University of Nevada, Reno
Variation in the Bimodal Precipitation Regime of Southern Nevada, USA
Measured in Tree-Ring Isotopes
A dissertation submitted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy in Ecology, Evolution, and Conservation Biology
by
Charles Myers Truettner
Dr. Franco Biondi – Major Advisor Dr. Adam Z. Csank – Co-advisor
August, 2020
THE GRADUATE SCHOOL
We recommend that the dissertation prepared under our supervision by
Charles Myers Truettner
entitled Variation in the Bimodal Precipitation Regime of Southern Nevada, USA Measured in Tree-Ring Isotopes
be accepted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy
Franco Biondi, Ph.D. Advisor Adam Z. Csank, Ph.D. Co-advisor Michael D. Dettinger, Ph.D. Committee Member Peter J. Weisberg, Ph.D. Committee Member Simon R. Poulson, Ph.D. Graduate School Representative
David W. Zeh, Ph.D., Dean Graduate School
August 2020
i
Abstract
The North American Monsoon System (NAMS) is a seasonal climatic phenomenon vital to ecosystems and human population centers in the southwestern USA and northwestern Mexico. Inter-annual variation of the NAMS bimodal precipitation regime is more predictable in its core region of northwestern Mexico, Arizona, and New Mexico,
USA than at its northwest boundary. The NAMS onset and strength is projected to shift during the 21st century because of climate change, yet there is high uncertainty of how and when that might happen. We combined extensive field measurements of micrometeorology, stable isotopes, and tree rings from an old-growth ponderosa pine
(Pinus ponderosa) stand in southern Nevada, USA. We tested if variation in the bimodal precipitation regime could be measured in the stable isotopes of tree rings. Hourly measurements from a meteorological station adjacent to the ponderosa pine stand were associated with regional seasonal patterns of integrated water vapor transport from
2011-2017. Biweekly measurements of stable isotopes in precipitation and xylem water were related to stable isotopes in the α-cellulose of tree rings to interpret subseasonal climatic patterns for the 2015 and 2016 growing season. Finally, a novel index calculated as the difference between the δ13C of α-cellulose in the latewood and earlywood
13 (δ CDIFF = latewood – earlywood) portions of a tree ring was highly influenced by the fraction (fSUMMER) of summer precipitation (June-September) to previous cool-season
13 precipitation (October-April). The δ CDIFF is an index that measures the water-stress of a tree throughout the growing season and could assist with dendroclimatic reconstructions near the boundary of the NAMS.
ii
Acknowledgments
I am in gratitude for the enthusiasm and guidance from my dissertation committee: Dr.
Franco Biondi, Dr. Adam Csank, Dr. Simon Poulson, Dr. Michael Dettinger, and Dr.
Peter Weisberg. Dr. Emanuele Ziaco, Dr. Scott Strachan, and Nicholas Miley were influential scientifically and emotionally during my experience at University of Nevada,
Reno. Alyssa Mineau and Daniel McCready assisted in the laboratory. Finally, I thank all the professors and graduate students in the EECB, Geography, and Hydrology Graduate
Programs for their support in this interdisciplinary project.
iii
Table of Contents List of Tables…………………………………………..…………………………………….…v List of Figures………………………………………………………...…..……………………vi General Introduction……………………………...…………………...………..….……….1 References……………………………………………………………….……………3 Chapter 1: Seasonal Analysis of the 2011–2017 North American Monsoon near its Northwest Boundary……………………………………………………..……....…….……5 Abstract………………………………………………………………………..…….…5 Introduction………………………………………………………………..…….……..6 Data and Methods………………………………………………………..…….……..8 Results……………………………………………………………………….……….11 Discussion……………………………………………………………………………16 Conclusions……………………………………………………………………….….22 Acknowledgments………………………………………………………….………..23 References………………………………………………………………….………..23 Figures………………………………………………………………………….…….32 Tables………………………………………………………………………….……..38 Chapter 2: Subseasonal Ponderosa Pine Radial Growth Response to Warm- Season Precipitation amid Drought in Southern Nevada………………..…….……39
Summary……………………………………………………………………………..39
Introduction…………………………………………………………………………..40
Methods………………………………………………………………………………44
Results………………………………………………………………………………..53
Discussion……………………………………………………………………………56
Conclusion……………………………………………………………………………60
Acknowledgments…………………………………………………………...………61
References…………………………………………………………………...………61
Figures………………………………………………………………………………..71
Tables…………………………………………………………………………………76 iv
Chapter 3: Bimodal Precipitation Variation Measured in Intra-Annual Tree-Ring Isotope Chronologies from Southern Nevada, USA ……………………....…….…...77
Abstract………………………………..…………………………….……….….…….77
Introduction…………………………………...…………………………..…………..78
Methods…………………………………………………………………….…………81
Results………………………………………………………….……………….…….84
Discussion……………………………………………………………………....…….88
Conclusion…………………………………………………………………….………90
References……………………………………………………………………………91
Figures………………………………………………………………………………...98
Appendix A…………………………………………………………………….……………104
Appendix B……………………………………………………………………….…………105
Appendix C…………………………………………………………………………….……107
Appendix D……………………………………………………………………………….…111
Appendix E………………………………………………………………………….………112
Appendix F……………………………………………………………………….…………113
Appendix G………………………………………………………………………..………..115
Appendix H…………………………………………………………………….……………118
Appendix I………………………………………………………………………...…………119
v
List of Tables
Chapter 1 Table 1. Summary of seasonal and water year precipitation totals (PPT) at our study site in the Sheep Range of southern Nevada.
Chapter 2 Table 1. Diameter at Breast Height (DBH), height, and final date of lignification for tree- ring growth quartiles for the 2015 and 2016 tree rings from the twelve ponderosa pines sampled
vi
List of Figures Chapter 1 Figure 1. Interpolated pseudo-color map of Pearson’s linear correlation between July– September total precipitation at our mountain site (solid white circle) and at each 800-m PRISM grid cell in the surrounding region during 1895–2015. All correlations were statistically significant (p-value < 0.01), with lower values usually corresponding to lower elevations, such as the Las Vegas valley (black circle). Figure 2. Time series plots of daily environmental variables measured at our study site. (a) Average dewpoint temperature and total precipitation; a 9.4 °C threshold of dewpoint temperature was used to identify the arrival of North American Monsoon (NAM) precipitation. (b) Volumetric water content at two soil depths (2–17 cm and 17–32 cm) and days when snow was present (see text for details). Figure 3. Precipitation percentages at our study site for cool (1 October–31 March), early warm (1 April to the day prior to 9.4 °C dewpoint threshold), and late warm (day of 9.4 °C dewpoint threshold to 30 September) seasons of the 2012–2017 water years. Figure 4. Average diurnal cycles of weather and soil variables by season. The variability of each hourly value is shown by vertical bars corresponding to two standard errors above and below the mean. Meteorological variables include: (a) total precipitation, (b) dewpoint temperature, (c) air temperature, (d) vapor-pressure deficit, (e) and solar radiation. Soil variables are (f) volumetric water content (VWC) at 2–17 cm soil depth, and (g) VWC at 17–32 cm soil depth. Figure 5. Total weekly precipitation plotted against average weekly dewpoint temperature for early and late warm season at our study site. The larger precipitation events occurred during the late warm season as dewpoint temperature increased. Figure 6. Seasonal averages of integrated water vapor transport (IVT) for the 2011– 2017 water years. Pseudo-color shading with overlaid vectors (arrows) was used to indicate the amount of moisture and the direction of transport. Cool-season IVT, and resulting precipitation was large in 2011, 2016, and 2017, as well as for the 2015 late warm season.
Chapter 2 Figure 1. Location of the study site (2320 m) in the Sheep Range of southern Nevada, U.S.A. 35 km north-northwest of Las Vegas. The study site is included in the Nevada Climate-Ecohydrological Assessment Network (NevCAN) located in the Desert National Wildlife Refuge. Figure 2. Microscopic image of the 2015 and 2016 tree rings from a sampled ponderosa pine at the NevCAN Montane site highlighting the variation in wood anatomy including lumen diameter (LD) and cell-wall thickness (CWT). Tree rings were sliced into quartiles (Q12015 - Q42016) and classified into subseasons depending on the tree-ring phenodate. Subseasons include: 2015 Monsoon Season (MS2015), 2015 Post-Monsoon Season (PM2015), 2016 Early Warm Season (EWS2016), 2016 Monsoon Season (MS2016), and vii
2016 Post-Monsoon Season (PM2016). A false ring is present in MS2015 and separates the subseasons.
Figure 3. The δ18O VSMOW (‰) and δ2H VSMOW (‰) measurements for precipitation, xylem water (large and small trees), and soil water (10 cm depth and 20 cm depth) collected during the 2015 and 2016 growing season at the NevCAN Montane Site in the Sheep Range of southern Nevada. Precipitation was classified into five subseasons (A) and were used to plot the local meteoric water line (δ2H = 8.27*δ18O + 9.7) compared to the global meteoric water line from Rozanski et al. (1993) (δ2H = 8.13*δ18O + 10.8). The δ18O VSMOW (‰) of precipitation, stem water, and soil water δ18O VSMOW (‰) are plotted with daily precipitation for the study time period (B). Figure 4. Subseasonal measurements of ponderosa pine radial growth during the 2015 and 2016 growing season at the NevCAN Sheep Range Montane Site. Subseasons were classified as the 2015 monsoon season (MS2015), 2015 post-monsoon season (PM2015), 2016 early warm season (EWS2016), 2016 monsoon season (MS2016), and 2016 18 18 post-monsoon season (PM2016). Precipitation (δ OPPT) and xylem water (δ OXW) were collected at the field site for large (A) and small (B) trees; the δ18O in α-cellulose (δ18O- CELL ) was measured from high resolution microtome slices of the 2015 and 2016 tree for 13 18 large (C) and small (D) trees; and δ C in α-cellulose (δ OCELL ) for large (E) and small (F) trees. Subseasons with the same letter are not significantly different (p < 0.05) using the Tukey Honest Significant Differences (HSD) test for large or small trees. Asterix (*) 18 denotes significantly different (p < 0.05) subseasonal δ OCELL between large vs. small trees using Wilcox rank sum test. Figure 5. Proportion of oxygen atoms from leaf water that have exchanged with xylem water at the site of cellulose synthesis (Pex) for large (A) and small (B) trees. Subseasons that have the same letter are not significantly different (p < 0.05) using the Tukey Honest Significant Differences (HSD) test. The HSD test was restricted to differences in subseason for large or small trees. Asterix (*) denotes significantly 18 different (p < 0.05) subseasonal δ OCELL between large vs. small trees using Wilcox rank sum test.
Chapter 3 Figure 1. Location of the Sheep Range Montane site in southern Nevada, USA compared to the fraction (fSUMMER) of summer precipitation (June-September) to non- summer precipitation (previous October-May) on a regional scale. Data from PRISM dataset (1980-2010 normals).
13 Figure 2. The δ CCELL earlywood and latewood isotope chronologies for the pre- Industrial Revolution (1640 – 1862 CE) and post-Industrial revolution (1820 – 2017 CE). The pre-industrial corrections method (PIN) was applied to the post-Industrial Revolution isotope chronologies. Figure 3. Tree-ring isotope chronologies from the Sheep Range, NV, USA: (a) the 13 13 δ CCELL for earlywood and latewood compared to the δ CCELL of annual tree rings from Bale et al., (2011), (b) Isotopic reconstruction of the fraction of summer precipitation to 13 non-summer precipitation (fSUMMER*) using δ CDIFF as a paleoclimate proxy, (c) dendroclimatic reconstruction of non-summer precipitation (*non-summer) using the viii earlywood ring width index as a paleoclimate proxy. The summer precipitation reconstruction (*summer) was calculated by multiplying the fSUMMER* by non-summer*. Five-year moving averages were used to show variations in the precipitation reconstructions. Figure 4. Results from treeclim that shows the moving correlation coefficient for June- September (JJAS), July-September (JAS), June-August (JJA), and cool-season 13 13 13 precipitation in relation to a) δ CLW, b) δ CLW, and c) δ CDIFF for the instrumental period (1948-2017 CE).
