R ES E A RC H | R E PO R TS

RE FE RENCES AND N OT ES

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

12.

13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

M.-H. Chiu et al., ACS Nano 8, 9649–9656 (2014). Y. Gong et al., Nat. Mater. 13, 1135–1142 (2014). S.-H. Su et al., Small 10, 2589–2594 (2014). X. Duan et al., Nat. Nanotechnol. 9, 1024–1030 (2014). C. Huang et al., Nat. Mater. 13, 1096–1101 (2014). Y.-H. Lee et al., Adv. Mater. 24, 2320–2325 (2012). J.-K. Huang et al., ACS Nano 8, 923–930 (2014). Y. Shi et al., Nano Lett. 12, 2784–2791 (2012). O. L. Krivanek et al., Nature 464, 571–574 (2010). Y. R. Shen, Annu. Rev. Phys. Chem. 40, 327–350 (1989). J. F. McGilp, D. Weaire. C. H. Patterson, Eds., Epioptics: Linear and Nonlinear Optical Spectroscopy of Surfaces and Interfaces (Springer-Verlag GmbH, Berlin, 1995). G. A. Reider, T. F. Heinz, in Photonic Probes of Surfaces: Electromagnetic Waves, P. Halevi, Eds. (Elsevier Science Limited, Amsterdam, 1995), vol. 2, chap. 9. N. Kumar et al., Phys. Rev. B 87, 161403 (2013). H. Zeng et al., Sci. Rep. 3, 1608 (2013). Y. Li et al., Nano Lett. 13, 3329–3333 (2013). W.-T. Hsu et al., ACS Nano 8, 2951–2958 (2014). Y. Y. Hui et al., ACS Nano 7, 7126–7131 (2013). L. Yang et al., Sci. Rep. 4, 5649 (2014). H. J. Conley et al., Nano Lett. 13, 3626–3630 (2013). C. R. Zhu et al., Phys. Rev. B 88, 121301 (2013). A. Castellanos-Gomez et al., Nano Lett. 13, 5361–5366 (2013). Q. Ji et al., Nano Lett. 13, 3870–3877 (2013). T. Heine, Acc. Chem. Res. 48, 65–72 (2015). H. Fang et al., Proc. Natl. Acad. Sci. U.S.A. 111, 6198–6202 (2014). M.-H. Chiu et al., Determination of band alignment in the single layer MoS2/WSe2 heterojunction. Nature Commun. (2015); available at http://arxiv.org/abs/1406.5137).

528

31 JULY 2015 • VOL 349 ISSUE 6247

26. H. R. Gutiérrez et al., Nano Lett. 13, 3447–3454 (2013). 27. H.-M. Li et al., Nat. Commun. 6, 6564 (2015). 28. A. Pospischil, M. M. Furchi, T. Mueller, Nat. Nanotechnol. 9, 257–261 (2014). ACKN OWLED GMEN TS

L.-J.L. acknowledges support from King Abdullah University of Science and Technology (Saudi Arabia), Ministry of Science and Technology (MOST) and Taiwan Consortium of Emergent Crystalline Materials (TCECM), Academia Sinica (Taiwan), and AOARD-134137 (USA). Y.-C.L. and K.S. acknowledge support from the Japan Science and Technology Agency research acceleration

program. W.-H.C. acknowledges the support from TCECM, MOST of Taiwan under grant NSC102-2119-M-009-002-MY3 and the Center for Interdisciplinary Science of Nation Chiao Tung University. SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/349/6247/524/suppl/DC1 Materials and Methods Figs. S1 to S11 References (29–38) 30 April 2015; accepted 24 June 2015 10.1126/science.aab4097

FOREST ECOLOGY

Pervasive drought legacies in forest ecosystems and their implications for carbon cycle models W. R. L. Anderegg,1,2* C. Schwalm,3,4 F. Biondi,5 J. J. Camarero,6 G. Koch,3 M. Litvak,7 K. Ogle,8 J. D. Shaw,9 E. Shevliakova,10 A. P. Williams,11 A. Wolf,1 E. Ziaco,5 S. Pacala1 The impacts of climate extremes on terrestrial ecosystems are poorly understood but important for predicting carbon cycle feedbacks to climate change. Coupled climate–carbon cycle models typically assume that vegetation recovery from extreme drought is immediate and complete, which conflicts with the understanding of basic plant physiology. We examined the recovery of stem growth in trees after severe drought at 1338 forest sites across the globe, comprising 49,339 site-years, and compared the results with simulated recovery in climate-vegetation models. We found pervasive and substantial “legacy effects” of reduced growth and incomplete recovery for 1 to 4 years after severe drought. Legacy effects were most prevalent in dry ecosystems, among Pinaceae, and among species with low hydraulic safety margins. In contrast, limited or no legacy effects after drought were simulated by current climate-vegetation models. Our results highlight hysteresis in ecosystem-level carbon cycling and delayed recovery from climate extremes.

A

nthropogenic climate change is projected to alter both climate mean conditions and climate variability, leading to more frequent and/or intense climate extremes such as heat waves and severe drought (1). Increasing variability is likely to profoundly affect ecosystems, as many ecological processes are more

1

Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ 08544, USA. 2Department of Biology, University of Utah, Salt Lake City, UT 84112, USA. 3 Center for Ecosystem Science and Society, Northern Arizona University, Flagstaff, AZ 86011, USA. 4School of Earth Sciences and Environmental Sustainability, Northern Arizona University, Flagstaff, AZ 86011, USA. 5DendroLab and Graduate Program of Ecology, Evolution, and Conservation Biology, University of Nevada–Reno, Reno, NV 89557, USA. 6 Instituto Pirenaico de Ecología, Consejo Superior de Investigaciones Científicas, Avda. Montañana 1005, 50192 Zaragoza, Spain. 7Department of Biology, University of New Mexico, Albuquerque, NM 87131, USA. 8School of Life Sciences, Arizona State University, Tempe, AZ 85287-4501, USA. 9Rocky Mountain Research Station, U.S. Forest Service, Ogden, UT 84401, USA. 10National Oceanic and Atmospheric Administration (NOAA) Geophysical Fluid Dynamics Laboratory, Princeton, NJ 08540, USA. 11Lamont-Doherty Earth Observatory of Columbia University, 61 Route 9W, Palisades, NY 10964, USA. *Corresponding author. E-mail: [email protected]

