1. Introduction
Vegetation in seasonally cold climates, with a pronounced springtime leaf emergence and summer growing season followed by leaf senescence with the onset of short, cold days in autumn, influences atmospheric seasonality. This is seen most prominently in atmospheric CO2 concentrations, which fluctuate seasonally as plant photosynthetic activity varies with the emergence and senescence of leaves in the Northern Hemisphere (Bacastow et al. 1985; Randerson et al. 1997). Relatively small changes in the timing of leaf emergence or senescence can impart large changes in net carbon uptake by terrestrial ecosystems (White et al. 1999; Barford et al. 2001). A less-known effect of leaf emergence is on spring temperature trends, as observed in the meteorological record (Schwartz and Karl 1990; Schwartz 1992, 1996; Hogg et al. 2000; van den Hurk et al. 2003). Rapid spring warming is found to slow at the time of deciduous plant leaf emergence, presumably due to a rise in plant transpiration (Hogg et al. 2000; Fitzjarrald et al. 2001; Schwartz and Crawford 2001). Humidity and other atmospheric variables also respond to leaf emergence (Freedman et al. 2001; Schwartz 1992).
Because of the importance of foliage in regulating surface climate, improved representations of leaf area and phenology (i.e., the emergence and senescence of foliage) are being implemented in regional and global climate models. In particular, prognostic models of leaf area that determine the phenology and amount of foliage according to temperature, precipitation, and plant productivity, all of which in turn depend on leaf area, are being included in the land models used with climate models (Dickinson et al. 1998; Lu et al. 2001; Tsvetsinskaya et al. 2001a,b). These prognostic leaf models generally improve the simulated climate by better representing the seasonal development of leaf area and usually lowering leaf area index (LAI) compared to alternative prescribed surface datasets. However, these studies have not examined the impact of leaf emergence on springtime temperature trends.
Here, we use an atmospheric general circulation model coupled to a land surface model with prognostic leaf area to study, in particular, the relationship between slower springtime warming and leaf emergence. We hypothesize that a coupled atmosphere–vegetation model is needed to reproduce the observed effect of leaf emergence on springtime temperature trends. We explore the effects on these trends caused by different modeling approaches to phenology, and we recommend our preferred approach.
2. Model description
Simulations were performed using the Community Atmosphere Model, version 2 (CAM2), a three-dimensional atmospheric general circulation model (Collins et al. 2002), coupled to the Community Land Model, version 2 (CLM2; Bonan et al. 2002a). In these simulations, we used climatological sea surface temperatures and sea ice, spectral T31 horizontal resolution (∼3.75° latitude × 3.75° longitude), the standard CAM2 vertical resolution of 26 atmospheric levels, and a 30-min time step.
CLM2 simulates land surface energy, moisture, and momentum fluxes. A prescribed time-invariant land cover dataset gives the fractional abundance of glaciers, lakes, wetlands, and various plant functional types within each model grid cell. Prescribed monthly average leaf area index datasets, linearly interpolated to daily values, give the leaf area index for each plant functional type in a grid cell (Bonan et al. 2002b). As an alternative, CLM can also grow plants to allow synchronous coupling between the atmosphere and vegetation (Bonan et al. 2003). With dynamic vegetation enabled, vegetation cover and leaf area index are simulated rather than obtained from prescribed surface datasets. Community composition and ecosystem structure are updated with an annual time step in response to establishment of new plants, resource competition, growth, mortality, and fire. An algorithm for leaf phenology (similar to Kucharik et al. 2000) updates leaf area index daily in response to air temperature for summergreen (cold deciduous) plants and soil water for raingreen (drought deciduous) plants. In the present study, we focus on the temperature effects of leaf emergence in middle to high latitudes, where summergreen trees are abundant.
3. Simulations
We used CAM2–CLM2 to study the relationship between springtime leaf emergence and air temperature and to highlight the role of leaf–atmosphere coupling in determining this relationship. We performed four 20-yr simulations to explore such plant–atmosphere interactions at the phenological (i.e., daily) time scale:
Prognostic leaf area (P)—Vegetation dynamics active.
