sharper distinction of different vegetation types in land cover classifications. Differential responses of various phenological metrics to various precipitation metrics are expected. The findings could improve our understanding of rainfall-vegetation relationships in this semi-arid environment and help disentangle rainfall from other driving forces of vegetation dynamics.
B51A-0054
Historical Phenological Observations: Past Climate Impact Analyses and Climate Reconstructions
Plant phenological observations have been found an important indicator of climate change impacts on seasonal and interannual vegetation development for the late 20th/early 21st century. Our contribution contains three parts that are essential for the understanding (part 1), the analysis (part 2) and the application (part 3) of historical phenological observations in global change research. First, we propose a definition for historical phenonolgy (Rutishauser, 2007). We shortly portray the first appearance of phenological observations in Medieval philosophical and literature sources, the usage and application of this method in the Age of Enlightenment (Carl von Linné, Charles Morren), as well as the development in the 20th century (Schnelle, Lieth) to present-day networks (COST725, USA-NPN) Second, we introduce a methodological approach to estimate 'Statistical plants' from historical phenological observations (Rutishauser et al., JGR-Biogeoscience, in press). We combine spatial averaging methods and regression transfer modeling to estimate 'statistical plant' dates from historical observations that often contain gaps, changing observers and changing locations. We apply the concept to reconstruct a statistical 'Spring plant' as the weighted mean of the flowering date of cherry and apple tree and beech budburst of Switzerland 1702- 2005. Including dating total data uncertainty we estimate 10 at interannual and 3.4 days at decadal time scales. Third, we apply two long-term phenological records to describe plant phenological response to spring temperature and reconstruct warm-season temperatures from grape harvest dates (Rutishauser et al, submitted; Meier et al, GRL, in press). http://www.giub.unibe.ch/~rutis/index.html
B51A-0055
A cross-taxa phenological dataset from Mohonk Lake, NY and its relationship to climate
We present a detailed analysis of a rare cross-taxa native species phenology dataset (plant flowering, insect first sighting, and amphibian first sighting) from Mohonk Lake, NY. This dataset is highly unusual in North America for its longevity of record, consistency of methodology and location, diversity of species available, and availability of local daily meteorological data. For each phenology series, we examined the flowering and first sighting Julian calendar dates for the existence of temporal trends. Only one of the five animal species (katydid) showed any evidence for a significant trend in first sighting. In contrast, the plant species showed a rich mixture of temporal trends in flowering that could be divided into four classes: woody plant (no trend), woody plant (negative trend), herbaceous plant (negative trend), and herbaceous plant (positive trend). Many of the trends were found to be statistically significant and robust to the method of trend estimation. The data within each of the four plant classes were also pooled as anomalies to provide more complete temporal coverage for tests of trend robustness. The results were strongly consistent with the individual species trends within each class and highly significant statistically. We next correlated the flowering and first sighting dates against growing degree-day (GDD) summations for each day of the year to measure the sensitivity of each species to this common form of climatic forcing on phenology. All species showed a significant sensitivity to GDD summations, with peak correlations falling on or near the varying median flowering or first sighting dates. These results were robust whether the GDD analyses were conducted over the complete set of observations for each species (beginning on or after 1928) or over a latter period with a more serially complete set of cross-taxa observations (1970-2002). The GDD correlations indicate significant climate sensitivity in all species, but the different magnitudes of correlation and timing of maximum correlation imply that plant and animal responses to climate changes in the future will not be homogeneous for the tested species at Mohonk Lake.