sensitive to climate extremes than to changes in mean states (2–4). In turn, the impacts of these extremes can have major effects on ecosystemlevel carbon cycling, feeding back to accelerate or limit climate change. The 2003 European heat wave, for example, led to the development of a strong anomalous carbon source, reversing 4 years of carbon uptake by terrestrial ecosystems on a continental scale (5). Forest ecosystems store nearly half of the carbon found in terrestrial ecosystems (6), but the fate of forests under climate change and with increasing climate extremes remains uncertain and controversial. Whereas some studies contend that large regions of forest are poised on the verge of collapse into an alternate state (7–9), others suggest that forests are relatively resilient and likely to experience only modest changes (10–12). The sensitivity of forests to climate extremes has become apparent in global patterns of widespread forest mortality (13), which highlight the possibility that the forest carbon sink could be weakened or could even transition rapidly to a carbon source in some regions (13–15). Thus, the response of forest growth and mortality to extreme drought and heat constitutes a sciencemag.org SCIENCE

Downloaded from www.sciencemag.org on August 10, 2015

at about 0.9 V under forward bias (fig. S10). A photovoltaic effect with an open-circuit voltage Voc of 0.22 V and short-circuit current Isc of 7.7 pA under white light illumination (power density Ew of 1 mW/cm2) is shown in the inset of Fig. 4C. The nearly symmetric I-V curves and barely photovoltaic effect for individual WSe2 (contacted with Pd) and MoS2 (contacted with Ti/Au) in Fig. 4D corroborate that the p-n junction from the heterostructure is predominant, rather than the small Schottky barriers between metal and TMDCs. We calculated the power conversion efficiency (PCE) of the device with the photon-to-electron conversion equation, PCE = ISCVOCFF/EWAC, where FF is the fill factor and AC is the effective area with energy conversion. The FF of 0.39 was extracted from the inset of Fig. 4C. The small FF might result from the high equivalent series resistance of the intrinsic TMDC layers. We estimated the maximum AC by considering both the depletion area of the junction and the adjacent diffusion area of each TMDC layers, giving rise to a maximum area about 32 mm2 (see the supplementary materials for details). The calculated PCE is at least 0.2%, comparable with the fewlayer MoS2 vertical p-n junction (27) and monolayer lateral WSe2 p-n junction (28). The p-n junction of WSe2-MoS2 is further corroborated with the results from another device (fig. S11), where the carrier transport is clearly through the heterojunction interface. The presence of depletion width (320 nm), rectifying behaviors, photoresponses, and photovoltaic effects confirms the intrinsic p-n junction properties for lateral WSe2-MoS2. The realization of controlled-edge epitaxy of TMDC opens the door to construct other monolayer components for future monolayer electronics.

Obs - pred growth

RE S EAR CH | R E P O R T S

0.05

0.05

0.05

0.00

0.00

0.00

-0.05

-0.05

-0.05

-0.10

-0.10

-0.10

-0.15

-0.15

-0.20

-0.15 Arid Wet

-0.20 1

2

3

Years post drought

4

1

2

3

-0.20 4

Years post drought

Fig. 1. Legacy effects are substantial and persist for 3 to 4 years. Legacy effects are quantified as the difference between observed and predicted growth (unitless index) after a 2-SD dry anomaly in the climatic water deficit (drought). (A) Legacy effects observed across all 1338 tree-ring chronologies (dashed line) and across 695 tree-ring chronologies at sites that correlate significantly with the climatic water deficit (solid line and red shaded region). (B) Legacy

1

2

3

4

Years post drought

effects at sites from among the above 695 that were categorized as arid (mean annual precipitation <500 mm) or wet (mean annual precipitation >1000 mm). (C) Legacy effects at sites from among the above 695 that support either of the two main families represented, Pinaceae and Fagaceae. Shaded regions in all panels represent the 95% confidence interval around the mean from bootstrapping (n = 5000 resamplings).

Fig. 2. Legacy effects are most prevalent in the southwestern and midwestern United States and parts of northern Europe. Legacy effects are quantified as the difference between observed and predicted growth (unitless index) after a 2-SD dry anomaly in the climatic water deficit across 1338 sites. (A) Site-level legacy effect summed over the first 4 years after drought. (B) Average correlation between tree growth (ring-width index) and the climatic water deficit (soil moisture from 0 to 100 cm minus potential evapotranspiration).

SCIENCE sciencemag.org

Fagaceae Pinaceae

large uncertainty in projections of terrestrial carbon cycle feedbacks (16). The treatment of drought in carbon cycle models is limited by a lack of representation of ecosystem response dynamics, such as recovery after drought and the potential for legacies or hysteresis—dynamics that are probably critical to predicting the future behavior of the system (17, 18). For example, lags in precipitation, particularly in semi-arid regions, have been shown to be important in the interannual variability of the land carbon sink (19). In current climate– carbon cycle models, plant physiological recovery from drought is often assumed to be complete and relatively fast. This is at odds with current understanding of physiological mechanisms in many ecosystems, particularly those with long-lived individual plants. Legacy effects and hysteresis after drought have been documented in stomatal conductance (20, 21), wood anatomy and density (22), xylem vulnerability to drought (23), drought-induced tree mortality (24, 25), and aboveground primary productivity (21, 26). As a biological legacy, the dynamics of recovery from severe drought can have a major influence on an ecosystem’s vulnerability to subsequent drought events, particularly if the drought return interval is shorter than the recovery time (17). The rate of recovery—for example, in the reestablishment of hydraulic function after drought— is largely unknown for the vast majority of tree species (24). We tested the occurrence, prevalence, and magnitude of legacy effects after severe drought using tree growth (i.e., tree-ring width) standlevel chronologies from 1338 sites across the globe, primarily in Northern Hemisphere extratropical forest ecosystems, which collectively represented 49,339 site-years. We selected tree-ring master chronologies (typically of 10 to 20 trees per site) from the International Tree Ring Data 31 JULY 2015 • VOL 349 ISSUE 6247

529

R ES E A RC H | R E PO R TS

530

31 JULY 2015 • VOL 349 ISSUE 6247

Fig. 3. Higher legacy 1.0 effects are associated with species 0.5 with low hydraulic safety margins. Integrated legacy 0.0 effects are quantified as the difference between observed and -0.5 predicted growth (unitless index) after a 2-SD dry anomaly, -1.0 summed over 1 to 4 years, averaged across all droughts within a -1.5 chronology, and aver0 1 2 3 4 5 aged across all chronologies for a given Hydraulic safety margin (MPa) species. Each point represents a species where legacies and hydraulic traits were both available. Error bars represent 1 SE. The regression line is in red.