Prescribed leaf area datasets (D)—This simulation disabled the dynamic vegetation so that CLM2 was run with prescribed vegetation datasets, as in the standard CLM2. However, vegetation datasets were obtained from simulation P rather than the standard satellite-derived datasets. Monthly leaf area index was prescribed with 20 yr of vegetation output from the prognostic leaf area simulation (P) and then interpolated to daily values as in the standard CLM2. The distribution of plant functional types was also prescribed as simulated in P for each of the simulation's 20 yr.
- Prognostic leaf area, slow phenology (P/4)—Though simulations P and D use the same monthly average leaf area index, simulation D, with the prescribed monthly datasets, has a smoother green-up than simulation P because the monthly leaf area index is interpolated to daily values. To allow the prognostic leaf area to mimic the slower phenological time scale of simulation D, a simulation was performed with prognostic leaf area, but leaf emergence time scales lengthened approximately four fold by modifying Eqs. (1) and (2), where emergence:and senescence:
Prescribed leaf area datasets, fast phenology (D×4)—This simulation used the same vegetation data as simulation D but with leaf emergence accelerated approximately four fold to mimic the time scale of phenology in simulation P.
The climate–vegetation system was initialized in two phases prior to these simulations. A 400-yr CLM2 simulation driven with present-day atmosphere data grew vegetation from initially bare ground. This provided initial conditions for an 80-yr coupled CAM2–CLM2 simulation that allowed the vegetation to come to equilibrium with the simulated climate.
Model output is analyzed relative to the “day of first leaf,” similar to Schwartz and Karl (1990). First leaf is defined in each year for each model grid cell as the last day before the leaf area index exceeds 2% of its annual range. Trends before and after first leaf are calculated for all variables by omitting the 15 days centered on first leaf in order to remove the bias of warm events leading to leaf emergence in simulations with prognostic phenology.
4. Results and discussion
We focus on three regions of interest selected for their simulated compositions of vegetation: northern Europe (60°–70°N, 5°–45°E), where the simulated vegetation is predominantly a mix of deciduous and evergreen trees (43% boreal needleleaf evergreen tree, 19% boreal summergreen tree, 29% temperate summergreen tree, and 8% grass); central Canada (51°–61°N, 102°–92°W), where the simulated vegetation is predominantly evergreen tree (74% boreal needleleaf evergreen tree, 17% boreal summergreen tree, and 8% grass); and eastern China (35°–45°N, 100°–130°E), where the simulated vegetation is predominantly deciduous tree (54% temperate summergreen tree, 11% boreal needleleaf evergreen tree, 8% boreal summergreen tree, 19% grass, and 7% unvegetated). The simulated vegetation is representative of the ecosystems expected in these three regions (Bailey 1996). Our goal is not to evaluate the atmosphere or vegetation simulations in these regions by focusing on specific biases but to assess the sensitivity of the springtime temperature trends to the timing of leaf emergence.
Simulated leaf phenology compares favorably to satellite-derived leaf area in northern Europe and central Canada (Fig. 1). Simulated leaf area index is systematically biased high in northern Europe, but the seasonal development of foliage concurs with the satellite observations. The average simulated day of first leaf for this region is 1–2 May. Observed leaf emergence occurs on 4 April, 7 May, and 23 May for three smaller areas encompassed by our region (Chmielewski and Rötzer 2001). Simulated leaf area index is also biased high in central Canada, but the gradual green-up of vegetation is fairly well reproduced. Simulated leaf area index is much higher than satellite observations in eastern China (Fig. 1), but this is a region in which the present-day vegetation has a high abundance of croplands. The vegetation dynamics used in the model only simulates natural vegetation, and this could account in part for the discrepancy with the observations. Monthly leaf area index behaves similarly for all four simulations within a region, differing only in the timing of leaf emergence (Fig. 2). Maximum leaf area index in the P/4 simulation is less than in the other simulations, especially in the cold climate of northern Europe, due to the shorter growing season resulting from slower leaf emergence. In addition, this simulation maintains a higher leaf area index than the other simulations in autumn due to comparably slower leaf drop.
a. Northern Europe
Surface air temperature shows a marked increase in the several days preceding leaf emergence in the P simulation (Fig. 3). This is consistent with observations and reflects the fact that leaf emergence in summergreen vegetation is triggered by warm temperatures (Schwartz and Karl 1990). This warm-up is less prominent in the P/4 simulation due to the longer time scale over which leaves emerge. It is absent in the D and D × 4 simulations because there is no relationship between air temperature and leaf emergence in these prescribed vegetation simulations.