B51A-0056
Using the Alexander Collection to measure the effects of climate change on the grasshoppers of the southern Rocky Mountains of Colorado
The current study utilizes the recently curated and databased Alexander Grasshopper Collection coupled with a new resurvey program to measure the effects of climate change on grasshoppers found along an elevational gradient in the southern Rocky Mountains of Colorado. The Alexander Collection is composed of approximately 19,000 pinned grasshoppers and a series of field data notebooks from a three year 1958-1960 survey project. During these survey years, Alexander processed over 65,000 grasshoppers from repeatedly sampled sites along an elevational gradient from Boulder (1530 m elev.) to Mt Evans (3900m elev.) in the Colorado Front Range. Data from 2006 shows that at mid-elevation sites grasshoppers are becoming adults 15-28 days earlier than they did nearly a half century ago. We found no changes in the time to reach adulthood at the high elevation sites. Preliminary data from 2007 (a year with milder spring temperatures) suggests that unlike the dramatic patterns documented in 2006, that the time to reach adulthood for grasshoppers at low and high elevation sites was not much different than it was 50 years ago. In 2007, several grasshopper species at mid-elevation did become adults earlier than they had a half century ago. http://alexander.colorado.edu/
B51A-0057
Sugar Maple Phenology: Anthocyanin Production During Leaf Senescence
The Northeastern United States is known for its brilliant fall foliage colors. Foliage is responsible for a billion dollar tourism industry. Many comment that past years have not resulted in the amazing color displays seen historically. As sugar maple trees senesce they contribute bright red leaves to the mural of oranges, yellows, and greens. The pigment that produces the red color, anthocyanin, is synthesized in the fall as chlorophyll slowly degrades. Remote sensing data from LandSat during fall senescence can help investigate this event by quantifying color change and intensity. This data can then be compared to ground validation efforts in several study plots. The results will help answer the question, "Why do leaves turn red?" One hypothesis is that this pigment acts as a photoprotectant and screens leaves from UV light. It is possible that an increase in tropospheric ozone has negatively affected fall foliage due to the increased reflection of UV light before it reaches the trees; thereby reducing the leaves need to produce anthocyanin. Another hypothesis is that production of anthocyanin is linked to temperature, with maximum synthesis occurring during cold evenings and moderate days. Temperature changes caused by climate change could also be affecting anthocyanin. Through observing these changes by remote sensing and ground experiments, more can be learned about this phenological stage and why it happens.
B51A-0058
Probabilistic Analysis of Spring Frost Risk
An important phenological event for vegetation is spring budbreak. The timing of this event represents an evolutionary adaptation to an acceptable risk of frost damage. We first present a general probabilistic analysis of spring frost risk to vegetation after budbreak based on a stochastic temperature process and a coupled model of biological development up to budbreak. Within this framework we consider the problem of how the risk regime for spring vegetation might change as a result of changes to the characteristics of the stochastic temperature process (e.g., variance and integral time scale).
B51A-0059
Effect of Ecosystem Warming on Boreal Black Spruce Bud Burst and Shoot Growth
The boreal forest is predicted to experience the greatest warming of any forest biome during the next 50-100 years, but the effects of warming on vegetation phenology are not well known. The objectives of this study were to (1) examine the effects of whole ecosystem warming on bud burst and annual shoot growth of black spruce trees in northern Manitoba, Canada and (2) correlate cumulative degree-days to the occurrence of bud burst. The experimental design was a complete randomized block design that consisted of four replicated blocks. Each replicate block was comprised of four treatments: soil warming only or heated outside (HO), soil and air warming or heated inside (HI), control outside (CO), and no soil or air warming inside the chamber, control inside (CI). On average shoot bud burst occurred seven days earlier in the HI than HO, CO, and CI treatments. Shoots required 71 more cumulative degree days (CDD0) to achieve bud bust in the HI than CO. In year one of the treatments, HI shoots burst before other treatments (HO, CO, CI), but annual shoot growth was less than CO. However in the second and third year of warming annual shoot growth was significantly greater for both HI (p = 0.001) and HO (p = 0.008) than CO and CI. Empirical results from this study suggest that climate warming may cause a possible shift in carbon allocation for boreal black spruce forests.