Integrated legacy effect

Bank (27) that contained at least 25 years of data between 1948 and 2008. We defined drought legacy as a departure of observed tree growth (ring-width index) from expected growth (based on the relationship between growth and climate) in the period after a drought episode. Wood growth is ideal to test for drought legacy effects, because it provides a long temporal record and has major implications for the carbon cycle. Wood is a carbon pool with slow turnover that stores immense amounts of ecosystem carbon (6), and wood growth is tightly correlated with net primary productivity (28). We further examined the extent to which observed legacy effects are simulated in current climate-vegetation models from the Coupled Model Intercomparison Project, Phase 5 (CMIP5). We asked: (i) Are legacy effects after extreme drought pervasive in tree growth? (ii) Are legacy effects more prominent in wet or dry environments? (iii) Do legacy effects vary among species with different hydraulic safety margins (29) [a measure of how closely a tree approaches catastrophic damage to its xylem during drought (25)]? (iv) Are the legacy effects simulated in CMIP5 climatevegetation models similar to those observed in tree rings? We quantified legacy effects in tree-ring width chronologies using two methods: (i) the departure of observed from predicted growth recovery after drought based on correlations with climate and (ii) partial autocorrelation coefficients. We focused primarily on sites where ring-width anomalies exhibited significant correlations (r > 0.3; mean correlation r = 0.51) with drought [climatic water deficit (30)], because our aim was to quantify the duration of growth suppression or enhancement after drought episodes. We found significant legacies in radial growth after severe drought (>2 SD from the mean climatic water deficit) that lasted 2 to 4 years (Fig. 1A and fig S1). These effects were substantial in magnitude: a ~9% decrease in observed versus predicted growth in year 1 and 5% in year 2 after drought (Fig. 1A). Legacy effects were observed regardless of the minimum climate correlation cutoff (fig. S1) or the drought variable used (figs. S2 and S3). Legacy effects were also observed in the partial autocorrelation analysis (fig S4). There did not appear to be a strong link between the magnitude of the legacy effect and the peak intensity of the observed drought [coefficient of determination (R2) = 0.01, P = 0.08] (fig S5). Legacy effects were most pronounced in arid ecosystems (Fig. 1B). Mean annual precipitation was the only significant predictor of the magnitude of drought legacy effects in tree growth; it explained a low proportion of the variance (R2 = 0.05, P = 0.0003) (fig. S6). Correlations with mean annual temperature and potential evapotranspiration were both insignificant (P > 0.05). Strong legacy effects also tended to occur in semi-arid regions in the Northern Hemisphere (Fig. 2A) and where correlations between growth and drought were higher (Fig. 2B). Tree-ring chronologies in the southwestern and midwestern United States and in parts of northern Europe exhibited par-

ticularly strong legacy effects (Fig. 2A). Positive legacy effects, where observed growth was higher than predicted after drought, were most frequent in California and the Mediterranean region (Fig. 2A). Gymnosperms exhibited legacy effects that were slightly but significantly larger (in terms of magnitude and duration) than those exhibited by angiosperms (t = 2.25, P = 0.02) (fig. S7). Among families, Pinaceae (pines) and Fagaceae (mostly oaks) were best represented in the data set, accounting for >90% of chronologies analyzed. Pines exhibited substantially larger legacies than did oaks (Fig. 1C). Although pines were typically found at drier locales than oaks (average mean annual precipitation for pines = 660 mm/year; average for oaks = 760 mm/year), a model allowing for interactions between precipitation and family was highly significant (t = 2.55, P = 0.01), indicating that such interactions were important. Both wet and dry pine sites exhibited strong negative legacy effects, whereas wet oak sites exhibited slightly negative legacy effects, and dry oak sites had strong positive legacy effects (fig. S8). Pines also had stronger negative legacy effects than the other main gymnosperm family in the database, Cupressaceae (fig. S9). This result is consistent with Cupressaceae’s generally higher drought tolerance relative to Pinaceae (31) and is supportive of a hydraulic damage mechanism underlying legacy effects. Several physiological mechanisms may underlie the observed legacy effects of reduced growth after drought. Loss of leaf area and/or stored nonstructural carbohydrates during drought may impair growth in subsequent years (25). Pest and pathogen impacts may lag drought or accumulate in drought-stressed trees, thereby lowering growth rates (25). Finally, stress-induced shifts in xylem anatomy and associated vulnerability to hydraulic dysfunction, or remnants of droughtinduced xylem cavitation, could impair water transport and, therefore, growth (25). Although data that could test the first two hypotheses are

not available, testing the third hypothesis is possible with an existing global hydraulic trait database (29). We found that species with lower hydraulic safety margins, defined as the water potential (Y) at which 50% conductivity is lost minus the minimum measured water potential (Y50 – Ymin), exhibited larger legacy effects (R2 = 0.33, F = 4.95, P = 0.04) (Fig. 3 and table S1). This indicates that the species most at risk of hydraulic damage are also those that have the slowest growth recovery after drought. Previous studies at individual sites have observed drought-induced shifts in plant hydraulics, especially in the first 3 to 4 years after drought in oaks and poplars (22, 25), and our results generalize these findings across many taxonomic groups and a broad geographic range. The CMIP5 models captured few to no detectable legacy effects from severe drought in grid cells where the tree-ring chronologies were located (Fig. 4). In many cases, interannual variability of wood carbon growth was low and more weakly correlated with water limitation or drought (mean correlations of R = 0.01 to 0.09) than were the observed tree-ring widths at the same locations (mean correlation R = 0.25). Only the Geophysical Fluid Dynamics Laboratory Earth System Model 2G (GFDL ESM2G) exhibited significant legacy effects of 1 to 2 years (Fig. 4A), and these were of lower magnitude than the observed legacies (Fig. 1A). GFDL ESM2G and CanESM (Canadian Centre for Climate Modeling and Analysis Second Generation Earth System Model) both use a dynamic carbon allocation scheme, but they use different approaches to allocate carbon, particularly under drought conditions. GFDL ESM2G’s scheme (32) is based on the pipe model for the relationship between sapwood area and leaf area (33) and allows drought-induced loss of living carbon, including from the sapwood pool, which may allow it to capture legacy effects. Most CMIP5-class models use constant fractional allocation among the vegetation pools and do not simulate plant hydraulic damage during drought, and these appears to sciencemag.org SCIENCE