The P, P/4, and D × 4 simulations have a reduction in springtime warming trend after leaf emergence compared to before leaf emergence (Table 1). This is in agreement with observations (Schwartz and Karl 1990). In contrast, the D simulation has an increased rate of warming following green-up.
Leaves emerge earlier in the year in the D simulation (9 days earlier than D × 4, 13 days earlier than P, and 16 days earlier than P/4), when the climate is colder than in the other simulations. This is an artifact of the smooth daily leaf area index, linearly interpolated from monthly means, in the D simulation compared to the potentially abrupt from one day to the next prognostic leaf emergence in simulations P and P/4. The rapid green-up of D × 4 resembles more closely the behavior of the prognostic phenology in P. Leaves in the P/4 simulation generally flush out at the latest date because of slower leaf emergence properties.
The ability of the model to replicate the observed reduction in springtime warming trend is related to the timing of leaf emergence relative to the prevailing environmental conditions. With prescribed leaf area, leaf emergence occurs independently of the surface climate; air temperature may be too cold, soils may be frozen, and photosynthesis and hence transpiration are precluded from occurring. In contrast, foliage emerges and photosynthesis begins with prognostic leaf area when environmental conditions are favorable. In particular, leaf emergence with prognostic leaf area is closely correlated with the end of snowmelt. Observations also show a strong correlation between day of leaf emergence and loss of snow cover (Schwartz 1992, 1996). This difference between prescribed and prognostic leaf area is seen in the simulated snow depth (Fig. 3). In the P and P/4 simulations, foliage does not emerge until much of the snowpack has melted. In these simulations, surface albedo (not shown) decreases markedly in the days preceding leaf emergence due to the melting of snow. By the time of green-up, much of the snowpack has melted and surface albedo is close to its seasonal low. In contrast, foliage emerges in the D simulation independent of prevailing snow cover, and there is a weaker correlation between time of leaf emergence and end of snowmelt (SM). With its more rapid green-up, the D × 4 simulation, though using prescribed leaf area, happens to better capture the timing of leaf emergence relative to the snowpack.
Because prognostic leaf emergence responds to the prevailing environment, conditions are favorable for stomata to open and leaves to photosynthesize. This is seen in the rapid increase in photosynthesis immediately following leaf emergence in the P and P/4 simulations (Fig. 3). In contrast, photosynthesis increases much more slowly following leaf emergence in the D simulation. Photosynthesis in D × 4 behaves more as in P and P/4 because of the closer correlation between leaf emergence and favorable environmental conditions.
The different environmental conditions at time of leaf emergence in the P, P/4, and D × 4 simulations compared to the D simulation are seen in the surface energy budget (Fig. 3). In the P and P/4 simulations, with their close correlation between end-of-snow melt and leaf emergence, less energy is used to melt snow and more is used to warm the soil in the days immediately prior to and after leaf emergence. The same is true for the D × 4 simulation, though the snowpack persists longer after leaf emergence. In contrast, snowmelt occurs for up to 28 days following leaf emergence in the D simulation. As a result of favorable environmental conditions at the time of leaf emergence in the P, P/4, and D × 4 simulations, stomata open for the leaves to photosynthesize and water is lost in transpiration. The flux of latent heat (LH) begins to increase in response to leaf emergence. The ratio of sensible heat flux (SH) to latent heat flux, called the Bowen ratio (not shown), decreases through much of spring as observed (e.g., Fitzjarrald et al. 2001), but especially near and after leaf emergence. In contrast, latent heat flux increases more gradually in the D simulation.