13 Figure 5. Comparison of season precipitation variable to the δ CCELL proxies that 13 13 13 included earlywood δ CCELL, latewood δ CCELL, and δ CDIFF compared to (a-c) June- September precipitation (JJAS), (d-f) June-August (JJA) precipitation, (g-i) July- September precipitation (JAS), (j-l) non-summer (previous October - May) precipitation, 13 (m-o) and fSUMMER precipitation. The strongest relationship is between δ CDIFF and fSUMMER (o).
13 Figure 6. Times series of the proxy of δ CDIFF was significantly correlated with fSUMMER 13 indicating that lower δ CDIFF are associated with more summer precipitation, and vice versa. 1
General Introduction
Isotopic measurements of δ13C and δ18O in paleoclimatological proxies are used to reconstruct different seasonal precipitation regimes and climatic patterns (Szejner et al.,
2018; Freund et al., 2019). Shifts in the isotopic signature of earlywood to latewood in the α-cellulose of annual tree-ring series represent cool-season and late warm-season precipitation, respectively, in the North American Monsoon System (NAMS) core region
(Griffin et al., 2013). Belmecheri et al. (2018) measured fluctuations in δ13C and δ18O in
α-cellulose of tree rings at higher resolution than early- and latewood in the NAMS core region. This led to a better understanding of how the NAMS bimodal precipitation pattern of cool-season and late warm-season precipitation is linked with tree-ring growth of ponderosa pine (Pinus ponderosa).
Xylogenesis is a technique that records intra-annual fluctuations of tree-ring growth over an entire growing season (Rossi et al., 2006). More specifically, the lumen area and cell wall thickness of tree xylem cells are measured to calculate cellular enlargement and cell wall thickening rates. This allows for precise measurements of precipitation events on tree-ring growth at a cellular scale (Ziaco et al., 2018). Combining measurements of
δ13C and δ18O in α-cellulose with cell-wall thickening rates provides the opportunity to measure mechanisms behind tree-ring formation.
13 Drought conditions in the deserts of the southwestern USA influence the δ C in the
α-cellulose of xylem tissue in tree rings (McDowell et al., 2008). A more enriched, or heavier, δ13C corresponds to water stress conditions when stomal and hydraulic conductivity are limited during the growing season. δ13C is depleted, or lighter, when these water-stressed conditions are relieved from precipitation events that lower vapor pressure deficits and increase stomatal and hydraulic conductivity. Water-stressed 2
conditions are more common when precipitation is sparse in the early warm-season of the southwestern USA. As the warm season continues into the summer, precipitation events increase with the northern extension of the intertropical convergence zone and the arrival of afternoon monsoonal thunderstorms and large remnant storms form hurricanes (Truettner et al., 2019). However, there is great variability in the amount of precipitation and the onset of the NAMS during the late warm season in the region
(Higgins et al., 1999).
The δ18O of source water needed for tree radial growth is frequently modeled from the α-cellulose in the cell walls of xylem tissue found in annual tree rings (Roden et al.,
2000, Barbour et al., 2004). The δ18O of precipitation samples and source water extracted from the stem water of trees collected in situ at a study site measures the influence of precipitation events on δ18O of α-cellulose (Roden et al., 2005). Continental and atmospheric processes affect the δ18O of precipitation at a specific location compared to the evaporative source regions from large bodies of water. The combination of δ18O and δ2H of precipitation at a study site for multiple seasons assists in identifying the global-to-local processes driving the fluctuations of δ18O in precipitation
(Craig 1961, Rozanski et al., 1993). On a global scale, precipitation originating from warmer regions closer to the equator has an enriched, or heavier, δ18O and δ2H while precipitation from cooler regions at higher latitudes have a depleted, or lighter, δ18O and
δ2H. In the southwestern USA, precipitation linked with the cool-season is generally depleted in δ18O and falls in the form of snow or rain from Pacific frontal storms. Late warm-season precipitation associated with the NAMS core region is more enriched than cool-season precipitation and originates in lower latitudes as hydroclimatic surges from the Gulf of California and Gulf of Mexico (Vera et al, 2015). Although, local conditions like elevation or amount effect can influence the stable isotope ratio in precipitation. 3
In my dissertation, we investigated variation in the bimodal precipitation regime in the Sheep Range of southern Nevada, USA using isotopes in tree rings. In chapter 1, we used hourly meteorological data and regional scale measurements of integrated water vapor transport to define the precipitation regime of our study site from April 2011-
October 2017. In chapter 2, we compared the meteorological data to stable isotope measurements in precipitation, stem water, and the α-cellulose of tree rings for the 2015 and 2016 growing season. Finally, in chapter 3, we tested a novel index of the difference
13 13 between δ C of α-cellulose in the earlywood and latewood (δ CDIFF) of a 377-year old tree-ring chronology to the fraction (fSUMMER) of late warm-season to cool-season precipitation totals at the Sheep Range. Variation in a bimodal precipitation regime was present at the Sheep Range and can be dominated by either cool-season or late warm- season precipitation. Precipitation that fell during the summer could be related to the
NAMS, extreme precipitation events like remnant storms from hurricanes, or a combination of the two.
References
Barbour, M. M., Roden, J. S., Farquhar, G. D. and Ehleringer, J. R. (2004) Expressing leaf water and cellulose oxygen isotope ratios as enrichment above source water reveals evidence of a Péclet effect, Oecologia, 138(3), pp. 426-435. Belmecheri, S., Wright, W. E., Szejner, P., Morino, K. A., & Monson, R. K. (2018). Carbon and oxygen isotope fractionations in tree rings reveal interactions between cambial phenology and seasonal climate. Plant, cell & environment, 41(12), 2758-2772. Craig, H. (1961). Isotopic variations in meteoric waters. Science, 133(3465), 1702-1703. Freund, M. B., Henley, B. J., Karoly, D. J., McGregor, H. V., Abram, N. J., & Dommenget, D. (2019). Higher frequency of Central Pacific El Niño events in recent decades relative to past centuries. Nature Geoscience, 1. Griffin, D., Woodhouse, C. A., Meko, D. M., Stahle, D. W., Faulstich, H. L., Carrillo, C., Touchan, R., Castro, C. L. and Leavitt, S. W. (2013) North American monsoon precipitation reconstructed from tree‐ring latewood, Geophysical Research Letters, 40(5), pp. 954-958. 4
Higgins, R. W., Chen, Y., & Douglas, A. V. (1999). Interannual variability of the North American warm season precipitation regime. Journal of Climate, 12(3), 653-680. McDowell, N., Pockman, W. T., Allen, C. D., Breshears, D. D., Cobb, N., Kolb, T., ... & Yepez, E. A. (2008). Mechanisms of plant survival and mortality during drought: why do some plants survive while others succumb to drought?. New phytologist, 178(4), 719- 739. Roden, J. S., Lin, G. and Ehleringer, J. R. (2000) A mechanistic model for interpretation of hydrogen and oxygen isotope ratios in tree-ring cellulose, Geochimica et Cosmochimica Acta, 64(1), pp. 21-35. Roden, J. S., Bowling, D. R., McDowell, N. G., Bond, B. J., & Ehleringer, J. R. (2005). Carbon and oxygen isotope ratios of tree ring cellulose along a precipitation transect in Oregon, United States. Journal of Geophysical Research: Biogeosciences, 110(G2). Rossi, S., Anfodillo, T. and Menardi, R. (2006) Trephor: a new tool for sampling microcores from tree stems, Iawa Journal, 27(1), pp. 89-97. Rozanski, K., Araguás‐Araguás, L. and Gonfiantini, R. (1993) Isotopic patterns in modern global precipitation, Climate change in continental isotopic records, 78, pp. 1-36. Szejner, P., Wright, W. E., Belmecheri, S., Meko, D., Leavitt, S. W., Ehleringer, J. R., & Monson, R. K. (2018). Disentangling seasonal and interannual legacies from inferred patterns of forest water and carbon cycling using tree‐ring stable isotopes. Global change biology, 24(11), 5332-5347. Truettner, C., Dettinger, M. D., Ziaco, E. and Biondi, F. (2019) Seasonal Analysis of the 2011–2017 North American Monsoon near its Northwest Boundary, Atmosphere, 10(7), pp. 420. Vera, C., Higgins, W., Amador, J., Ambrizzi, T., Garreaud, R., Gochis, D., Gutzler, D., Lettenmaier, D., Marengo, J. and Mechoso, C. (2006) Toward a unified view of the American monsoon systems, Journal of climate, 19(20), pp. 4977-5000. Ziaco, E., Truettner, C., Biondi, F., & Bullock, S. (2018). Moisture‐driven xylogenesis in Pinus ponderosa from a Mojave Desert mountain reveals high phenological plasticity. Plant, cell & environment, 41(4), 823-836.
5
Seasonal Analysis of the 2011–2017 North American Monsoon near its Northwest
Boundary
Charles Truettner 1, Michael D. Dettinger 2,3, Emanuele Ziaco 1 and Franco Biondi 1,*
1 DendroLab, Department of Natural Resources and Environmental Science, University
of Nevada, Reno, NV 89557, USA
2 U.S. Geological Survey (Retired), Carson City, NV 89701, USA
3 Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA
92037, USA
Abstract: The seasonal extent of the North American Monsoon (NAM) is highly variable and potentially sensitive to future climate change. Our objective was to determine how regional monsoonal patterns influence mountain precipitation near the NAM northwest boundary. Among the data we analyzed, a unique opportunity was provided by hourly observations collected on the Sheep Range (2300 m asl), in the Mojave Desert of southern Nevada, during 2011–2017. Long-term 800-m Parameter-elevation
Relationships on Independent Slopes Model (PRISM) precipitation time series showed that the site is representative of mountain areas in the NAM northwest region. Based on in situ observations, we divided the water year into three seasons: cool (1 October through 31 March), early warm (1 April through last day with dewpoint <9.4°C), and late warm (first day with dewpoint ≥9.4°C through 30 September). Dewpoint temperature differed by about 8°C between early warm season (mean of -6.3°C) and late warm season (mean of 2.3°C). According to ANCOVA model results, increasing hourly dewpoint associated with afternoon thunderstorms in the late warm season had the greatest relationship with hourly precipitation (F-value = 237.8, p-value < 0.01). Except for 2016, more precipitation fell at our study site during the late than the early warm 6
season. Late warm season precipitation contributed the most (43-56%) to total water- year precipitation during the 2012–2015 extended drought. Southwestern USA regional composites of vertically integrated water vapor transport (IVT) suggested that water vapor in the cool and early warm season originated from the Pacific Ocean to the west, while a transition to a NAM-like pattern of northward IVT coincided with the late warm season.