RE S EAR CH | R E P O R T S

Obs - pred growth

0.05

0.05

NorESM

0.00

0.00

0.00

-0.05

-0.05

-0.05

-0.10

-0.10

-0.10

-0.15

-0.15

-0.15

-0.20

-0.20

-0.20

1 0.05

Obs - pred growth

0.05

GFDL ESM2G

2

3

4

5

6

1 0.05

CanESM

2

3

4

5

6

1 0.05

BCC-CSM

0.00

0.00

0.00

-0.05

-0.05

-0.05

-0.10

-0.10

-0.10

-0.15

-0.15

-0.15

-0.20

-0.20

-0.20

1

2

3

4

5

6

1

2

3

CESM

4

5

6

2

3

4

5

6

4

5

6

Had-GEM

1

2

3

Years post drought Fig. 4. Legacy effects after drought are not captured in predictions of woody biomass by Earth system models. (A to F) Legacy effects after a 2-SD dry anomaly in grid cells that correlated significantly with drought and overlapped with the real-world locations of the 1338 tree-ring chronologies. Shaded regions represent the 95% confidence interval around the mean from bootstrapping (n = 5000 resamplings). Models used included GFDL ESM2G (A), the Norweigan Earth System Model (NorESM) (B), the Community Earth System Model (CESM) (C), CanESM (D), the Beijing Climate Center Climate System Model (BCC-CSM) (E), and the Hadley Centre Global Environmental Model (Had-GEM) (F).

be crucial limitations to capturing legacy effects of drought. The response of terrestrial ecosystems to drought has been reported to be one of the largest uncertainties in the carbon cycle (34) and is not well represented in current climate-vegetation models, as evidenced by our model-data comparison. Current models lack representation of some basic physiological and structural properties of plants, such as the vulnerability of xylem transport to hydraulic water stress, that lead to growth suppression, legacy effects, and drought-induced mortality (35). Mortality is generally not measured or reported at these sites, so our analysis does not examine drought-induced mortality; however, mortality or canopy dieback of surrounding trees could generate some of the positive legacy effects in surviving trees that we observed via increased resource availability. Although the impacts of climate extremes on plant mortality and species turnover will also influence carbon cycling (14), we detected a strong, pervasive, and previously undocumented legacy effect of drought on tree growth, especially in dry regions. That is, even when climatic conditions return to normal, surviving trees do not recover their expected growth rates for an average of 2 to 4 years. Given that (i) woody plant growth is a central component of carbon storage and often correlated with productivity and (ii) semi-arid regions SCIENCE sciencemag.org

play a prominent role in the variability of the global carbon cycle (19), these legacy effects have potential ramifications for the interannual variability of ecosystem-level carbon cycling and for long-term carbon storage. For example, a simple conservative estimate based on forests in the southwestern United States revealed that legacy effects could lead to 3% lower carbon storage in semi-arid ecosystems over a century, equivalent to 1.6 metric gigatons of carbon when considering all semi-arid ecosystems across the globe (30). Drought could lead to changes in carbon allocation by trees, with less being allocated to bole growth and more to roots or leaves (36, 37), which would mean that growth declines might not immediately reflect decreases in carbon uptake by forests. The fast turnover of leaves and roots, however, would still result in overall decreases in ecosystem-level carbon storage relative to ecosystems without legacy effects (37). The prominence of legacy effects in tropical forests, where tree-ring analyses are challenging, is a major remaining question. There are some indications of legacy effects in the Amazon rainforest in satellite (38) and time-series inventory plot analyses (39) after the severe 2005 and 2010 droughts. The lack of legacy effects in CMIP5 models indicates that drought impacts and their effect on carbon cycling are not accurately captured. These

findings reveal the critical roles of contingency and hysteresis in ecosystem response after climate extremes. REFERENCES AND NOTES

1. Intergovernmental Panel on Climate Change, in Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, T. F. Stocker et al., Eds. (Cambridge Univ. Press, Cambridge, 2013), pp. 3–29. 2. D. R. Easterling et al., Science 289, 2068–2074 (2000). 3. D. Medvigy, S. C. Wofsy, J. W. Munger, P. R. Moorcroft, Proc. Natl. Acad. Sci. U.S.A. 107, 8275–8280 (2010). 4. M. D. Smith, J. Ecol. 99, 651–655 (2011). 5. P. Ciais et al., Nature 437, 529–533 (2005). 6. G. B. Bonan, Science 320, 1444–1449 (2008). 7. P. M. Cox et al., Theor. Appl. Climatol. 78, 137–156 (2004). 8. M. Hirota, M. Holmgren, E. H. Van Nes, M. Scheffer, Science 334, 232–235 (2011). 9. A. C. Staver, S. Archibald, S. A. Levin, Science 334, 230–232 (2011). 10. Y. Malhi et al., Proc. Natl. Acad. Sci. U.S.A. 106, 20610–20615 (2009). 11. C. Huntingford et al., Nat. Geosci. 6, 268–273 (2013). 12. G. E. Ponce Campos et al., Nature 494, 349–352 (2013). 13. C. D. Allen et al., For. Ecol. Manage. 259, 660–684 (2010). 14. W. A. Kurz et al., Nature 452, 987–990 (2008). 15. O. L. Phillips et al., Science 323, 1344–1347 (2009). 16. S. Sitch et al., Glob. Change Biol. 14, 2015–2039 (2008). 17. V. P. Gutschick, H. BassiriRad, New Phytol. 160, 21–42 (2003).