B51A-0060
Impact of Selective Logging on Phenology in Amazon Rain-Forests
Selective logging (SL) in the Brazilian Amazon was recently shown in analyses of Landsat ETM+ data at high spatial resolution to be occurring at rates of about 12000-–20000 km 2 per year, thus indicating the central role today of selective logging in disturbance of tropical forests. There has been few studies analyzing the effects of SL on forest phenology and no previous systematic study. Deforestation leads to persistent conversion of forest to another land cover/use type, which typically follows a significantly different phenological trajectory than the previous forest. Such changes in timing, growing season length, and trajectories of growth provide an integrated signal of altered ecosystem functionality. In contrast to deforestation, rapid closure of relatively small canopy gaps following SL may appear similar to natural regeneration phenomena in forests. As a result, forest biospheric processes and functions, as represented by phenological trajectories seem unaffected. We investigated the assumption that no significant change in forest function follows SL by analyzing a time-series of MODIS satellite data at 1-km scale in selectively logged forests in Mato Grosso, Brazil. The area studied is nearly 670000 km 2 and is characterized by a large number of small SL events that occurred between 1999 and 2000. We tracked the phenological changes induced by these events using a time series of two MODIS vegetation indexes (VI), the Enhanced Vegetation Index, and the Normalized Difference Water Index. First, we show that even relatively low levels (5-–10%) of canopy damage cause significant and long-lasting (> 3 years) changes in forest phenology. Partial clearing impedes forest green-up in the dry season, progressively dehydrates the canopy, and induces overall seasonal deficits in canopy moisture and greenness. We outline the advanced optimized contextual statistical method used to detect phenological impacts, involving estimation of spectral changes at each 16-day time step. We discuss the SL impact in terms of phenological metrics derived from the observed and predicted time series and in relation to the SL intensity. Given the large and constantly increasing geographic extent of selective logging throughout Amazonia, phenological disturbances may have far- reaching impacts on carbon and water fluxes, nutrient dynamics, and other functional processes in Amazon forests.
B51A-0061
Quantifying the growing season dynamics and phenology of a boreal black spruce wildfire chronosequence: Coupling field measurements with MODIS
The boreal forest is the second largest forested biome and the vast area and large carbon stores in the soil makes these forests important to the global carbon, water and energy cycles. Analysis of global coverage, coarse resolution satellite Vegetation Index (VI) data have provided considerable information on the seasonal cycles of vegetation in the mid-to high-latitudes, including the boreal forest, with evidence of an increase in the magnitude of vegetation greenness and a lengthening of the active growing season, which has been attributed to climate warming. However, boreal forests are prone to extensive wildfire disturbance that influence canopy dynamics (i.e. species composition, LAI, and phenology) and separating the direct affect of warming from the indirect affect of increased wildfire frequency on the patterns of boreal phenology and seasonal greeness requires further analysis coupled to ground measurements. In this research we address the need for detailed information on the growing season dynamics and phenological patterns of boreal vegetation. We evaluate whether MODIS reflectance data can resolve small inter-annual variations in canopy phenology and growing season dynamics of boreal forests. We quantified the seasonality and inter-annual differences of the overstory and understory vegetation by optically measuring the LAI and light harvesting potential (FPAR) during the 2004-2006 growing seasons. An automated continuously operating system is used to monitor growing season PAR transmittance. We focused on a boreal wildfire chronosequence of sites comprising a range of forest ages (1-154 years since fire) to quantify the differences in vegetation dynamics and phenology between the deciduous/mixed and coniferous forests. The spatial and temporal characteristics of LAI / FPAR within the chronosequence were examined by comparing both the in situ measurements and the relevant MODIS products. A statistical curve fitting procedure is used to derive the key phenological transition periods of vegetation phenology across the chronosequence for both the in situ and MODIS time series data. In addition, we assessed the uncertainty in these estimates using Monte Carlo simulations to obtain 95% confidence intervals for each modeled transition date. This information is used to examine the effects of temporal aggregation, seasonal cloud and other residual atmospheric effects on determining phenological dates with MODIS. Collectively these data help discern the role warming and increased wildfire have in modifying boreal growing season dynamics while extending our ongoing long-term work to systematically link field measurements, remote sensing and ecosystem modeling to quantify the effects of global change on the carbon budget of boreal forests.