Keywords: integrated water vapor transport; NevCAN; North American monsoon; hydrometeorology; Southern Nevada; Sheep Range; Desert National Wildlife Refuge
Introduction
Large-scale precipitation patterns are projected to shift as the Earth’s atmosphere and oceans warm during the 21st century, and monsoonal precipitation is one such pattern (Cook and Seager, 2013). The North American Monsoon (NAM: Adams and
Comrie, 1997) has been a subject of intense research in recent years, with multiple field campaigns and modeling studies (see Jana et al., 2018 for a recent review). Douglas et al. (1993) identified one of its moisture sources when winds shift northwards on the west side of the Sierra Madre Occidental, transporting water vapor along the Gulf of
California, and causing summertime precipitation that progressively moves from Central
America in May to western Mexico in June (Vera et al., 2006; Higgins et al., 1999).
Further east, moisture originating from the Gulf of Mexico and Caribbean Sea is forced to rise by the complex mountain landscape, and eventually is transported northward into the NAM region (Adams and Comrie, 1997). Some moisture may be recycled through terrestrial vegetation evapotranspiration (Hu and Dominguez, 2015), even though this hypothesis was generated from an atmospheric reanalysis, and validation field data are still lacking. Depending on the location, any combination of four major sources of 7
moisture—Gulf of California, Gulf of Mexico, land, and the Pacific Ocean—can be responsible for the bulk of summer rainfall (Jana et al., 2018).
NAM precipitation typically arrives in the southwestern USA during early July, even though onset dates, total amounts, seasonal distribution, and affected regions can vary considerably from year to year (Higgins et al., 1999). Convective thunderstorms develop over the topographically diverse terrain of the southwestern USA, driven by influxes of low-level moisture, strong surface heating, and resulting thermodynamic instability
(Adams and Souza, 2009). Deep convection over the elevated topography of the region
(Demko et al., 2009) can organize into mesoscale convective systems, which produce intense precipitation usually in the afternoon, but also in the evening or night, especially at lower elevations (Higgins et al., 2006; Castro et al., 2007; Carbone et al., 2002).
Moisture surges from the Gulf of California, as well as tropical Pacific cyclones originating off the southern coast of Mexico, are responsible for precipitation during late summer, especially at the northwest boundary of the NAM, where individual storms can produce a large fraction of the total monsoon rainfall in just one event (Hereford et al.,
2006).
Summertime precipitation linked to the NAM may also reflect thermal effects of western US snowpack on the high-pressure system located where the states of Arizona,
Utah, Colorado, and New Mexico meet (Hawkins et al., 2002). The duration of the winter snowpack can influence the start of the monsoon season (Notaro et al., 2011; Zhu et al.,
2005). Heavy winter snowpack from the previous winter may increase soil moisture and decrease surface temperatures in the summertime, thereby limiting convective energy available for monsoonal thunderstorms (Lo and Clark, 2002; Gutzler, 2000). Even though the correlation between cool and warm season precipitation is generally weak and 8
temporally unstable in paleoclimatic records, such “opposite sign” anomalies have been increasingly observed during the mid-to-late 20th century (Griffin et al., 2013).
The NAM domain has been subdivided in eight regions, as defined by the North
American Monsoon Experiment Forecast Forum (Gochis et al., 2009). Southern Nevada is part of region 7, which occupies the northwest boundary of the NAM region (Griffin et al., 2013). Relatively few studies have focused on NAM boundary regions compared to the core area, especially when considering the western-northwestern edge of the NAM.
Our main objective was to determine how large-scale monsoonal patterns influence precipitation in mid-to-upper elevations near the NAM northwest boundary. With that goal, we quantified hourly, inter-seasonal, and inter-annual variability using three independent data sources. Among the data we analyzed, a unique opportunity was provided by hourly observations collected on a remote mountain site during 2011–2017. In our three-pronged investigation, we addressed the following questions. (1) Was our study site representative of the NAM northwest boundary based on long-term precipitation data? (2) What were the meteorological variables most closely linked to NAM precipitation at our study site? (3)
Which moisture sources were likely responsible for NAM precipitation at our study site?
Data and Methods
Study Site and Classification of Three Seasonal Intervals
We had an opportunity to investigate monsoonal dynamics at hourly intervals thanks to an automated weather station that operated continuously since April 2011 at 2300 m asl in a remote location on the Sheep Range, ~35 km north-northwest of Las Vegas
(Figure 1). This unique site is part of the Nevada Climate-ecohydrological Assessment
Network (NevCAN), an observing system managed by the Nevada System of Higher
Education (Mensing et al., 2013). The Sheep Range is a ‘sky-island’ within the Mojave
Desert, and our site is within the Desert National Wildlife Refuge, which is managed by 9
the US Fish and Wildlife Service. Given the relatively high elevation of our study site, vegetation is dominated by ponderosa pine (Pinus ponderosa Lawson & C. Lawson) and single-leaf pinyon pine (Pinus monophylla Torr. & Frém.), with sparse Utah juniper
(Juniperus osteosperma (Torr.) Little). Shrub and herbaceous cover is almost absent.
Soil type includes colluvium and/or residuum from weathered dolomite and limestone classified as loamy-skeletal, mixed, superactive, mesic Aridic Lithic Haplustolls (Johnson et al., 2014).
Daily meteorological data collected at the site were obtained from the Western
Regional Climate Center (https://wrcc.dri.edu/). Weather variables included barometric pressure, precipitation, air temperature, and relative humidity at 2-m height. Soil temperature and moisture probes had been installed around the NevCAN weather station at the time of its construction, and we obtained soil temperature at 5-cm depth and soil volumetric water content (VWC) at two soil depths: 2–17 cm and 17–32 cm.
High-resolution images taken at noon local time by a remotely operated camera placed at the top of the station tower were used to determine the presence of snow, as well as the last day of the year when snow was present. Vapor pressure deficit (VPD) and dewpoint temperature were calculated using formulas included in (Murray, 2010;
Brutsaert, 1982; Biondi et al., 2010; Kunkel, 1989) in Appendix A.
Careful evaluation of the 2011–2017 meteorological data indicated that warm- season precipitation events occurred at the study site after average daily dewpoint temperatures reached 9.4 °C. Dewpoint thresholds have indeed been used operationally to define monsoon onset by local National Weather Service Forecast Offices in the
Southwest. These thresholds can vary between locations, as Phoenix used 12.8 °C and
Tucson 12.2 °C, while regional analyses have suggested 10 °C (Ellis et al., 2004), but no single value is able to capture the onset of the monsoon across space and time. We 10
used our threshold to distinguish between early warm season (pre-monsoon) and late warm season (monsoonal). Specifically, we performed seasonal analyses for the 1 April
2011 to 30 September 2017 period using three subdivisions of the water year (previous
October through current September): (a) cool season (1 October through 31 March), (b) early warm season (1 April through last day with dewpoint < 9.4 °C), and (c) late warm season (first day with dewpoint ≥ 9.4 °C through 30 September). We also calculated average diurnal (or diel) cycles by season for weather and soil variables using hourly data aggregated from NevCAN measurements. These graphs normally do not include error bars (see Prakash et al., 2015 for a recent example) because their purpose is to visualize and summarize observations in an effective way, rather than to formally test for statistical significance (Biondi and Rossi, 2015). However, in order to highlight the variability of each hourly average, and also to indicate significant differences among seasons, we added two-standard error bars above and below each mean.
Analysis of NAM-related variability
We first tested whether July through September precipitation at our study site is representative of precipitation variability elsewhere at the northwest boundary of the
NAM. For this analysis, we used the 800-m grid cell version of the Parameter-elevation
Relationships on Independent Slopes Model (PRISM) dataset (Daly et al., 2008). We then extracted July–September precipitation totals during 1895–2015 for the grid cell containing the study site as well as for 475,415 other grid cells across the surrounding region. Overall temporal synchronicity was estimated using Pearson’s simple linear correlation between each grid cell precipitation and that of our study site.
We then investigated the linkage between afternoon thunderstorms occurring at the study site during the late warm season and other in-situ meteorological variables.
Because the variables we considered were highly correlated, we identified relationships 11
with precipitation using analysis of covariance (ANCOVA). Total hourly precipitation was used as predictand, whereas hourly solar radiation, air temperature, vapor pressure deficit, and dewpoint temperature were used as predictors, with each variable having its own ANCOVA. Seasons, as well as hour of the day, were considered independent categorical variables in each ANCOVA, and an interaction term was introduced between hour of the day and the meteorological variable in each model. All statistical analyses were performed using the R computing environment (R Core Team, 2015).
Finally, we evaluated whether the late warm season precipitation at the study site was indeed NAM-derived using reanalysis fields of atmospheric water vapor. Vertically integrated water vapor transport (IVT, kg m−1 s−1) data were obtained from the web site maintained by NOAA Earth System Research Laboratory, Physical Sciences Division, to make available the National Centers for Environmental Prediction/National Center for
Atmospheric Research (NCEP/NCAR) Reanalysis products
(https://www.esrl.noaa.gov/psd/data/composites/day/derived.html). Seasonal IVT composites were then calculated and mapped to represent both the amount of water vapor in the atmosphere and the directions along which water vapor was being transported. These results were used to determine whether late warm-season IVTs to the study site bore the signature northward patterns that characterize NAM conditions.
Results
Precipitation patterns
Variations in July–September total precipitation during 1895–2015 at our study site were significantly correlated (p-value < 0.01) with all 475,416 PRISM grid cells used in our analysis (Figure 1). Within this region, the mid-to-upper elevations of the Sheep
Range and Spring Mountains (southwest of the Sheep Range) were the most correlated locations (r > 0.8), whereas precipitation in the Las Vegas metropolitan area (south- 12
southeast of the Sheep Range) and other low-elevation valleys were less correlated (r <
0.4; Figure 1). Because relatively high correlations (cyan to blue colors in Figure 1) were found over most of the NAM northwest sector, our site is representative of historical variations in late-warm season precipitation for most areas, and especially those covered by woody vegetation, near the northwest boundary of the NAM.
The 9.4 °C daily mean dewpoint temperature threshold was met in early July (1–5) during 2011, 2014, 2015, and 2016, while it was reached in mid-July (11–17) in 2012,
2013, and 2017 (Figure 2a, Table 1). Late warm-season precipitation, following the dewpoint temperature threshold, reached the Sheep Range in July of every year. These dates fall within the 21 June–3 August window, which is when NAM precipitation arrives in its core region (Higgins et al., 1997). Daily dewpoint temperature and precipitation were significantly correlated (p-value < 0.001) for all three seasons, with the late warm- season relationship being slightly higher (r = 0.33) than that for the cool season (r =
0.29) and early warm season (r = 0.24).
A severe drought in 2012–2015 affected California as well as southern Nevada, including our study site. The number of days when snow was present for each cool season from October 2012 to March 2015 was less than for the cool seasons of October
2015 through March 2017. In concurrence, snow disappeared earlier from 2012 to 2015
(16 April 2012 to 7 March 2015) than from 2016 to 2017 (30 April 2016 to 2 April 2017)
(Figure 2b). There was no significant correlation between daily precipitation and soil volumetric water content, which, however, increased substantially at both depths during the wet 2016 and 2017 water years (Figures 2b and 3). Soil moisture generally remained higher at the greater depth during wet (2016–2017) and dry (2014–2015) years, whereas greater water content was observed in the shallower soil layer during most of the initial recording period (2011–2013; Figure 2b). Soil temperature varied little (data not shown) 13
and was ≤0 °C rarely (105 days out of the 2375 included in the study period) even at the shallow (5 cm) depth we considered. All the days with average soil temperature below freezing point occurred in the cool season, with 95% of them recorded from December to
February. December had an average of seven days (range from 0 to 12 days), January had average of six days (range from 0 to 21), and February had an average of three days (range from 0 to 6). No days with frozen soil were observed in February during
2015–2017, and 2017 was completely frost-free, most likely because of increased snow cover during those years (Figure 2b).