Because leaf emergence and the end of snowmelt are so closely correlated when using prognostic leaf area, it is possible that the loss of snow cover, not leaf emergence, controls surface air temperature (Schwartz 1992, 1996). However, one would expect an accelerating warming trend with loss of the snowpack because progressively less energy is used to melt highly reflective snow and more is used to warm the darker soil. This would operate against the cooling effect of leaf emergence on the temperature trend. Following the disappearance of snow, the effect of leaf emergence should be uninhibited. We confirm this by examining the warming trend in only those grid cells where leaf emergence occurs after the end of snowmelt. The warming trend shows similar and normally greater changes before and after leaf emergence as compared to the overall regional average (Table 1).
b. Central Canada
Results for central Canada are similar to those for northern Europe. Leaf emergence with prognostic leaf area is preceded by a period of rapid warming (Fig. 4). The P, P/4, and D × 4 simulations have a reduction in springtime warming trend after leaf emergence compared to before leaf emergence (Table 1). This reduction is greater with prognostic leaf area (P and P/4) than with prescribed leaf area (D × 4). The D simulation has an increase in warming trend after leaf emergence. Analyzing only those grid cells where leaf emergence occurs after the end of snowmelt does not change the results except that now the D simulation also shows a reduced temperature trend.
Leaf emergence in the D simulation occurs 16 calendar days earlier and when temperatures are colder than in the P, P/4, and D × 4 simulations (Table 1). Leaf emergence in the P, P/4, and D × 4 simulations is characterized by a reduction in snowmelt and an increase in photosynthesis and latent heat (Fig. 4). In the D simulation, foliage emerges when snow is still on the ground and conditions are unfavorable for photosynthesis, so the latent heat flux increases more slowly.
c. Eastern China
As in the other two regions, leaf emergence with prognostic leaf area is preceded by a period of rapid warming (Fig. 5). In this region, all four simulations show the reduced springtime warming trend (Table 1). All four simulations show increased photosynthesis and latent heat upon leaf emergence, though this is weaker for the D simulation (Fig. 5). Though foliage emerges 10–18 days earlier in the D simulation than in the other simulations, the warmer climate and less snow cover of this region compared to northern Europe and central Canada ensure that conditions are favorable for photosynthesis and transpiration upon leaf emergence in all simulations. In addition, this region has a greater abundance of broadleaf deciduous trees than the other regions. Both broadleaf deciduous and needleleaf evergreen trees can photosynthesize at cold air temperatures if the soil is unfrozen, but broadleaf deciduous trees have a greater rate of stomatal conductance and photosynthesis, for a given temperature, than needleleaf evergreen trees. As a result, transpiration in this region has a greater sensitivity to changes in leaf area index than in the other regions with greater coverage of needleleaf evergreen trees.
Surface albedo (not shown) increases slightly following leaf emergence in all four simulations. For example, surface albedo is 0.13 immediately prior to leaf emergence (from −5 to −1 days) and 0.14 after leaf emergence (from 6 to 10 days) in simulation P. This is in agreement with observations and arises because leaf reflectivity in the near infrared is generally higher than that of soils (Schwartz 1992; Moore et al. 1996). This behavior in the albedo slows the rise in net radiation and contributes to the slower warming trend after first leaf.
5. Conclusions
Observations show that emergence of foliage in springtime cools surface air temperature (Schwartz and Karl 1990; Schwartz 1992, 1996; Hogg et al. 2000; Fitzjarrald et al. 2001). The simulated changes in temperature trend following leaf emergence are consistent with these observations. In particular, they agree with observations by Schwartz and Karl (1990) showing a reduced rate of daily temperature warming in springtime after leaf emergence. Analysis of the surface energy budget shows that the cooling coincides with the latent heat flux rising faster than other terms in the energy balance and often becoming the dominant term after leaf emergence. This, too, is consistent with observational studies, which show rising latent heat flux compared to sensible heat flux following leaf emergence and attribute the temperature cooling to greater transpiration following leaf emergence (Hogg et al. 2000; Wilson and Baldocchi 2000; Fitzjarrald et al. 2001; Schwartz and Crawford 2001).