B51A-0062
Intensive Landscape Phenology
Plant phenology is typically observed for selected individuals at discrete in-situ locations, lacking representation of complex local ecological relationships. Land Surface Phenology recorded by satellite sensors integrates ecosystem information over large areas, but with less detail and accuracy. A new approach-Landscape Phenology is proposed to study plant phenology in the context of landscape ecology, by intensive sampling of an entire vegetation patch along with the relevant biological and microenvironmental factors. This ongoing research project is collecting intensive tree phenology data near the WLEF AmeriFlux tall-tower site in the northern mixed forest of Wisconsin. Two years (2006 and 2007) spring tree phenologies were observed with bi-daily temporal resolution for 216 forest trees in a 300m by 600m area using cyclic sampling design. Tree phenologies were scored with a customized protocol optimizing the eyeball observation accuracy. Ground cover phenologies were sampled with both eyeball and digital cameras in 2007. Intensive air temperature and humidity were also measured concurrently across the entire sampling area with on-site HOBOs. Additionally, LiDAR data provided high resolution topographic and tree heights information, which are integrated into the analysis. Results so far suggest that at this scale, the phenological behaviors of forest trees are rather individualistic, with no spatial autocorrelation detected. The temporal differences of leaf onset among individuals across the landscape are likely affected by biological differences of plants, rather than microclimatic variations. However, for each species, the average phenological advance rates are apparently correlated with the energy availability over the previous two days. In order to reconstruct the Landscape Phenology time series from the sample data, a simple average may be sufficient for each species, and a spatial average based on plant community distribution and high resolution satellite imagery (IKONOS/Quickbird) should be used to complete the quantification for the entire study area or selected portions of it. Landscape Phenology expands the scope and depth of current phenological studies. A broad application of this project is to calibrate the currently available medium-resolution satellite-based phenological indices for more accurate monitoring of global climatic change impact on terrestrial ecosystem functionalities.
B51A-0063
Phenology as an Integrative Science for Assessment of Global Climate Change Impacts
Phenology is the study of periodic plant and animal life cycle events and how these are influenced by seasonal and interannual variations in climate. Examples include the timing of leafing and flowering, agricultural crop stages, insect emergence, and animal migration. All of these events are sensitive measures of climatic variation and change, are relatively simple to record and understand, and are vital to both the scientific and public interest. Integration of spatially-extensive phenological data and models with both short and long-term climatic forecasts offer a powerful agent for human adaptation to ongoing and future climate change. However, a new data resource of national scale is needed to capture the valuable information potential of phenological responses to climate change; to study its nature, pace and the effects of ecosystem function; and to understand connectivity and synchrony among species. The USA National Phenology Network (USA-NPN) is being designed and organized to engage federal agencies, environmental networks and field stations, educational institutions, and mass participation by citizen scientists to create this data resource, and develop phenology research potential. This presentation illustrates the variety of source, scale, and use of phenology in assessing current and future global climate change impacts.
B51A-0064
The Northeast Phenology Network: A Regional Coordinating Group of the National Phenology Network
Recent studies conducted in northeastern North America reveal an overall pattern of increasing growing season length in response to recent warming trends, and also highlight considerable, and poorly understood, spatial variability in the species-level and land surface phenologies of the region's predominately forested ecosystems. To obtain the spatially-explicit understanding of phenology that is necessary for predicting the ecosystem-level impacts of climate change and climate variability in this region, we formed the Northeast Phenology Network (NEPN) as the first regional coordinating group of the National Phenology Network (NPN). This presentation will report the organizational and preliminary scientific progress made by the NEPN through its first workshop held in November 2007 with over 35 of the region's phenology researchers and representatives from citizen-science groups. We report progress made toward four immediate goals of the NEPN: (1) seeking out and inviting the region's scientific and citizen science groups to contribute new and archived phenological measurements to the NPN (2) vetting protocols for phenology measurements in the region, (3) developing region-specific phenology monitoring training materials and "train the trainer" workshops, and (4) preparing scientific syntheses of the multiple forms of phenological data (e.g. satellite and field) collected in the region.