Increased snow cover from October 2015 through December 2017 corresponded to higher percentages of cool-season precipitation (Figure 3) during the 2016 (58% of total) and 2017 (62% of total) water years. The greatest contribution of early warm-season total precipitation to the water year was in 2016 (32% of total), following a wet cool season, which led to the termination of the multi-year severe drought. Increased precipitation in the 2016 cool season and early warm season was followed by reduced late warm-season precipitation (10% of total; Figure 3). Except for 2016, more precipitation fell at our study site during the late warm season than the early warm season (Table 1), highlighting the NAM influence. Late warm-season precipitation was relatively similar (173–190 mm) in 2012–2014, during a severe drought, but was considerably lower during the large snowpack years of 2011 (72 mm) and 2016 (46 mm), and also (albeit less so) in 2017 (135 mm). Late warm-season precipitation in fact accounted for the majority of total water-year precipitation in 2012 (55%), 2013 (53%),
2014 (56%), and 2015 (43%; Figure 3). The 2017 water year was very wet at the site, mostly due to storms originating from the Pacific Ocean that also brought record precipitation to the Sierra Nevada. While the 2017 cool season was notably wet, a large percentage of precipitation also fell during the late warm season (35%; Figure 3). 14
Diurnal cycle and relationship with precipitation
The quality of our dataset was supported by the extremely low number of missing observations (no more than 8 out of 59208 hourly records). Hourly precipitation was extremely variable, as shown by its large error bars (greatest during the late warm season; Figure 4a). Most precipitation in the cool and early warm season fell in the morning and afternoon, from 08:00 to 16:00 (79% of the total daily value in the cool season, with a peak at 11:00, and 74% in the early warm season; Figure 4a). In the late warm season, less precipitation (53%) fell in this interval, and mostly in the later hours, from 11:00 to 16:00 (49%, with a peak at 15:00), likely indicative of convective afternoon thunderstorms that are characteristic of NAM precipitation. A second surge of precipitation in this season fell at night, from 23:00 to 01:00 (Figure 4a), accounting for
18% of the average daily total.
Dewpoint temperature differed by about 8 °C between early warm season (mean of
−6.3 °C) and late warm season (mean of 2.3 °C) for every hour of the diel cycle (Figure
4b). This was the greatest difference between seasons for any of the variables we analyzed and was also statistically significant for every hour (Figure 4b). Both dewpoint and air temperature increased significantly from cool season to early warm season to late warm season (Figure 4b,c). Early and late warm-season hourly VPD values were comparable during the night but were significantly greater in the late warm season during the morning and early afternoon (08:00–15:00). Early and late warm-season VPD were both significantly greater than cool season values (Figure 4d). Solar radiation was significantly lower during the cool season; it reached a maximum in the early warm season, and then significantly decreased in the late warm season likely due to increased cloud cover during the day (Figure 4e). Volumetric water content at both depths was quite stable during the diel cycle, and was highest in the early warm season, especially 15
in deep soil, and lowest in the late warm season (Figure 4f,g), likely because of the seasonal balance between snowmelt infiltration and evapotranspiration.
Statistical relationships derived from the hourly dataset of meteorological variables, seasons, and hour of the day, as quantified by ANCOVA, showed that each season had a similar influence on hourly precipitation (F-values for ANCOVA models ranged from
28.9 to 29.6, p-value < 0.01). Hour of the day was not significant in any of the
ANCOVAs. Dewpoint temperature had the greatest relationship with hourly precipitation
(F-value = 237.8, p-value < 0.01), followed by VPD (F-value = 100.3, p-value < 0.01), solar radiation (F-value = 42.8, p-value < 0.01), and air temperature (F-value = 9.3, p- value < 0.01). The only significant interaction between hour of the day and meteorological variables was with dewpoint temperature (F-value = 4.7, p-value = 0.03).
Thus, dewpoint temperature had the strongest relationship with hourly precipitation, while VPD, solar radiation, and air temperature lagged far behind.
An association between increasing dewpoint temperature and precipitation was quite evident when plotting total weekly precipitation and average weekly dewpoint temperature at our study site (Figure 5). This relationship became noisier, albeit still visible, when variables were calculated on a daily or hourly time scale (Appendix B).
Individual precipitation events can be very intense and drop large rainfall amounts within short periods, as shown by a thunderstorm that produced 57 mm of precipitation in less than five hours during the afternoon of 1 August 2015 (Figure 2a).
Analysis of integrated water vapor transport (IVT)
Sources of water vapor that reached the NevCAN site differed between the cool and early warm seasons versus the late warm season. Cool-season and early warm-season
IVT originated over the North Pacific and resulted in precipitation at our site from Pacific frontal storms, whereas late warm-season IVT arrived from the Gulf of California and the 16
eastern tropical Pacific (Figure 6), as expected for the NAM. The Sierra Madre
Occidental appeared as a terrestrial source of IVT, possibly because of mixing of moisture due to deep convection resulting from water vapor transport from the Gulf of
California and eastern Pacific. Some additional moisture may have originated from local evapotranspiration as well as from convective activity further south over Mexico (Hu and
Dominguez, 2015; Dominguez et al., 2008), but reanalysis data are of low geographic resolution that does not distinguish such divisions well, particularly over a complex region such as northwest Mexico where few long-term reliable datasets exist (Serra et al., 2016).
The 2011, 2016, and 2017 cool seasons, which were outside the drought period, experienced greater IVT off the Pacific coast, likely associated with more frequent
Pacific frontal storms resulting in greater cool-season precipitation at our site. In fact,
2016 and 2017 were the only years when cool-season precipitation was the greatest contributor to total water-year precipitation (Figure 3, Table 1). Early warm-season IVT for all years except 2011 was similar or less than for the cool season, as also reflected in decreased precipitation (Figure 3, Table 1). Late warm-season IVT composite indicated water vapor reaching southern Nevada from the south each year (Figure 6).
Discussion
NAM precipitation variability
In situ weather data are rarely available at high elevations in topographically complex regions over several years (Bradley et al., 2004), but they can often generate novel insights and hypotheses on meteorological processes (Biondi et al., 2009) or be used to test accuracy of interpolated datasets (Strachan and Daly, 2017). The continuous hourly observations, in relation to the IVT seasonal composites, provided an 17
opportunity to assess how summertime precipitation in southern Nevada reflects the large-scale meteorology and expression of the NAM northwest extension. Local meteorological variables, especially dewpoint temperature, were correlated with the amount of precipitation falling at the study site. Therefore, dewpoint temperature was appropriate to define the boundary between early and late warm-season precipitation.
These findings support the National Weather Service usage of dewpoint temperature as the atmospheric variable to define the NAM onset.
The transition from early to late warm season with the arrival of NAM precipitation at our site corresponds to an increase in frequency of convective afternoon thunderstorms linked with moisture surges from the Gulf of California (Favors and Abatzoglou, 2013;
Pascale et al., 2018). These surges are anomalous southerly flow events with the duration of a few days (Bordoni and Stevens, 2006) that can generate intense precipitation (Adams and Comrie, 1997), reaching as far as the western-northwestern
NAM boundary. During strong monsoonal years, like 2015, water vapor in the late warm season associated with the NAM precipitated at the study site from afternoon convective thunderstorms. Solar radiation also decreased in diurnal cycles of the late warm season, likely because of cloud cover from afternoon thunderstorms. The arrival of afternoon convective thunderstorms begins in early-to-middle July, but the amount of precipitation that falls in each storm varies greatly. Gochis et al., (2004) found that upper-to-middle elevations in the NAM core region in western Mexico have more intense and less frequent precipitation events than the highest elevations, but more frequent and less intense precipitation events than at lower elevations. These elevation-related differences in precipitation patterns are presumably responsible for the lack of correlation we uncovered between our study site and the Las Vegas valley (Figure 1). During the late warm season, Gochis et al., (2004) also measured nocturnal storms, whose 18
environmental conditions and thermodynamics have been studied by other authors
(Smith and Gall, 1989; Farfán and Zehnder, 1994; Lang et al., 2004). Our study site, which is in the upper-to-middle elevation of the Sheep Range, displayed similar patterns.
Future research could investigate monsoonal precipitation at our study site compared to other NevCAN sites to better evaluate the role of elevational differences. In addition, both local and regional NAM precipitation could be evaluated using Convective Available
Potential Energy (CAPE) or other convectively important atmospheric parameters (Brook et al., 2007, Riemann-Campe et al., 2011)
Regional IVT patterns indicated that precipitation regime at the study site is associated with a shift in the source region from the cool/early warm season to the late warm season. Water vapor reaching the southwestern USA during the cool season and early warm season of 2011 to 2017 originated every year from large areas in the Pacific
Ocean, albeit with different strengths from one year to the next. This water-vapor transport is associated with synoptic-scale seasonal patterns of atmospheric circulation over the Pacific, e.g., variations of the North Pacific High and Aleutian Low, as well as atmospheric rivers that can strongly influence the amount of cool-season precipitation in the southwestern USA (Dettinger et al., 2011; Rutz et al., 2014). The transition from early to late warm season, as defined here by the 9.4 °C dewpoint threshold, coincided every year with a large-scale shift in vapor sources from the Pacific Ocean to the Gulf of
California and the Sierra Madre Occidental. This shift in IVT and associated winds is one of the most characteristic meteorological patterns of the NAM (Douglas et al., 1993).
Late warm-season IVT arrived at our site from the south every year from 2011 to 2017, with lower IVT values directly north and west of it. This pattern was not affected by the strength of the monsoon season in the NAM core region, which varied during these years, and also affected the amount of precipitation measured at the study site. Because 19
NCEP reanalysis features a horizontal resolution of 2.5° × 2.5°, future research could be aimed at refining the connection between our study site and NAM circulation features, such as the Gulf of California low-level jet or the Gulf of California moisture surges, using products with smaller grid cells, including the North American regional reanalysis
(Mesinger et al., 2006), the Modern-Era Retrospective analysis for Research and
Applications, Version 2 (MERRA-2; Gelaro et al., 2017) or the European Centre for
Medium-Range Weather Forecasts' (ECMWF's) ERA5 (Hoffman et al., 2019) reanalysis.
Afternoon thunderstorms associated with the NAM are crucial for maintaining the sky-island ecosystems found in the region (Peltier et al., 2019), as they recharge topsoil water for biotic use after the dry early warm season. Mature ponderosa pines predominantly use winter precipitation but they are also able to exploit monsoonal precipitation to relieve summer drought stress caused by high VPD (Kerhoulas et al.,
2017). Studies conducted on ponderosa pine at our site have indicated that summertime precipitation pulses are primary drivers of stem radial expansion and growth (Ziaco et al., 2018). While ecohydrological processes appear to recycle precipitation through evapotranspiration on a storm-scale (Xiang et al., 2018), data from an individual location can only provide anecdotal information on such processes. For instance, the arrival of
Hurricane Dolores in 2015 relieved southern Nevada from the high VPD conditions of the early warm season. The late warm-season precipitation from Hurricane Dolores was linked to the beginning of radial growth for ponderosa pine at the study site (Ziaco et al.,
2018), and the largest precipitation event occurred about one week later.
The largest influence on soil moisture in these sky-island ecosystems is given by cool-season precipitation (Figure 2b), essentially because of snowpack melting. Soil water eventually percolates to deeper soils and is incorporated into the groundwater that flows downslope into the surrounding valleys (Moreo et al., 2014). Soil moisture 20
recharge is lessened throughout the warm season, as shown by stable isotopic ratios measured by Ingraham et al. (1991) in the adjacent Spring Mountains, and by Earman et al. (2006) in Arizona and New Mexico. Stable isotope analysis has indeed been used to infer that most groundwater recharge in the southwestern US derives from snowmelt rather than warm-season precipitation (Earman et al., 2006; Winograd et al., 1998). This is supposed to happen even in areas, such as ours, where cool-season precipitation may be equal, or even less, than warm-season precipitation. It should be noted that our study site is the only location within the northwest NAM region where in situ data can be used to test the distinction in sources of groundwater recharge that has been derived from isotopic studies (see also Devitt et al., 2018).