The failure of the simulation with prescribed leaf area (D) to capture the reduced warming trend is not due to its longer period of leaf emergence compared to the prognostic leaf area simulation (P). A simulation with prognostic leaf area but with a longer time scale of leaf emergence (P/4) still shows the reduced warming trend. Prescribed leaf area with rapid green-up (D × 4) also has a reduced warming trend, suggesting that prognostic leaf area is not required to replicate the reduced warming trend. However, this may be a fortuitous result in that the D × 4 simulation merely replicates prognostic phenology better.
The primary difference between prescribed leaf area and prognostic leaf area relates to the timing of leaf emergence. Leaf emergence occurs early in the year in simulation D due to the smoothing effect of monthly averaged leaf area index linearly interpolated to the daily one. Early leaf emergence has a weaker effect on latent heat because cooler conditions may not support intense photosynthesis and, therefore, transpiration. This is particularly evident in northern Europe and central Canada but not in eastern China, where a warmer climate allows photosynthesis despite early leaf emergence. In contrast, with prognostic leaf area foliage emerges when environmental conditions are favorable for photosynthesis. Simulation P/4, with its lengthened time scale for green-up, presents the extreme example of this, with leaf emergence occurring later than in the other simulations. With warmer ambient temperatures, photosynthesis and transpiration are more sensitive to leaf emergence, resulting in a stronger effect of latent heat on temperature.
Climate model simulations with prognostic leaf area reveal a tight coupling between atmospheric and ecological processes. The ability of the model to reproduce the reduction in springtime warming trend following leaf emergence depends on the timing of leaf emergence relative to optimal conditions for photosynthesis. Prognostic leaf area ensures that foliage emerges when environmental conditions are right for photosynthesis. With prescribed monthly leaf area, foliage may emerge when conditions such as cold air temperatures or frozen soil preclude photosynthesis from occurring. These results demonstrate that the onset of spring seen in temperature records is a manifestation of dynamic vegetation–atmosphere coupling and occurs when photosynthesis, stomatal conductance, and leaf emergence are synchronized with atmospheric conditions.
Acknowledgments
The authors thank two anonymous reviewers and Richard Betts for their valuable comments. This work was supported in part by the NASA Land Cover Land Use Change program through Grant W-19,735.
REFERENCES
Bacastow, R. B., C. D. Keeling, and T. P. Whorf, 1985: Seasonal amplitude increase in atmospheric CO2 concentration at Mauna Loa, Hawaii, 1959–1982. J. Geophys. Res, 90 , 10529–10540.
Bailey, R. G., 1996: Ecosystem Geography. Springer, 204 pp.
Barford, C. C., and Coauthors, 2001: Factors controlling long- and short-term sequestration of atmospheric CO2 in a mid-latitude forest. Science, 294 , 1688–1691.
Bonan, G. B., 1996: A land surface model (LSM version 1.0) for ecological, hydrological, and atmospheric studies: Technical description and user's guide. NCAR Tech. Note NCAR/TN-417+STR, National Center for Atmospheric Research, Boulder, CO, 150 pp.
Bonan, G. B., S. Levis, L. Kergoat, and K. W. Oleson, 2002a: Landscapes as patches of plant functional types: An integrating concept for climate and ecosystem models. Global Biogeochem. Cycles,16, 1021, doi:10.1029/2000GB001360.
Bonan, G. B., K. W. Oleson, M. Vertenstein, and S. Levis, 2002b: The land surface climatology of the Community Land Model coupled to the NCAR Community Climate Model. J. Climate, 15 , 3123–3149.
Bonan, G. B., S. Levis, S. Sitch, M. Vertenstein, and K. W. Oleson, 2003: A dynamic global vegetation model for use with climate models: Concepts and description of simulated vegetation dynamics. Global Change Biol, 9 , 1543–1566.
Chmielewski, F-M., and T. Rötzer, 2001: Response of tree phenology to climate change across Europe. Agric. For. Meteor, 108 , 101–112.
Collins, W. D., J. K. Hackney, and D. P. Edwards, 2002: An updated parameterization for infrared emission and absorption by water vapor in the National Center for Atmospheric Research Community Atmosphere Model. J. Geophys. Res.,107, 4664, doi:10.1029/2001JD001365.