31 JULY 2015 • VOL 349 ISSUE 6247

531

R ES E A RC H | R E PO R TS

18. J. A. Arnone III et al., Nature 455, 383–386 (2008). 19. A. Ahlström et al., Science 348, 895–899 (2015). 20. L. Virlouvet, M. Fromm, New Phytol. 205, 596–607 (2015). 21. K. Ogle et al., Ecol. Lett. 18, 221–235 (2015). 22. L. Corcuera, J. J. Camarero, E. Gil-Pelegrín, Trees (Berlin) 18, 83–92 (2004). 23. U. G. Hacke, V. Stiller, J. S. Sperry, J. Pittermann, K. A. McCulloh, Plant Physiol. 125, 779–786 (2001). 24. W. R. L. Anderegg, J. A. Berry, C. B. Field, Trends Plant Sci. 17, 693–700 (2012). 25. W. R. L. Anderegg et al., Glob. Change Biol. 19, 1188–1196 (2013). 26. Y. Zhang et al., J. Geophys. Res. Biogeosci. 118, 148–157. 27. H. D. Grissino-Mayer, H. C. Fritts, Holocene 7, 235–238 (1997). 28. D. A. Clark et al., Ecol. Appl. 11, 356–370 (2001). 29. B. Choat et al., Nature 491, 752–755 (2012). 30. Materials and methods are available as supplementary materials on Science Online. 31. T. J. Brodribb, S. A. McAdam, G. J. Jordan, S. C. Martins, Proc. Natl. Acad. Sci. U.S.A. 111, 14489–14493 (2014).

32. E. Shevliakova et al., Glob. Biogeochem. Cycles 23, GB2022 (2009). 33. K. Shinozaki, K. Yoda, K. Hozumi, T. Kira, Jap. J. Ecol. 14, 97–105 (1964). 34. M. Reichstein et al., Nature 500, 287–295 (2013). 35. T. L. Powell et al., New Phytol. 200, 350–365 (2013). 36. R. Dybzinski, C. Farrior, A. Wolf, P. B. Reich, S. W. Pacala, Am. Nat. 177, 153–166 (2011). 37. C. E. Farrior, R. Dybzinski, S. A. Levin, S. W. Pacala, Am. Nat. 181, 314–330 (2013). 38. S. Saatchi et al., Proc. Natl. Acad. Sci. U.S.A. 110, 565–570 (2013). 39. R. J. W. Brienen et al., Nature 519, 344–348 (2015). ACKN OWLED GMEN TS

Funding for this research was provided by NSF (grant no. DEB EF-1340270). W.R.L.A. was supported in part by a NOAA Climate and Global Change Postdoctoral Fellowship, administered by the University Corporation for Atmospheric Research. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the opinions or policies of the funding agencies. All tree-ring data are available at

GLOBAL WARMING

Recent hiatus caused by decadal shift in Indo-Pacific heating Veronica Nieves,1,2* Josh K. Willis,2 William C. Patzert2 Recent modeling studies have proposed different scenarios to explain the slowdown in surface temperature warming in the most recent decade. Some of these studies seem to support the idea of internal variability and/or rearrangement of heat between the surface and the ocean interior. Others suggest that radiative forcing might also play a role. Our examination of observational data over the past two decades shows some significant differences when compared to model results from reanalyses and provides the most definitive explanation of how the heat was redistributed. We find that cooling in the top 100-meter layer of the Pacific Ocean was mainly compensated for by warming in the 100- to 300-meter layer of the Indian and Pacific Oceans in the past decade since 2003.

I

t has been widely established that Earth is absorbing more energy from the Sun than it is radiating back to space (1). Furthermore, this has been attributed to anthropogenic greenhouse gases (2). Although global surface temperatures have risen over the previous century, several recent papers have documented a slowdown in the rate of surface warming since 2003 (3). Most efforts to explain this surface temperature “hiatus” have suggested that it is compensated for by more rapid warming at deeper levels in the ocean in either the Pacific (3–6) or Atlantic (7) Oceans or a combination of the Southern, Atlantic, and Indian Oceans (8). More recently, a study pointed out that heat is piling up in the depths of the Indian Ocean (9). These studies suggest that the net rate of ocean heat uptake has continued unabated and that the surface hiatus signature is due to an internal rearrangement of heat within the ocean between the surface and 1

Joint Institute for Regional Earth System Science and Engineering, University of California, Los Angeles, CA, USA. Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA.

2

*Corresponding author. E-mail: [email protected]

532

31 JULY 2015 • VOL 349 ISSUE 6247

some deeper layer of the ocean. This has implications for the net radiative forcing of Earth, because the ocean is the dominant reservoir for storage of excess heat on time scales longer than 1 year, and ocean heat content increases should approximately equal the net radiative imbalance at the top of the atmosphere (1). An alternative hypothesis is that approximately half of the surface hiatus is caused by changes in solar and stratospheric aerosol forcing (10). In other words, half of the hiatus signal is caused by reduced uptake of heat by the ocean, as opposed to an internal redistribution of heat. A vigorous debate has grown over this subject, but it has not yet been informed by comprehensive analysis of the available data over the past two decades. We considered both ocean observational data and widely used reanalysis products (that is, numerical simulations constrained by ocean and sometimes atmospheric observations) that span both of the previous decades. A recent study based on Argo data pointed out that net warming continued unabated, despite the exchange of heat between the top 100-m layer and the thermocline on interannual time scales. (11). It did

www.ncdc.noaa.gov/data-access/paleoclimatology-data/datasets/ tree-ring. We thank all data contributers at the International TreeRing Data Bank. All CMIP5 data are available at http://cmip-pcmdi. llnl.gov/cmip5/data_portal.html. We acknowledge the World Climate Research Programme's Working Group on Coupled Modelling, which is responsible for CMIP, and we thank the climate modeling groups (listed in the supplementary materials) for producing and making available their model output. For CMIP, the U.S. Department of Energy's Program for Climate Model Diagnosis and Intercomparison provides coordinating support and led the development of software infrastructure in partnership with the Global Organization for Earth System Science Portals. SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/349/6247/528/suppl/DC1 Materials and Methods Figs. S1 to S10 Table S1 References (40–67) 24 March 2015; accepted 3 July 2015 10.1126/science.aab1833