Variations of total July–September precipitation during 1895–2015 at the study site were well correlated with those in surrounding areas, especially at similar or higher elevations. Because monsoon precipitation is expected to be different at a ‘sky island’ location than at a lowland or basin location (such as the Las Vegas valley in this case), our results were consistent with monsoonal dynamics. We relied on the PRISM dataset because it is commonly used in the United States for studying climate change and its impacts (Lobell et al., 2014). At the same time, every interpolated dataset is subject to potential sources of bias, and one of them was recently identified in the high-elevation
SNOwpack TELemetry (SNOTEL) network, which is particularly relevant for the topographically complex landscape of the western US. SNOTEL data were found to be affected by warming artifacts and sensor biases that, when propagated into climate datasets that incorporate them, such as PRISM, have most likely amplified the “1981–
2012 western U.S. elevation-dependent warming by +217 to +562%” (Oyler et al., 2015).
Since we were using PRISM precipitation data, it is unlikely that such temperature bias could have produced the lower correlations between our study site and the surrounding 21
valleys. Also, we decided to use the entire length of the PRISM record, starting in year
1895, to make the comparison less sensitive to potentially anomalous historical periods, which is always a concern when relatively few years are used. The IVT analysis suggested that water vapor associated with the NAM pattern reaches the site, and afternoon thunderstorms characteristic of NAM precipitation occurred at the site. The connection between the transition from cool- and early warm-season IVT from the
Pacific Coast to the Gulf of California, the highly correlated historical trends in July–
September total precipitation for the surrounding area, and the presence of afternoon convective thunderstorms provide clear evidence that our study site is highly influenced by, hence representative of, the NAM northwest boundary.
Recent droughts and pluvials
The 2012–2015 severe warm drought in California and Nevada has been compared to the intense droughts of the medieval climate anomaly (Hatchett et al., 2015), although it had shorter duration (Cook et al., 2016). This drought was clearly reflected by the low
VWC at both soil depths, which were notably drier than in either 2016 or 2017. Soil moisture is vital to the vegetation and to the iconic organisms that depend on it (e.g., desert big horn sheep and Agassiz’s desert tortoise). The low VWC during the 2012–
2015 severe warm drought was linked with decreased snow cover, leading to higher soil infiltration rates earlier in the warm season because snow was melting earlier in the year.
Late warm-season precipitation was the greatest source of precipitation during the
2012–2015 drought, highlighting the importance of precipitation associated with the
NAM, especially in a severe drought period. The termination of the 2012–2015 drought at the study site began with the arrival of precipitation from late warm-season water vapor originating from the Gulf of California following the warmest period we observed, 22
during the 2015 early warm season. Water vapor originating from Hurricane Dolores off the Mexico coast, and strong thunderstorms in August 2015, brought surges of precipitation at the site. Then, a very wet 2015–2016 cool season and 2016 early warm season terminated the 2012–2015 drought. Cerezo-Mota et al. (2016) found that stronger monsoon seasons may be related to a more northward, intense, and larger inter-tropical convergence zone, which could explain the intense monsoon season of
2015.
Cool seasons with more precipitation, longer snow-cover periods, and later snowpack melting allow greater opportunities for recharging soil moisture at both shallow and deep soil levels. They may also influence NAM precipitation during the following warm season (Hawkins et al., 2002.), which may have been the case for NAM precipitation in 2016. A weak monsoon following abundant cool-season precipitation has been observed in the NAM core region, particularly in the mid-to-late 20th century, even though the dynamics of this reversal are uncertain (Griffin et al., 2013; Stahle et al.,
2009).
Conclusions
Summer precipitation measured at the NevCAN site is representative of surrounding areas and NAM processes. During every year of our study, the NCEP/NCAR IVT analysis suggested that the Sheep Range falls within the northwest boundary of the
NAM. This pattern was not affected by the strength of the monsoon season in the NAM core region, which was stronger in 2011, 2013, and 2015. Because the local increase in daily dewpoint temperature with the arrival of NAM-associated IVT during the late warm season had the strongest relationship to increased precipitation events at our site, daily dewpoint temperature thresholds are particularly well suited to define the local onset of
NAM precipitation. 23
Late warm-season precipitation contributed the greatest amount of precipitation during the 2012–2015 water years, highlighting the importance of NAM precipitation amid drought. The 2012–2015 drought was followed by a wet spell that started with the influence of Hurricane Dolores and included a monsoonal thunderstorm that produced the largest precipitation event since our site began operating. The 2015–2016 cool and early warm seasons ended the drought period, and a weak monsoon was then apparent in 2016. Future hydroclimatic variability in the NAM region, including the frequency and intensity of tropical Pacific cyclones, is highly uncertain. Shifts in global monsoon precipitation (Seidel et al., 2008) could lead to major changes from the historical water balance, affecting not only the hydrology of monsoonal areas but also biotic and human communities. Better knowledge of current and past influences of the NAM upon water reserves and ecosystem responses, especially at its highly variable northwest border, provides crucial baseline information for anticipating future impacts of potential shifts in precipitation regimes.
Acknowledgments: We thank A. Csank, D.K. Adams, and S. Pascale for helpful comments and advice on an earlier version of this manuscript, S. Strachan for NevCAN maintenance, and A. Sprunger for collaborating with the NevCAN team on the Sheep
Range transect. NevCAN data, including those we used in this study, are freely available from http://sensor.nevada.edu/SENSORDataSearch/.
References
Adams, D.K.; Souza, E.P. CAPE and convective events in the Southwest during the North American Monsoon. Mon. Weather Rev. 2009, 137, 83–98, doi:10.1175/2008mwr2502.1.
Brutsaert, W. Evaporation into the Atmosphere; D. Reidel Publishing Company: Boston, MA, USA, 1982; p. 299.
24
Biondi, F.; Hartsough, P. Using automated point dendrometers to analyze tropical treeline stem growth at Nevado de Colima, Mexico. Sensors 2010, 10, 5827–5844, doi:10.3390/s100605827.
Biondi, F.; Rossi, S. Plant-water relationships in the Great Basin Desert of North America derived from Pinus monophylla hourly dendrometer records. Int. J. Biometeorol. 2015, 59, 939–953, doi:10.1007/s00484-014-0907-4.
Biondi, F.; Hartsough, P.; Galindo Estrada, I.G. Recent warming at the tropical treeline of North America. Front. Ecol. Environ. 2009, 7, 463–464, doi:10.1890/09.wb.028.
Bradley, R.S.; Keimig, F.T.; Diaz, H.F. Projected temperature changes along the American cordillera and the planned GCOS network. Geophys. Res. Let. 2004, 31, L16210.
Bordoni, S.; Stevens, B. Principal component analysis of the summertime winds over the Gulf of California: A gulf surge index. Mon. Weather Rev. 2006, 134, 3395–3414, doi:10.1175/mwr3253.1.
Gochis, D.J.; Jimenez, A.; Watts, C.J.; Garatuza-Payan, J.; Shuttleworth, W.J. Analysis of 2002 and 2003 warm-season precipitation from the North American Monsoon Experiment event rain gauge network. Mon. Weather Rev. 2004, 132, 2938–2953, doi:10.1175/mwr2838.1.
Ellis, A.W.; Saffell, E.M.; Hawkins, T.W. A method for defining monsoon onset and demise in the southwestern USA. Int. J. Climatol. 2004, 24, 247–265, doi:10.1002/joc.996.
Carbone, R.E.; Tuttle, J.D.; Ahijevych, D.A.; Trier, S.B. Inferences of predictability associated with warm season precipitation episodes. J. Atmos. Sci. 2002, 59, 2033–2056, doi:10.1175/1520-0469(2002)059<2033:iopaww>2.0.co;2.
Castro, C.L.; Pielke Sr., R.A.; Adegoke, J.O. Investigation of the summer climate of the contiguous United States and Mexico using the Regional Atmospheric Modeling System 25
(RAMS). Part I: Model climatology (1950–2002). J. Clim. 2007, 20, 3844–3865, doi:10.1175/jcli4211.1.
Cook, B.I.; Seager, R. The response of the North American Monsoon to increased greenhouse gas forcing. J. Geophys. Res. Atmos. 2013, 118, 1690–1699. Adams, D.K.; Comrie, A.C. The North American Monsoon. Bull. Am. Meteorol. Soc. 1997, 78, 2197–2213.
Daly, C.; Halbleib, M.; Smith, J.I.; Gibson, W.P.; Doggett, M.K.; Taylor, G.H.; Curtis, J.; Pasteris, P.P. Physiographically sensitive mapping of climatological temperature and precipitation across the conterminous United States. Int. J. Climatol. 2008, 28, 2031–2064.
Dominguez, F.; Kumar, P.; Vivoni, E.R. Precipitation recycling variability and ecoclimatological stability—A study using NARR data. Part II: North American Monsoon region. J. Clim. 2008, 21, 5187–5203, doi:10.1175/2008jcli1760.1.
Demko, J.C.; Geerts, B.; Miao, Q.; Zehnder, J.A. Boundary layer energy transport and cumulus development over a heated mountain: An observational study. Mon. Weather Rev. 2009, 137, 447–468, doi:10.1175/2008mwr2467.1.
Higgins, R.W.; Chen, Y.; Douglas, A.V. Interannual variability of the North American warm season precipitation regime. J. Clim. 1999, 12, 653–680.
Douglas, M.W.; Maddox, R.A.; Howard, K.; Reyes, S. The Mexican Monsoon. J. Clim. 1993, 6, 1665–1677, doi:10.1175/1520-0442(1993)006<1665:tmm>2.0.co;2.
Gutzler, D.S. Covariability of spring snowpack and summer rainfall across the Southwest United States. J. Clim. 2000, 13, 4018–4027, doi:10.1175/1520- 0442(2000)013<4018:cossas>2.0.co;2.
Griffin, R.D.; Woodhouse, C.A.; Meko, D.M.; Stahle, D.W.; Faulstich, H.L.; Carrillo, C.; Touchan, R.; Castro, C.L.; Leavitt, S.W. North American monsoon precipitation reconstructed from tree-ring latewood. Geophys. Res. Let. 2013, 40, 954–958. 26
Gochis, D.; Schemm, J.-K.; Shi, W.; Long, L.; Higgins, R.W.; Douglas, A.V. A forum for evaluating forecasts of the North American Monsoon. Eos Tran. Am. Geophys. Union 2009, 90, 249–251.
Favors, J.E.; Abatzoglou, J.T. Regional surges of monsoonal moisture into the Southwestern United States. Mon. Weather Rev. 2013, 141, 182–191, doi:10.1175/mwr- d-12-00037.1.
Hawkins, T.W.; Ellis, A.W.; Skindlov, J.A.; Reigle, D. Intra-annual analysis of the North American snow cover–monsoon teleconnection: Seasonal forecasting utility. J. Clim. 2002, 15, 1743–1753, doi:10.1175/1520-0442(2002)015<1743:iaaotn>2.0.co;2.
Hu, H.; Dominguez, F. Evaluation of oceanic and terrestrial sources of moisture for the North American Monsoon using numerical models and precipitation stable isotopes. J. Hydrometeorol. 2015, 16, 19–35, doi:10.1175/jhm-d-14-0073.1.