Dickinson, R. E., M. Shaikh, R. Bryant, and L. Graumlich, 1998: Interactive canopies for a climate model. J. Climate, 11 , 2823–2836.
Fitzjarrald, D. R., O. C. Acevedo, and K. E. Moore, 2001: Climatic consequences of leaf presence in the eastern United States. J. Climate, 14 , 598–614.
Freedman, J. M., D. R. Fitzjarrald, K. E. Moore, and R. K. Sakai, 2001: Boundary layer clouds and vegetation–atmosphere feedbacks. J. Climate, 14 , 180–197.
Hogg, E. H., D. T. Price, and T. A. Black, 2000: Postulated feedbacks of deciduous forest phenology on seasonal climate patterns in the western Canadian interior. J. Climate, 13 , 4229–4243.
Kucharik, C. J., and Coauthors, 2000: Testing the performance of a dynamic global ecosystem model: Water balance, carbon balance, and vegetation structure. Global Biogeochem. Cycles, 14 , 795–825.
Lu, L., R. A. Pielke Sr., G. E. Liston, W. J. Parton, D. Ojima, and M. Hartman, 2001: Implementation of a two-way interactive atmospheric and ecological model and its application to the central United States. J. Climate, 14 , 900–919.
Moore, K. E., D. R. Fitzjarrald, R. K. Sakai, M. L. Goulden, W. Munger, and S. C. Wofsy, 1996: Seasonal variation in radiative and turbulent exchange at a deciduous forest in central Massachusetts. J. Appl. Meteor, 35 , 122–134.
Randerson, J. T., M. V. Thompson, T. J. Conway, I. Y. Fung, and C. B. Field, 1997: The contribution of terrestrial sources and sinks to trends in the seasonal cycle of atmospheric carbon dioxide. Global Biogeochem. Cycles, 11 , 535–560.
Schwartz, M. D., 1992: Phenology and springtime surface-layer change. Mon. Wea. Rev, 120 , 2570–2578.
Schwartz, M. D., 1996: Examining the spring discontinuity in daily temperature ranges. J. Climate, 9 , 803–808.
Schwartz, M. D., and T. R. Karl, 1990: Spring phenology: Nature's experiment to detect the effect of “green-up” on surface maximum temperatures. Mon. Wea. Rev, 118 , 883–890.
Schwartz, M. D., and T. M. Crawford, 2001: Detecting energy-balance modifications at the onset of spring. Phys. Geogr, 22 , 394–409.
Tsvetsinskaya, E. A., L. O. Mearns, and W. E. Easterling, 2001a: Investigating the effect of seasonal plant growth and development in three-dimensional atmospheric simulations. Part I: Simulation of surface fluxes over the growing season. J. Climate, 14 , 692–709.
Tsvetsinskaya, E. A., L. O. Mearns, and W. E. Easterling, 2001b: Investigating the effect of seasonal plant growth and development in three-dimensional atmospheric simulations. Part II: Atmospheric response to crop growth and development. J. Climate, 14 , 711–729.
van den Hurk, B. J. J. M., P. Viterbo, and S. O. Los, 2003: Impact of leaf area index seasonality on the annual land surface evaporation in a global circulation model. J. Geophys. Res.,108, 4191, doi:10.1029/2002JD002846.
White, M. A., S. W. Running, and P. E. Thornton, 1999: The impact of growing-season length variability on carbon assimilation and evapotranspiration over 88 years in the eastern U.S. deciduous forest. Int. J. Biometeor, 42 , 139–145.
Wilson, K. B., and D. D. Baldocchi, 2000: Seasonal and interannual variability of energy fluxes over a broadleaved temperate deciduous forest in North America. Agric. For. Meteor, 100 , 1–18.