not, however, consider the redistribution of heat between these layers on decadal time scales, its geographic distribution, or how it might differ before and after the start of the hiatus in 2003. Our analysis indicates that during the most recent decade, cooling in the top 100-m layer of the Pacific Ocean is compensated for by warming in the 100- to 300-m layer of the Western Pacific and Indian Oceans, with the largest contribution in the tropics. The Southern Ocean plays a secondary role in warming the 100- to 300-m layer, but this warming has been steady over both of the past decades. The Atlantic Ocean does show a switch from warming to cooling, but its area is so small that it cannot meaningfully contribute to the hiatus signal in surface temperature over the past decade (figs. S1 and S2). Finally, we find little evidence for any change in warming rates below 700 m between the past decade and the previous one—or that the net ocean heat uptake has slowed in the most recent decade. Observed temperature trends in the oceans were estimated using objectively analyzed subsurface temperature fields from the World Ocean Atlas (WOA) (12), Ishii (13), and the Scripps Institution of Oceanography (14). Simulated temperature trends, based on reanalyses that assimilate ocean data, have also been examined. We compared observed trends with simulated trends from the Simple Ocean Data Assimilation (SODA), National Centers for Environmental Prediction Global Ocean Data Assimilation System (NCEP GODAS), and the latest European Centre for Medium-Range Weather Forecasts ocean reanalysis system 4 (ECMWF ORAS4) (15–17) (SODA, NCEP, and ECMWF). We considered global and basin-averaged temperature trends for the periods from 1993 to 2002 (the 90s) and from 2003 onward (the 00s). The periods for the past decade are slightly different depending on the product (table S1). Nevertheless, our results are insensitive to the choice of somewhat different time periods. These periods were chosen based on the assumption that the hiatus began in approximately 2003, when there was a small local maximum in the 5-year moving average of global surface sciencemag.org SCIENCE

www.sciencemag.org/content/349/6247/528/suppl/DC1

Supplementary Materials for Pervasive drought legacies in forest ecosystems and their implications for carbon cycle models W. R. L. Anderegg,* C. Schwalm, F. Biondi, J. J. Camarero, G. Koch, M. Litvak, K. Ogle, J. D. Shaw, E. Shevliakova, A. P. Williams, A. Wolf, E. Ziaco, S. Pacala *Corresponding author. E-mail: [email protected] Published 31 July 2015, Science 349, 528 (2015) DOI: 10.1126/science.aab1833 This PDF file includes: Materials and Methods Figs. S1 to S10 Table S1

Materials and Methods International Tree Ring Data Bank We downloaded all available tree ring chronologies (n=7,560) from the International Tree Ring Data Bank (ITRDB) on November 1, 2014. The ITRDB serves as a database and clearinghouse where hundreds of research groups have generously made their tree ring data freely available, along with meta-data containing the latitude, longitude, and species of the tree ring chronology (27). We restricted our analysis to only those chronologies that contained at least 25 years during 1948-2008 and removed duplicate chronologies from the same sites that measured other wood characteristics than total ring width (e.g. wood density, earlywood width, etc.). Ring-width measurements are converted to site-level chronologies of standardized ring-width indices via detrending to remove low-frequency ring-width fluctuations related to increasing tree size/age or to stand dynamics, and then averaging to obtain a single time series per species and site (27). Several types of standardization options are available, and we used the site-level chronologies generated by each research group under the assumption that those investigators selected the most suitable standardization for their data. To avoid removing key information on legacy effects, we did not select sites that had only “prewhitened” chronology files, i.e. series where an autoregressive model was fit either to the individual ring-width series or to the average site chronology to emphasize interannual variation (40, 41) (i.e. chronologies had to have detrended-only files to be included). This led to a final sample size of 1,338 chronologies. A similar approach has been implemented by other analyses of ITRDB chronologies in recent years (42–44). Drought datasets Because no single definition and metric of drought is likely to be applicable or useful in all ecosystems, we used multiple drought metrics to examine legacy effects after severe droughts. We used three monthly global precipitation datasets: (i) Climate Research Unit version 3.22 (CRU; (45)), (ii) Global Precipitation Climatology Centre (GPCC; (46)), and (iii) National Oceanographic and Atmospheric Administration Precipitation Reconstruction over Land (NOAA PREC/L; (47)). The GPCC and NOAA PREC/L datasets had geographic resolutions of 1.0°. CRU precipitation product had a geographic resolution of 0.5°, which we aggregated up to 1.0° resolution. We also utilized grids of monthly potential evapotranspiration (PET) and modeled soil moisture. PET was calculated by Sheffield et al. (2006) at 1.0° resolution (48) using the Penman-Monteith equation (49) and was accessed from on 12 December 2014. We also calculated an alternate version of PET, where interannual variability was only affected by variability in surface air temperature. In this calculation, we used the Penman-Monteith equation to recalculate PET while holding all non-temperature variables at their climatological mean annual cycles for each grid cell, based on the 1961–2000 period. This alternate version of Penman-Montheith PET was calculated as a compromise between the primary calculation, which may be inaccurate due to consideration of variables that have relatively high levels of uncertainty (vapor pressure, wind speed, solar insolation), and the overly simple Thornthwaite formulation (50), which is known to introduce 2

inaccuracies when temperature is high (e.g., (51)). Modeled soil moisture for 0–100 cm soil depth was calculated by the Noah land surface model version 3.3 (52), forced by the meteorology from the Global Land Data Assimilation System version 2 (GLDAS2; (53)). The GLDAS2 data had a geographic resolution of 0.25° and we aggregated to 1.0° resolution. We also calculated several metrics of the synthetic drought variable Climatic Water Deficit (CWD; (54)). We calculated CWD as (a) Precipitation (CRU) minus potential evapotranspiration (PET) calculated using the Penman-Monteith approach, (b) Precipitation (CRU) minus PET calculated using the Thornthwaite approach, (c) soil moisture summed over 0-100 cm minus PET (Penman-Monteith), and (d) soil moisture summed over 0-100 cm minus PET (Thornthwaite). We calculated these values monthly, and then summed them over the course of a year to get annual values. We performed our analyses on multiple drought variables and found similar qualitative patterns in all cases (e.g. Fig S3). Based on correlations between tree growth and drought variables across all 1,338 sites (Fig S2), the CWD calculated from soil moisture of 0-100 cm minus PET calculated from the full Penman-Monteith equation performed best. It had the highest mean correlation (0.254), it is immune to criticisms concerning temperature-only PET (55), and it is widely considered one of the most biologically relevant drought variables. Thus we used that drought index for all analyses presented in the main text. Calculating legacy effects and statistics We quantified the presence of legacy effects using two complementary methods: partial autocorrelation coefficients and a null model based on climate regressions. Partial autocorrelation coefficients (PACs; (56)) provide an estimate of the predictand (in our case tree-ring width) as a function of a predictor with a specified time lag, while conditioning on all smaller lags of the predictor variable. For example, the PAC of lag 3 estimates the regression parameter between yt+3 and xt, while conditioning on xt+1, xt+2, and xt+3. Thus, PACs capture the influence of time lags in regressions, and explicitly account for all shorter lags. For each site, we calculated the PACs of tree-ring chronologies as a function of the annual drought variable. The second method relied on generating a prediction of what growth post-drought should be, based on regressions between drought variables and pre-whitened tree growth. For each tree-ring chronology, we first performed pre-whitening using the default option of the dplR software package (57) to highlight interannual growth variability, and then calculated its linear regression with a drought variable (e.g. Climatic Water Deficit) over the 1948-2008 climate data period. Normalized values of the drought variable that exceeded two standard deviations were used to define a drought “event”, although other thresholds (1.5 and 2.5 standard deviations) were used and exhibited similar qualitative patterns. The regression model estimated over the entire period of overlap between the chronology and the drought variable was then used to predict tree growth after each drought event. The difference between the observed post-drought growth (with detrended-only chronologies) and the predicted post-drought growth (based on prewhitened tree growth) was used to determine the legacy effect. The usefulness of this null model for predicted growth depends on the strength of the regression between climate and growth. An uninformative null model (i.e. regression slope is not significantly different from zero) would always predict growth recovers 3