Hereford, R.; Webb, R.H.; Longpré, C.I. Precipitation history and ecosystem response to multidecadal precipitation variability in the Mojave Desert region, 1893–2001. J. Arid Environ. 2006, 67, 13–34, doi:10.1016/j.jaridenv.2006.09.019.
Higgins, R.W.; Ahijevych, D.; Amador, J.; Barros, A.; Berbery, E.H.; Caetano, E.; Carbone, R.; Ciesielski, P.; Cifelli, R.; Cortez-Vazquez, M., et al. The NAME 2004 field campaign and modeling strategy. Bull. Am. Meteorol. Soc. 2006, 87, 79–94.
Jana, S.; Rajagopalan, B.; Alexander, M.A.; Ray, A.J. Understanding the dominant sources and tracks of moisture for summer rainfall in the Southwest United States. J. Geophys. Res. Atmos. 2018, 123, 4850–4870, doi:10.1029/2017jd027652.
Higgins, R.W.; Yao, Y.; Wang, X.L. Influence of the North American Monsoon system on the U.S. summer precipitation regime. J. Clim. 1997, 10, 2600–2622.
27
Lo, F.; Clark, M.P. Relationships between spring snow mass and summer precipitation in the southwestern United States associated with the North American Monsoon System. J. Clim. 2002, 15, 1378–1385.
Johnson, B.G.; Verburg, P.S.J.; Arnone III, J.A. Effects of climate and vegetation on soil nutrients and chemistry in the Great Basin studied along a latitudinal-elevational climate gradient. Plant Soil 2014, 382, 151–163, doi:10.1007/s11104-014-2144-3.
Kunkel, K.E. Simple procedures for extrapolation of humidity variables in the mountainous western United States. J. Clim. 1989, 2, 656–669, doi:10.1175/1520- 0442(1989)002<0656:spfeoh>2.0.co;2.
Luong, T.M.; Castro, C.L.; Chang, H.-I.; Lahmers, T.; Adams, D.K.; Ochoa-Moya, C.A. The more extreme nature of North American Monsoon precipitation in the Southwestern United States as revealed by a historical climatology of simulated severe weather events. J. Appl. Meteorol. Clim. 2017, 56, 2509–2529, doi:10.1175/jamc-d-16-0358.1.
Notaro, M.; Zarrin, A. Sensitivity of the North American monsoon to antecedent Rocky Mountain snowpack. Geophys. Res. Let. 2011, 38, doi:10.1029/2011gl048803.
Mensing, S.; Strachan, S.; Arnone, J.; Fenstermaker, L.; Biondi, F.; Devitt, D.; Johnson, B.; Bird, B.; Fritzinger, E. A network for observing Great Basin climate change. Eos Tran. Am. Geophys. Union 2013, 94, 105–106, doi:10.1002/2013eo110001.
Murray, F.W. On the computation of saturation vapor pressure. J. Appl. Meteorol. 1967, 6, 203–204.
Pascale, S.; Kapnick, S.B.; Bordoni, S.; Delworth, T.L. The influence of CO2 forcing on North American Monsoon moisture surges. J. Clim. 2018, 31, 7949–7968, doi:10.1175/jcli- d-18-0007.1.
28
Prakash, S.; Shati, F.; Norouzi, H.; Blake, R. Observed differences between near-surface air and skin temperatures using satellite and ground-based data. Theor. Appl. Climatol. 2019, 137, 587-600, doi:10.1007/s00704-018-2623-1.
R Core Team. R: A Language and Environment for Statistical Computing, 3.0.2; R Foundation for Statistical Computing: Vienna, Austria, 2015.
Serra, Y.L.; Adams, D.K.; Minjarez-Sosa, C.; Moker Jr., J.M.; Arellano, A.F.; Castro, C.L.; Quintanar, A.I.; Alatorre, L.; Granados, A.; Vazquez, G.E., et al. The North American Monsoon GPS Transect Experiment 2013. Bull. Am. Meteorol. Soc. 2016, 97, 2103–2115, doi:10.1175/bams-d-14-00250.1.
Strachan, S.; Daly, C. Testing the daily PRISM air temperature model on semiarid mountain slopes. J. Geophys. Res. Atmos. 2017, 122, 5697–5715, doi:10.1002/2016JD025920.
Vera, C.; Higgins, R.W.; Amador, J.; Ambrizzi, T.; Garreaud, R.D.; Gochis, D.; Gutzler, D.S.; Lettenmaier, D.P.; Marengo, J.A.; Mechoso, C.R., et al. Toward a unified view of the American monsoon systems. J. Clim. 2006, 19, 4977–5000.
Zhu, C.; Lettenmaier, D.P.; Cavazos, T. Role of antecedent land surface conditions on North American Monsoon rainfall variability. J. Clim. 2005, 18, 3104–3121, doi:10.1175/jcli3387.1.
Smith, W.P.; Gall, R.L. Tropical squall lines of the Arizona Monsoon. Mon. Weather Rev. 1989, 117, 1553–1569, doi:10.1175/1520-0493(1989)117<1553:tslota>2.0.co;2.
Farfán, L.M.; Zehnder, J.A. Moving and stationary mesoscale convective systems over northwest Mexico during the Southwest Area Monsoon Project. Weather Forecast. 1994, 9, 630–639, doi:10.1175/1520-0434(1994)009<0630:masmcs>2.0.co;2.
29
Lang, T.J.; Ahijevych, D.A.; Nesbitt, S.W.; Carbone, R.E.; Rutledge, S.A.; Cifelli, R. Radar- observed characteristics of precipitating systems during NAME 2004. J. Clim. 2007, 20, 1713–1733, doi:10.1175/jcli4082.1.
Brooks, H.E.; Anderson, A.R.; Riemann, K.; Ebbers, I.; Flachs, H. Climatological aspects of convective parameters from the NCAR/NCEP reanalysis. Atmos Res. 2007, 83, 294– 305, doi:10.1016/j.atmosres.2005.08.005.
Riemann-Campe, K.; Blender, R.; Fraedrich, K. Global memory analysis in observed and simulated CAPE and CIN. Int. J. Climatol. 2011, 31, 1099–1107, doi:10.1002/joc.2148.
Dettinger, M.D.; Ralph, F.M.; Das, T.; Neiman, P.J.; Cayan, D.R. Atmospheric rivers, floods and the water resources of California. Water 2011, 3, 445–478, doi:10.3390/w3020445.
Rutz, J.J.; Steenburgh, W.J.; Ralph, F.M. Climatological characteristics of atmospheric rivers and their inland penetration over the western United States. Mon. Weather Rev. 2014, 142, 905–921, doi:10.1175/mwr-d-13-00168.1.
Mesinger, F.; DiMego, G.; Kalnay, E.; Mitchell, K.; Shafran, P.C.; Ebisuzaki, W.; Jović, D.; Woollen, J.; Rogers, E.; Berbery, E.H., et al. North American Regional Reanalysis. Bull. Am. Meteorol. Soc. 2006, 87, 343–360, doi:10.1175/bams-87-3-343.
Gelaro, R.; McCarty, W.; Suárez, M.J.; Todling, R.; Molod, A.; Takacs, L.; Randles, C.A.; Darmenov, A.; Bosilovich, M.G.; Reichle, R., et al. The Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2). J. Clim. 2017, 30, 5419–5454, doi:10.1175/jcli-d-16-0758.1.
Hoffmann, L.; Günther, G.; Li, D.; Stein, O.; Wu, X.; Griessbach, S.; Heng, Y.; Konopka, P.; Müller, R.; Vogel, B., et al. From ERA-Interim to ERA5: The considerable impact of ECMWF's next-generation reanalysis on Lagrangian transport simulations. Atmos. Chem. Phys. 2019, 19, 3097–3124, doi:10.5194/acp-19-3097-2019.
30
Peltier, D.M.P.; Ogle, K. Legacies of La Niña: North American monsoon can rescue trees from winter drought. Glob. Chang. Biol. 2019, 25, 121–133, doi:10.1111/gcb.14487.
Kerhoulas, L.P.; Kolb, T.E.; Koch, G.W. The influence of monsoon climate on latewood growth of southwestern ponderosa pine. Forests 2017, 8, 140, doi:10.3390/f8050140.
Ziaco, E.; Biondi, F. Stem circadian phenology of four pine species in naturally contrasting climates from sky-island forests of the western USA. Forests 2018, 9, 396, doi:10.3390/f9070396.
Xiang, T.; Vivoni, E.R.; Gochis, D.J. Influence of initial soil moisture and vegetation conditions on monsoon precipitation events in northwest México. Atmósfera 2018, 31, 25– 45, doi:10.20937/ATM.2018.31.01.03.
Ziaco, E.; Truettner, C.; Biondi, F.; Bullock, S. Moisture-driven xylogenesis in Pinus ponderosa from a Mojave Desert mountain reveals high phenological plasticity. Plant Cell Environ. 2018, 41, 823–836, doi:10.1111/pce.13152.
Moreo, M.T.; Senay, G.B.; Flint, A.L.; Damar, N.A.; Laczniak, R.J.; Hurja, J. Hydroclimate of the Spring Mountains and Sheep Range, Clark County, Nevada; Scientific Investigations Report 2014-5142; U.S. Geological Survey: Reston, VA, USA, 2014; p 38.
Ingraham, N.L.; Lyles, B.F.; Jacobson, R.L.; Hess, J.W. Stable isotopic study of precipitation and spring discharge in southern Nevada. J. Hydrol. 1991, 125, 243–258. Earman, S.; Campbell, A.R.; Phillips, F.M.; Newman, B.D. Isotopic exchange between snow and atmospheric water vapor: Estimation of the snowmelt component of groundwater recharge in the southwestern United States. J. Geophys. Res. Atmos. 2006, 111, doi:10.1029/2005jd006470.
Winograd, I.J.; Riggs, A.C.; Coplen, T.B. The relative contributions of summer and cool- season precipitation to groundwater recharge, Spring Mountains, Nevada, USA. Hydrogeol. J. 1998, 6, 77–93, doi:10.1007/s100400050135.
31
Devitt, D.; Bird, B.; Lyles, B.; Fenstermaker, L.; Jasoni, R.; Strachan, S.; Arnone, J.A., III; Biondi, F.; Mensing, S.; Saito, L. Assessing near surface hydrologic processes and plant response over a 1600 m mountain valley gradient in the Great Basin, NV, U.S.A. Water 2018, 10, 420, doi:10.3390/w10040420.
Lobell, D.B.; Roberts, M.J.; Schlenker, W.; Braun, N.; Little, B.B.; Rejesus, R.M.; Hammer, G.L. Greater sensitivity to drought accompanies maize yield increase in the U.S. midwest. Science 2014, 344, 516–519, doi:10.1126/science.1251423.
Oyler, J.W.; Dobrowski, S.Z.; Ballantyne, A.P.; Klene, A.E.; Running, S.W. Artificial amplification of warming trends across the mountains of the western United States. Geophys. Res. Let. 2015, 42, 153–161, doi:10.1002/2014GL062803.
Hatchett, B.J.; Boyle, D.P.; Putnam, A.E.; Bassett, S.D. Placing the 2012–2015 California- Nevada drought into a paleoclimatic context: Insights from Walker Lake, California- Nevada, USA. Geophys. Res. Let. 2015, 42, 8632–8640, doi:10.1002/2015gl065841.
Cook, B.I.; Cook, E.R.; Smerdon, J.E.; Seager, R.; Williams, A.P.; Coats, S.; Stahle, D.W.; Díaz, J.V. North American megadroughts in the Common Era: Reconstructions and simulations. Wires Clim. Change 2016, 7, 411–432, doi:10.1002/wcc.394.