Observed and 20-yr average simulated monthly LAI for northern Europe (60°–70°N, 5°–45°E; simulated as predominantly broadleaf deciduous forest); central Canada (51°–61°N, 102°–92°W; simulated as predominantly needleleaf evergreen forest); and eastern China (35°–45°N, 100°–130°E; simulated as predominantly broadleaf deciduous forest). The simulated values are from simulation P, described as a prognostic vegetation simulation using the model's default time scales of leaf emergence. Observed LAI is derived from satellite observations (Bonan et al. 2002a)
Citation: Journal of Climate 17, 23; 10.1175/3218.1
Observed and 20-yr average simulated monthly LAI for northern Europe (60°–70°N, 5°–45°E; simulated as predominantly broadleaf deciduous forest); central Canada (51°–61°N, 102°–92°W; simulated as predominantly needleleaf evergreen forest); and eastern China (35°–45°N, 100°–130°E; simulated as predominantly broadleaf deciduous forest). The simulated values are from simulation P, described as a prognostic vegetation simulation using the model's default time scales of leaf emergence. Observed LAI is derived from satellite observations (Bonan et al. 2002a)
Citation: Journal of Climate 17, 23; 10.1175/3218.1
Observed and 20-yr average simulated monthly LAI for northern Europe (60°–70°N, 5°–45°E; simulated as predominantly broadleaf deciduous forest); central Canada (51°–61°N, 102°–92°W; simulated as predominantly needleleaf evergreen forest); and eastern China (35°–45°N, 100°–130°E; simulated as predominantly broadleaf deciduous forest). The simulated values are from simulation P, described as a prognostic vegetation simulation using the model's default time scales of leaf emergence. Observed LAI is derived from satellite observations (Bonan et al. 2002a)
Citation: Journal of Climate 17, 23; 10.1175/3218.1
The 20-yr average monthly LAI for the three regions of interest from simulations P, P/4, D, and D × 4
Citation: Journal of Climate 17, 23; 10.1175/3218.1
The 20-yr average monthly LAI for the three regions of interest from simulations P, P/4, D, and D × 4
Citation: Journal of Climate 17, 23; 10.1175/3218.1
The 20-yr average monthly LAI for the three regions of interest from simulations P, P/4, D, and D × 4
Citation: Journal of Climate 17, 23; 10.1175/3218.1
The 20-yr average (a) LAI, (b) surface air temperature, (c) snow depth, and (d) photosynthesis (abbreviated A, for assimilation) for simulations P, P/4, D × 4, and D. Also, the 20-yr average net radiation (Rnet), SH, LH, ground heat flux (G), and SM for (e) simulation P, (f) simulation P/4, (g) simulation D × 4, and (h) simulation D. Variables are averaged over vegetated land points of the northern European region shown in Fig. 1. Variables are aligned with day of first leaf at the origin as opposed to 1 Jan
Citation: Journal of Climate 17, 23; 10.1175/3218.1
The 20-yr average (a) LAI, (b) surface air temperature, (c) snow depth, and (d) photosynthesis (abbreviated A, for assimilation) for simulations P, P/4, D × 4, and D. Also, the 20-yr average net radiation (Rnet), SH, LH, ground heat flux (G), and SM for (e) simulation P, (f) simulation P/4, (g) simulation D × 4, and (h) simulation D. Variables are averaged over vegetated land points of the northern European region shown in Fig. 1. Variables are aligned with day of first leaf at the origin as opposed to 1 Jan
Citation: Journal of Climate 17, 23; 10.1175/3218.1
The 20-yr average (a) LAI, (b) surface air temperature, (c) snow depth, and (d) photosynthesis (abbreviated A, for assimilation) for simulations P, P/4, D × 4, and D. Also, the 20-yr average net radiation (Rnet), SH, LH, ground heat flux (G), and SM for (e) simulation P, (f) simulation P/4, (g) simulation D × 4, and (h) simulation D. Variables are averaged over vegetated land points of the northern European region shown in Fig. 1. Variables are aligned with day of first leaf at the origin as opposed to 1 Jan
Citation: Journal of Climate 17, 23; 10.1175/3218.1
Means and/or trends per day of 20-yr averages of the following simulated variables: calendar day of first leaf, LAI, surface air temperature (T ), and surface air temperature when first leaf occurs after snowmelt. Except for the calendar day of first leaf, the means and trends are calculated over the three weeks preceding and following first leaf, excluding the first week immediately before and after leaf emergence (i.e., for days 8–28 before and after first leaf )
The National Center for Atmospheric Research is sponsored by the National Science Foundation.