B51A-0065
Development of Cyberinfrastructure to Support the USA National Phenology Network
The USA National Phenology Network (USA-NPN) is an emerging and exciting partnership among academic communities, federal agencies, and volunteers. The USA-NPN consists of four components, representing different levels of spatial coverage and quality/quantity of phenological and related environmental information: 1) Locally intensive sites focused on process studies; 2) Spatially extensive scientific networks focused on large- scale phenomena; 3) Volunteer and Education Networks; and 4) remote sensing products that can be validated against ground observations and assimilated to extend surface phenological observations to the continental- scale. A critical challenge for the USA-NPN is the development of a robust and cost-effective cyberinfrastructure which supports the distributed collection and management of phenological data across these differing components. In this presentation, we will present the Phase I cyberinfrastructure for USA-NPN, designed to support the most critical needs across these tiers, as well as the structure planned for Phases II and III. Phase I is based on a range of technologies leveraged from several efforts, including the Plant Phenology Network, the National Biological Information Infrastructure Metadata Clearinghouse, Project BudBurst, and the ORNL Distributed Active Archive Center. http://www.usanpn.org
B51A-0066
Spatial Variability in Wheat Phenology Across a Landscape
Winter wheat phenology varies among shoots on the plant to main stems on plants within a plot to locations across a landscape. Most often phenological measurements have focused on small treatment plots under presumably similar soils and topography. Many models exist to predict wheat phenology for small treatment plots based on the primary drivers of temperature and photoperiod, with secondary factors of water deficits, nutrient availability, and atmospheric CO2 levels occasionally considered. When considering winter wheat phenology across a landscape with varying topography and soils, differences in phenology would be expected as the drivers vary. We conducted a 6-year experiment measuring winter wheat phenology across approximately a 100-ha landscape with topographic and soil variation. At each of 10 landscape positions per growing season, wheat emergence and biomass at different growth stages were sampled in randomly nested patterns. Air and soil temperature and soil water content were measured at each location. Preliminary analysis on 3 of the 6 years of data shows that depending on location, the time of main stem jointing, flag leaf completing growth, heading, anthesis, and physiological maturity varied as much as 23, 17, 15, 18, and 9 days, respectively, within a year. The location rankings often did not remain consistent across growth stages. Maturity growth stage had the lowest range of variation, and in our environments may reflect that regardless of location and year, resources universally across locations became sufficiently limiting at nearly the same time causing growth to cease. Given that the same cultivar was used, what are the sources of this variation? The photoperiod factor can be eliminated from consideration since it was similar at all locations. One important source of variation was planting date that greatly impacted seedling emergence patterns. In the one year with two planting dates, the variation decreased by 17 days for the jointing growth stage within a planting date, and often for all stages other than maturity, typically 10 or more days. Variation due to different air and soil temperatures will be estimated by conversion to thermal time. In addition, the influence of soil water content on thermal time will be considered.
B51A-0067
Phenological indicators of forest composition in northern deciduous forests
The phenology of deciduous forests in temperate climates is triggered by winter and spring temperatures. While the exact forcing mechanism is still unclear and spatial-temporal models are still in development, the early 20th century Hopkins Law, which estimates leaf emergence from latitude and elevation, still yields roughly accurate predictions. A newly parameterized version of this simple correlation using MODIS satellite phenology suggests that average leaf phenology in New England can be predicted effectively if deciduous-dominated forests are parsed into two communities. Where the Hopkins prediction and satellite phenology indicate discrepancies corresponds closely to the boundary between northern and central hardwood forests. It is suggested that despite the absence of an accurate phenology-climate model, simple average temperature relationships can still be used to explore spatial phenological patterns and detect fundamental forest compositional boundaries. If, in fact, compositional forest boundaries in temperate forests are primarily determined by climatological parameters, this research may further imply that the ability for different forest types to respond phenologically to spring climates plays a key role determining forest composition. The climate-defined boundary zones between forest types may function as a key indicator of climate change on forest function.