Cerezo-Mota, R.; Cavazos, T.; Arritt, R.; Torres-Alavez, A.; Sieck, K.; Nikulin, G.; Moufouma-Okia, W.; Salinas-Prieto, J.A. CORDEX-NA: Factors inducing dry/wet years on the North American Monsoon region. Int. J. Climatol. 2016, 36, 824–836, doi:10.1002/joc.4385.
Stahle, D.W.; Cleaveland, M.K.; Grissino-Mayer, H.D.; Griffin, R.D.; Fye, F.K.; Therrell, M.D.; Burnette, D.J.; Meko, D.M.; Villanueva-Diaz, J. Cool- and warm-season precipitation reconstructions over western New Mexico. J. Clim. 2009, 22, 3729–3750.
Seidel, D.J.; Fu, Q.; Randel, W.J.; Reichler, T.J. Widening of the tropical belt in a changing climate. Nat. Geosci. 2008, 1, 21, doi:10.1038/ngeo.2007.38. 32
Figure 1. Interpolated pseudo-color map of Pearson’s linear correlation between July–
September total precipitation at our mountain site (solid white circle) and at each 800-m
PRISM grid cell in the surrounding region during 1895–2015. All correlations were statistically significant (p-value < 0.01), with lower values usually corresponding to lower elevations, such as the Las Vegas valley (black circle).
33
Figure 2. Time series plots of daily environmental variables measured at our study site.
(a) Average dewpoint temperature and total precipitation; a 9.4 °C threshold of dewpoint temperature was used to identify the arrival of North American Monsoon (NAM) precipitation. (b) Volumetric water content at two soil depths (2–17 cm and 17–32 cm) and days when snow was present (see text for details). 34
Figure 3. Precipitation percentages at our study site for cool (1 October–31 March), early warm (1 April to the day prior to 9.4 °C dewpoint threshold), and late warm (day of
9.4 °C dewpoint threshold to 30 September) seasons of the 2012–2017 water years. 35
Figure 4. Average diurnal cycles of weather and soil variables by season. The variability of each hourly value is shown by vertical bars corresponding to two standard errors above and below the mean. Meteorological variables include: (a) total precipitation, (b) dewpoint temperature, (c) air temperature, (d) vapor-pressure deficit, (e) and solar radiation. Soil variables are (f) volumetric water content (VWC) at 2–17 cm soil depth, and (g) VWC at 17–32 cm soil depth. 36
Figure 5. Total weekly precipitation plotted against average weekly dewpoint temperature for early and late warm season at our study site. The larger precipitation events occurred during the late warm season as dewpoint temperature increased.
37
Figure 6. Seasonal averages of integrated water vapor transport (IVT) for the 2011–
2017 water years. Pseudo-color shading with overlaid vectors (arrows) was used to indicate the amount of moisture and the direction of transport. Cool-season IVT, and resulting precipitation was large in 2011, 2016, and 2017, as well as for the 2015 late warm season. 38
Table 1. Summary of seasonal and water year precipitation totals (PPT) at our study site in the Sheep Range of southern Nevada.
Monsoon Year * Cool Season Early Warm Season Late Warm Season Water Year Onset
PPT (mm) Days PPT (mm) Days Date PPT (mm) Days PPT (mm)
2011 - - 38.07 95 7/5/2011 71.63 88 -
2012 110.98 183 46.73 103 7/13/2012 189.97 80 347.68
2013 138.65 182 21.83 101 7/11/2013 177.28 82 337.76
2014 110.22 182 27.43 95 7/5/2014 172.73 88 310.38
2015 120.83 182 58.39 92 7/2/2015 133.85 91 313.07
2016 260.83 183 141.88 91 7/1/2016 45.97 92 448.68
2017 243.57 182 11.93 107 7/17/2017 134.87 76 390.37
*Cool season and total precipitation for 2011 were not included because meteorological measurements began on 1 April 2011.
39
Subseasonal Ponderosa Pine Tree-Ring Growth Response to Warm-Season
Precipitation amid Drought in Southern Nevada
Truettner, C.1, S.R. Poulson2, E. Ziaco1, and A.Z. Csank3
1DendroLab, Department of Natural Resources and Environmental Science, University of
Nevada, Reno, NV
2Department of Geological Sciences and Engineering, University of Nevada, Reno, NV
3Department of Geography, University of Nevada, Reno, NV
Keywords: Stable isotopes, ecohydrology, tree physiology, wood anatomy, high- resolution dendrochronology, Sheep Range, NevCAN, monsoon, tropical cyclones, extreme precipitation events
Summary
• The isotopic composition of source water and fractionations due to natural
processes, including photosynthesis and cellulose synthesis, affect the relative
abundance of stable isotopes in different plant tissues
• We measured stable isotopes (δ18O and δ2H) in precipitation, stem water and
annual tree-ring quartiles (δ18O and δ13C) while collecting microcores for
xylogenesis measurements in an old-growth ponderosa pine grove in southern
Nevada during the 2015 and 2016 warm season to measure how small versus
large trees respond to large precipitation events
• Subseasonal Pex, the proportion of oxygen molecules from sugars formed in
leaves that mixes with soil water at the site of cellulose synthesis in a tree,
differed between large (> 40.0 cm DBH and > 375 years old) and small (< 40.0
cm DBH and < 285 years old) trees among subseasonal climate 40
• Intrinsic water-use efficiency only differed in small trees during the post-
monsoonal subseason in 2015
• The isotopic signatures in small trees are more responsive to variation in
subseasonal hydroclimate in response to changes in water-use strategies,
although our evidence suggests large trees utilize deep soil water doing drier
subseasons
• If precipitation regimes shift with warmer temperatures, then deep soil water
recharge is likely less consistent intensifying the effects of drought on large trees
Introduction
High-resolution dendrochronology studies measure mechanisms of tree-ring growth in diverse forest types and the instruments needed to measure are becoming more accessible (Verheyden et al., 2004; Gebrekirstos et al., 2014; Treydte et al., 2014;
Szejner et al., 2020). Field studies combining measurements of stable isotope ratios
(δ13C and δ18O) in plant tissue with quantitative wood anatomy traits, such as cell wall thickness and lumen diameter, in tree rings allows for a more precise understanding of tree physiological response to seasonal hydroclimates (Belmecheri et al., 2018; Pacheco et al., 2020). Trees from subtropical climates influenced by monsoonal rains during the warm season (June-September) have clear seasonal isotopic signatures measured in the α-cellulose of their rings (Liu et al., 2008; Grießinger et al., 2011; Szejner et al.,
2016; Sano et al., 2017; Pumijumnong et al., 2020). The North American Monsoon
System (NAMS) core region has a bimodal precipitation regime of cool-season and late warm-season precipitation (Douglas et al., 1993; Adams and Comrie, 1997; Higgins,
Yao and Wang, 1997; Fuller and Stensrud, 2000; Vera et al., 2006; Higgins and Gochis,
2007), and this bimodal precipitation regime has been studied using the δ13C and δ18O 41
13 18 of alpha-cellulose (δ Ccell and δ Ocell, respectively) in the xylem tissue of tree rings
(Leavitt et al., 2002; Leavitt et al., 2011; Szejner et al., 2018). However, the subseasonal isotopic signature for the region, especially at the boundaries of the core region has not been completely resolved.
Earlywood, which is the lighter band of a tree ring with relatively wide and thin-walled cells, commonly starts at the beginning of the warm season (April – October) in the
NAMS core region and is associated with cool-season (previous November – March) precipitation from storms originating from midlatitude westerlies over the U.S. Pacific
Coast (Stahle et al., 2009; Carrillo et al., 2016). Latewood, which is the darker band of a tree ring with relatively narrow and thick-walled cells, begins to form during the late warm-season (July – September) and is associated with NAMS precipitation (Griffin et al., 2013; Meko and Baisan, 2001). False rings, also termed intra-annual density fluctuations (IADFs) in quantitative wood anatomy, can occur when a dry period occurs after initial radial growth and xylem cells begin to thicken like those found in latewood
(Battipaglia et al., 2016). False rings form in the transition zone when a tree begins growing from thicker to wider cells again after an influx of water stimulating cambial activity from increased stomatal conductivity (Battipaglia et al., 2014), usually from monsoonal precipitation events in the NAMS core region (Griffin et al., 2011). Stable
13 18 13 18 isotopes in the δ C and δ O of α-cellulose (δ CCELL and δ OCELL, respectively) in tree rings related to precise measurements of subseasonal tree-ring growth are used to investigate specific phenological questions like false-ring formation.
18 An important factor in the δ OCELL signature is the effect of PEX, which is the proportion of exchangeable oxygen at the site of cellulose synthesis from source water and leaf sugars (Farquhar and Lloyd, 1993; Barbour et al., 2004; Sternberg et al., 2006). 42
Source water is first pulled from soil water, taken up by the roots, and transported to the
18 canopy of the tree through xylem cells (McCarroll and Loader, 2004). The δ O of source water becomes more enriched in the leaf of the tree driven by evaporation and photosynthesis, and results in an enrichment of the δ18O in leaf water (Dongmann et al.,
18 1974; Farquhar and Lloyd, 1993; Cernusak et al., 2016). The δ O signature of leaf water is incorporated into sugars formed in the leaves of a tree, which are essentially non- structural carbohydrates (NSCs), and are loaded into the phloem and transported down the tree to the site of cellulose synthesis (Song et al., 2014). Gessler et al. (2009) summarized their field intensive study of Scots pine (Pinus sylvestris) and concluded
18 there was ~2 weeks lag effect from the enrichment of δ O in leaf water to when the sugars are transported down the stem through the phloem to the site of cellulose synthesis. The tree continues to pull source water from the soil during this lag effect, and
18 18 the source water has a more depleted δ O signature compared to the δ O of sugars in the phloem. The proportion of oxygen atoms from source water that mix with those from the sugars formed in the tree canopy is the value given for PEX. The value of PEX is usually around 0.42 (Sternberg et al., 1986; Cernusak et al., 2005), although theoretically it can range from 0.1 to 0.9 depending on the amount of evaporative enrichment of the δ18O in leaf water (Belmecheri et al., 2018) and the relatively depleted
18 δ O in source water (Gessler et al., 2009). In dendrochronological studies, tree cores are commonly extracted within 2 meters of the base of large trees. Therefore, the PEX derived from tree rings represents cellulose synthesis at the height of sample extraction of the tree, which is where the oxygen atoms from source water exchange with the oxygen atoms from sugars in the phloem. 43
Annual tree rings in temperate latitudes are comprised of xylem cells and the δ-
18 OCELL of α-cellulose in the cell walls of xylem cells can be measured at a subseasonal scale. The quantitative wood anatomy measurements of the formation of those cells is referred to as xylogenesis (Fukuda, 1996; Cuny and Rathgeber, 2016). Xylogenesis rates measured by quantitative wood anatomy throughout a growing season provides the timing of when cell walls thicken and produce α-cellulose. Intrinsic water-use efficiency (iWUE), the ratio of the rate of carbon assimilation to the stomatal
13 conductance to water in the leaf mesophyll tissue, is derived from the δ CCELL of tree rings (McCarroll and Loader, 2004). Xylogenesis and iWUE measurements quantify characteristics of tree-ring growth driven by transpiration from the tree canopy including bending stress and likelihood of cavitation (Martin-Benito et al., 2017).