perfectly after a two standard deviation drought, even if a lower severity (e.g. one standard deviation) drought occurred in the next year. Thus, to ensure our results were not driven by uninformative null models, we restricted our analysis to sites that had a significant correlation between the drought variable and growth (r>0.3), a strong correlation (r>0.5), and a very strong correlation (r>0.7). The trade-off in this approach is a loss of sites and lower sample size with higher correlation thresholds. All correlation cut-offs exhibited legacy effects of roughly the same length, and the magnitude of the effect increased with the correlation cut-off (Fig S1). To test if potential non-drought drivers of positive autocorrelation in ring-width time series (e.g. lower frequency variation driven by age/size effects, standardization options, site dynamics, and other disturbance events) were influencing our calculations of legacy effects, we developed an additional null model. In this model, we performed exactly the same calculations to quantify legacy effects, but instead of examining growth after two standard deviation droughts, we examined growth after randomly selected years. We kept the same number of “drought events” analyzed per chronology, so that each chronology was similarly weighted as in the above analyses. We repeated this 100 times and calculated the confidence interval via bootstrapping across all sites and iterations. We found that there was a slight positive autocorrelation in the null model, but that the confidence interval overlapped zero, and the observed drought legacy effect was outside the null model interval for at least 2 years in the sites with r>0.3 presented in Fig. 1 (Fig. S10). To test if tree legacy effects were driven solely by legacy effects in climate variables (e.g. precipitation, soil moisture, etc.), we performed partial autocorrelation analysis on all drought indices. Precipitation variables exhibited no significant partial autocorrelations for our sites (PAC R<0.2, p>0.1), nor did CWD calculated from P – PET (PAC R<0.2, p>0.1). CWD from SM – PET exhibited slightly significant legacies at 1year lag (PAC R=0.25, p=0.05), but not for longer lags (PAC R<0.1, p>0.5). This indicates that legacy effects in climate, including soil moisture storage, are highly unlikely to drive the legacy effects of 3-4 years that we uncovered in tree-ring chronologies. Determinants of legacy effects We calculated mean annual precipitation (CRU), mean annual temperature (CRU), and mean annual PET (Penman-Monteith formulation) for each of the chronologies based on the 1970-2000 period. We examined the relationship between these mean climate normal and legacy effects via linear mixed models, including site as a random effect. To examine the potential interaction between family and climate, we included family as a fixed effect in the mixed models. All models met the assumptions of normality. We calculated the uncertainty around the magnitude of legacy effects through bootstrapping with N=5000 iterations. To examine whether xylem hydraulic legacies may provide a mechanistic explanation of legacy effects, as has been observed in field studies (22, 25), we used a global database of hydraulic safety margin (29). Hydraulic safety margin for a species is calculated as the water potential at which point 50% of hydraulic conductivity is lost minus the minimum observed water potential in the field: Ψ50 – Ψmin. This trait provides a general estimate of how much a species avoids “allowing” its xylem water 4

potential to approach levels that are potentially damaging and lethal by regulating Ψmin using stomatal control, leaf shedding, and other processes (58). We compared each species’ hydraulic safety margin to the average legacy effect observed for that species, averaging over all drought events and all sites where the species was sampled. All statistical analyses were performed in the R computing environment (59), using the dplR for tree ring analysis (57), lme4 for mixed model analysis (60), RNetCDF for drought variable and land surface model output calculations (61), and boot (62) packages. Wood growth in land surface models We obtained land surface model outputs from the CMIP5 multi-model ensemble archive for one realization for each of six models – GFDL-ESM2G, NorESM, CESM, CanESM, BCC-ESM, and HADGEM2-ES. Only one realization of each model is needed because legacy effects should be a product of model structure and insensitive to initial conditions, and thus present in all realizations. We downloaded the wood carbon content per unit land area, precipitation, and soil moisture from historical runs driven by all forcings during 1850-2005 (1860-2005 in GFDL ESM2G) (63). We then subsetted only those pixels in the model output where ITRDB chronologies used in our analysis were found. We then calculated legacy effects using the same null model approach detailed above, both on all sites and on those with climate correlations greater than 0.3. Of course, direct comparisons between tree–ring chronologies and land surface models are difficult because models do not represent trees directly, and instead tree growth is measured as the change in a pool of carbon in wood. In addition, using the same pixels as ITRDB sites relies on the assumption that roughly similar plant functional types are simulated in these pixels as the species that were cored for ITRDB chronologies. We believe this is a reasonable assumption because one of the primary goals and benchmarks of dynamic global vegetation models in a land surface model is to predict the spatial distribution of ecosystems. Nevertheless, because wood carbon has a long turnover time, examining sensitivity of this pool to climate and drought in these models in comparison to observed sensitivities in field-based data is still useful. Indeed, the ability of the GFDL ESM2G model to simulate moderate legacy effects, likely due to its allocation scheme, is indicative that this comparison is useful and that simple modifications in land surface models may better capture drought legacies. First-order estimate of legacy effects’ impact on ecosystem carbon storage Scaling up changes in tree ring width to ecosystem carbon storage is incredibly challenging (64, 65) and requires additional information generally not present in the ITRDB dataset (e.g. stand demography, bark depth, diameter at breast height, etc.). Thus, to provide a first-order estimate of the potential impacts of legacy effects on ecosystem storage, we performed simulations using a simple demographic model of carbon pools calibrated using data from 20 years of 2,985 forest plots in the United States Forest Service’s Forest Inventory and Analysis (FIA) for the state of Utah. Based on data from 1993-2012, which included both wet and severe drought periods, we calculated the average annual gross growth from diameter measurements, roughly equal to NPP, and mortality. We then ran 100-year simulations, a typical length for carbon cycle modeling, both with and without drought legacy effects, starting from the initial aboveground forest C stock from the FIA data. Every 16 years, we imposed a drought that decreased gross 5

growth to 25% of average. Without legacy effects, growth recovered to average the following year. With legacy effects, we applied the percentage decrease to growth shown in Fig 1b for arid ecosystems, the vast majority of these forests, during 2 years postdrought. We compared total C storage between the two cases at the end of the 100-year simulation. Assuming that this effect could occur in all semi-arid ecosystems globally (consistent with the lagged climate-productivity relationships presented in refs (19, 66)), we multiplied the fractional difference in carbon storage by current carbon storage in semi-arid ecosystems (woody savannas and savannas in (67)) to estimate that legacy effects lead to 1.6 Pg C less carbon stored in these ecosystems at the end of a century. This is likely a conservative estimate of the global potential impact, because it includes only semi-arid ecosystems (where our sampling density was highest; Fig 2), and legacy effects were also detected in other ecosystems with larger carbon storage.

6

Fig. S1.

Legacy effects are observed regardless of climate correlation of plots included in analysis. Legacy effects are observed minus predicted growth (unitless index) after a two standard deviation of the climatic water deficit drought index (soil moisture 0-100 cm minus potential evapotranspiration from Penman-Monteith). Error bars regions represent the 95% confidence interval around the mean from bootstrapping (n=5000).

7

Fig. S2.

Correlations of the different drought variables with annual ring width index. Colors indicate different types of variables. Red: temperature (MAT) only; blue: precipitation (MAP) only; brown: soil moisture (SM) ; green: Climatic Water Deficit calculated as either precipitation or soil moisture minus potential evapotranspiration from PenmanMonteith (PETa) or Thornthwaite (PETt). Numbers in upper right indicate mean correlation.

8

Fig. S3.

Legacy effects after two standard deviation drought are observed regardless of the drought index examined. (a) Drought as precipitation anomaly from CRU precipitation data, (b) Drought as Climatic Water Deficit = precipitation (CRU) minus potential evapotranspiration. (c) Drought as Climatic Water Deficit = soil moisture from 0-100 cm minus potential evapotranspiration. Black lines are all 1,338 sites and red lines are only those sites with significant (r>0.3) correlations with the drought variable.

9

Fig. S4.

Legacy effects of up to 3-4 years are observed in partial autocorrelation function coefficients as well. (a) All 1,208 chronologies and (b) the chronologies that are significantly correlated with Climatic Water Deficit. Polygons estimate the 95% confidence intervals around the mean from bootstrapping.

10

Fig. S5.

Magnitude of legacy effect is weakly associated with drought severity (here the z-score of CWD anomaly). Legacy effects are the sum of observed minus predicted growth (unitless index) for years 1-4 years after the drought event.

11

Fig. S6.

Low mean annual precipitation of a site is associated with larger legacy effects (Observed – predicted growth; unitless) after a two standard deviation drought of the climatic water deficit drought index. Each point is a site.

12

Fig. S7.

Gymnosperms experience significantly higher legacy effects than angiosperms after a two standard deviation drought of the climatic water deficit drought index (soil moisture 0-100 cm minus potential evapotranspiration from Penman-Monteith).

13

Fig. S8.

Pines exhibit legacy effects at both wet (>500 mm mean annual precipitation) and dry (<500 mm) sites while oaks exhibit slight legacy effects at wet sites and negative legacy effects at dry sites. Legacy effects are observed minus predicted growth (unitless index) after a two standard deviation drought of the climatic water deficit drought index (soil moisture 0-100 cm minus potential evapotranspiration from Penman-Monteith). Error bars regions represent the 95% confidence interval around the mean from bootstrapping (n=5000).

14

Fig. S9.

Pinaceae experiences higher legacy effects than Fagaceae or Cuppresaceae. Legacy effects after a two standard deviation drought of the climatic water deficit drought index (soil moisture 0-100 cm minus potential evapotranspiration from Penman-Monteith). Error bars regions represent the 95% confidence interval around the mean from bootstrapping (n=5000).

15

Fig. S10.

Legacy effects in a null model of randomly selected years compared to observed legacies after a two standard deviation drought of the climatic water deficit drought index (soil moisture 0-100 cm minus potential evapotranspiration from Penman-Monteith). Error bars and shaded regions represent the 95% confidence interval around the mean from bootstrapping (n=5000). Black dashed line is all 1,338 sites and red dashed line is the 695 sites that correlated significantly with drought (r>0.3) presented in Fig 1.

16

Table S1. Table S1: Species and regions included in the ITRDB analysis that experienced a drought during the time period studied (1948-2008). Genus Species Regions Abies concolor western North America Abies magnifica western North America Abies pindrow Asia Acer saccharum eastern North America Juniperus occidentalis western North America, eastern North America Juniperus phoenicea Asia Juniperus virginiana eastern North America Phyllocladus aspleniifolius Australia Picea abies Europe Picea engelmannii western North America Picea glauca western North America, eastern North America Picea orientalis Asia Pinus albicaulis western North America Pinus balfouriana western North America Pinus banksiana western North America Pinus cembroides western North America Pinus contorta western North America Pinus coulteri western North America Pinus echinata eastern North America Pinus edulis western North America Pinus flexilis western North America Pinus jeffreyi western North America Pinus lambertiana western North America Pinus monophylla western North America Pinus nigra Europe, Asia, eastern North America Pinus palustris eastern North America Pinus pinea Europe Pinus ponderosa western North America Pinus rigida eastern North America Pinus strobiformis western North America Pinus strobus eastern North America Pinus sylvestris Europe, Asia Pseudotsuga macrocarpa western North America, eastern North America Pseudotsuga menziesii western North America Quercus alba eastern North America, Europe Quercus douglasii western North America Quercus kelloggii western North America 17

Quercus Quercus Quercus Quercus Quercus Quercus Quercus Taxodium Tsuga

lobata lyrata macrocarpa petraea prinus robur stellata distichum canadensis

western North America eastern North America western North America, eastern North America Europe eastern North America Europe eastern North America eastern North America eastern North America

18