Ponderosa pine (P. ponderosa ex. Lawson) is a widespread conifer species found in mid-elevation forests of western USA and is commonly investigated in tree physiology and paleoclimatological studies (Graham and Jain, 2005; Griffin et al., 2013; Williams et al., 2013; Truettner et al., 2018; Peltier and Ogle, 2019). It is a hydrophilic species, which makes it an ideal candidate tree species for studying plant-water-soil interactions
(Maherali and DeLucia, 2000; Remke et al., 2020). Rooting depth and tree canopy size affect PEX and iWUE. Yet, few studies have compared high-resolution dendrochronology in large (> 40.0 cm diameter at breast height [DBH] and > 375 years old) and small (<
40.0 cm DBH and < 285 years old) ponderosa pine trees, especially in old-growth forests in arid hydroclimates affected by seasonal precipitation patterns. Therefore, our primary research objective was to understand the ecophysiological response of ponderosa pine to pulses of precipitation in an arid hydroclimate amid drought. Our more specific
18 hypothesis tests if there are no differences in subseasonal δ OCELLl, iWUE, and PEX across a tree ring in large vs. small trees. 44
Methods Field site and collections
This study was conducted at the Sheep Range Montane site, hereafter referred to as Montane site. The site includes a monitoring station as part of the Nevada Climate- ecohydrological Assessment Network (NevCAN: Mensing et al., 2013), located in the
Sheep Range of southern Nevada, USA (~ 35 km north-northwest of Las Vegas). The
Sheep Range is a desert “sky-island” that supports conifer-dominated forests at mid-to- high elevation in the hyper-arid Mojave Desert of southern Nevada (Figure 1). A highly instrumented meteorological station located at the Montane site has been recording sub- hourly ecohydrological parameters since April 2011. The Montane site (2320 m a.s.l.) is positioned below wind-exposed mountain tops and on top of a ridgeline to limit discharge from adjacent watersheds (Mensing et al., 2013). Vegetation cover is composed of a mixed conifer stand including ponderosa pine (Pinus ponderosa), single-leaf pinyon pine
(P. monophylla) and sparse Utah junipers (Juniperus osteosperma). Shrub and herbaceous cover are present but at low densities. The soil type is classified as loamy- skeletal, mixed, superactive, mesic Aridic Lithic Haplustolls and colluvium and/or residuum from weathered dolomite and limestone (Johnson et al., 2014).
Microcores were sampled on a biweekly interval for twelve ponderosa pine individuals of different sizes located near the meteorological station. Two microcores (2 mm in diameter and ~ 35 mm long) were extracted from each tree using a Trephor
(Rossi et al., 2006) from March to October during 2015 and 2016. Microcores were collected from the same spot following a right-upward spiraling pattern, spacing each sampling point by ~5 cm to avoid the formation of traumatic resin ducts between successive collections. Microcores for isotopic measurements were placed in 2.0 mL 45
microcentrifuge tubes and sealed with parafilm, while those for xylogenesis were stored in a 50% alcohol/water solution and transported to the lab. One 12 mm short tree short core (~3 cm long) was also extracted from each tree on 7 November 2016 after the 2016 warm season.
A funneled precipitation collector was installed on 31 October 2014. Mineral oil was used in the collector to limit evaporative enrichment of the precipitation samples
(Friedman et al., 1992; Berkelhammer and Stott, 2008). Starting 4 April 2015, water from precipitation during the two-week sampling interval was collected, and mineral oil was replaced in the collector for the next two-week interval. Precipitation samples were collected on the same sampling dates as the microcores and represents the cumulative isotopic value for all precipitation events during the two-week interval.
Soil water was collected at the beginning and end of the growing seasons at two different depths: 10 cm and 20 cm. The initial installation was on 31 October 2014. To sample soil water, we used passive capillary wick lysimeters adapted after the design of Frisbee et al. (2008). A small amount of mineral oil was added to two 1 L plastic bottles, to prevent evaporation from the bottles. Bottles were placed inside a 5 gal bucket with a 1/2” diameter fiberglass wick sheathed in plastic inserted through a hole in the bucket lid and then into the top of each 1L bottle (one for 10 cm and one for 20 cm).
The bucket was sealed and buried at a depth slightly below 25 cm. The ends of the two wicks that were not inserted into the bottles were coiled into a flat disk and carefully inserted into the undisturbed soil face, one at 10 cm depth and the other at 20 cm depth.
Precipitation, stem, and soil water isotopes
Water samples were placed in a 2 ml microcentrifuge tube and centrifuged for 10 minutes at 10,000 RPM to separate water from mineral oil. The δ18O and δ2H 46
measurements were performed using a Picarro Cavity Ring-down Mass Spectrometer
(CRDMS) (L2102) water isotope analyzer housed in the Nevada Stable Isotope Lab at the University of Nevada, Reno. The samples were standardized to VSMOW (Vienna
Standard Mean Oceanic Water) using equation 1:
R δ = ( Sample – 1) X 1000 (1) RStandard
Where δ (in ‰) is the oxygen or hydrogen isotope composition (18O/16O or 2H/1H) in the water sample (RSample) relative to the standard VSMOW (RStandard). A local meteoric water line (LMWL) was derived from precipitation samples, to compare to the global meteoric water line (GMWL) published by Rozanski et al. (1993).
Regional scale measurements of integrated water vapor transport (IVT) from
NOAA Earth System Research Laboratory’s NCEP/NCAR Reanalysis portal
(www.esrl.noaa.gov/psd/data/composites/day/derived.html) for derived meteorological variables was used to assist in tracking the water vapor source for precipitation events
(Truettner et al., 2019). On an annual basis since 2011, most of the warm-season precipitation occurs during the late-summer after the onset of the NAMS in early July.
Precipitation events (> 2 mm in 24 hours) at the Montane site were investigated individually through analysis of regional shifts in IVT two days prior to and the day of the precipitation event (appendix C Storm Track Analysis).
The microcores collected for stem water extraction were pooled into two groups; one group included the microcores from large trees (DBH > 40.0 cm and > 375 years old) and the other for small trees (DBH < 40.0 cm and < 285 years old). The microcores from each group were combined to obtain an adequate water sample for isotopic analysis. There was likely a mix of phloem water and xylem water in the samples, 47
however the vast majority of plant tissue in microcores is xylem tissue. Therefore, to contribute our findings to modern leaf water isotope theory (Barbour et al., 2004;
Barbour et al., 2017), we consider the water extracted from the microcores as xylem
18 18 18 water. The δ O of xylem water (δ OXW) has been proven to be the same as δ O of soil water (i.e., source water) because no fraction was measured during the root uptake of
18 18 soil water (Ehleringer and Dawson, 1992). Hereon, we use the δ OXW in place of δ O of source water. Xylem water was extracted from each group of microcores (large and small) using a cryogenic extraction line at the Environmental Geochemistry Lab at the
Desert Research Institute. ChemCorrect™ software was used to identify and quantify any potential organic contamination in the xylem water samples.
18 13 High resolution δ OCELL and δ CCELL measurements
The 12-mm short cores were sliced into columnar rectangles and sanded on all sides. The 2015 and 2016 tree rings were measured on both radial sides of the sample.
These were averaged and measurements were divided into quartiles for the individual tree ring. We then used a microtome to slice the samples into tree-ring quartiles, wetting the samples in deionized water before every slice. Tree ring quartiles were sliced after the false ring (when present) to preserve the IADF that formed the false ring (Table 1).
The tree-ring quartiles were then processed from wholewood to α-cellulose.
Samples were processed using a modified version of the Leavitt and Danzer (1993) cellulose processing method after Rinne et al. (2005). Samples sat in an acidified sodium chlorite solution for 4 hours or overnight at 70˚C, the solution was removed via pipette, fresh solution was added, and samples then sat for another two hours at 70˚C in the solution. Samples were then rinsed in deionized water and ethanol. In order to process the resultant cellulose samples to purified α-cellulose, samples were placed in a 48
17% NaOH solution at 80˚C for 45 minutes followed by 10% HCl solution to neutralize the samples (Rinne et al., 2005; Anchukaitis et al., 2008). The α-cellulose samples were washed using deionized water and ethanol and dried overnight.
Isotope analyses of α-cellulose were performed using a Eurovector elemental analyzer interfaced to a Micromass IsoPrime stable isotope ratio mass spectrometer.
18 The δ OCELL analyses were performed using the pyrolysis method of Koziet (1997), were calibrated using benzoic acid standards IAEA-601 and IAEA-602 and standardized
13 to VSMOW using equation 1. The δ CCELL analyses were performed using a combustion method, were calibrated using an internal laboratory standard of acetanilide (previously calibrated vs. IAEA reference standards) and standardized to Vienna Pee Dee
Belemnite (VPDB) using equation 1.
Subseason classification
Six subseasons throughout the 2015 and 2016 warm season were classified using meteorological data from the Montane site and the IVT analysis. The six subseasons include: 2015 early warm season (EWS2015); 2015 monsoon season
(MS2015); 2015 post-monsoon (PM2015); 2016 early warm season (EWS2016); 2016 monsoon season (MS2016); and 2016 post-monsoon season (PM2016). April 1 of both years marked the beginning of the warm season and, therefore, the first day of EWS2015 and EWS2016. The monsoon onset threshold of 9.4 ˚C daily dewpoint temperature from
Truettner et al., (2019) was used to separate the early warm season from the monsoon season. The monsoon onset date for MS2015 was 2 July 2015, and the monsoon onset date for MS2016 was 1 July 2016. Our IVT analysis for sourcing water vapor for precipitation events at the Montane site determined the end of the monsoon season and beginning of the post-monsoon season for the 2015 and 2016 warm season (Appendix 49
C). IVT during the monsoon season originated from subtropical sources, while IVT during the post-monsoon season was a mixture of moisture from both subtropical and mid-latitude eastern Pacific sources. The PM2015 began on 14 September 2015 after a strong monsoon season in 2015, and the PM2016 began on 25 August 2016 after a weak monsoon season in 2016. Both post-monsoon seasons ended on October 31 before the start of the cool season.
Phenodate calculations
A “phenodate” is the term used to define the temporal period over which unique plant tissue is formed during the growing season. This is similar to the “xylogenesis windows” used in Belmecheri et al. (2018) but can be used for any unique plant tissue that formed during the growing season. In this study, the phenodate of cell wall tissue was measured in the xylem sliced into annual tree-ring quartiles. Biweekly wood anatomical measurements of lumen diameter and cell wall thickness allowed for precise dating of the tree-ring quartile phenodate. Lumen diameter measures the diameter of the space in xylem tissue cells found in a tree ring, and cell wall thickness is the length of the primary cell wall for xylem tissue cells. For each tree and year, the date of appearance (expressed as day of the year, or DOY) of the first cell in each tree-ring quartile was determined using a Gompertz function calculated for that individual tree
(Ziaco, 2020) The ending date for each tree-ring quartile was defined as the day in which the last cell of the quartile had completed its maturation according to the Gompertz function (Table 1).
18 Modeling PEX for subseasonal δ OCELL
Tree-ring growth quartile phenodates, hereon referred to as phenodates, allowed us to assign which of the six subseasons a specific tree-ring growth quartile formed. 50
Figure 2 is a visual example of tree-ring growth quartiles from the 2015 and 2016 tree ring, which included a “false ring” in the 2015 tree ring. Little tree-ring growth was measured during the EWS2015 (Ziaco et al., 2018). Therefore, EWS2015 was not analyzed in the modeling component of this study. Tree-ring growth quartiles were classified into subseasonal groups according to phenodate.
The PEX for each tree-ring growth quartile was calculated using the model presented by Szejner et al., (2020) that combined the mechanistic models for modeling
18 18 δ OCELL (Roden et al., 2000; Barbour et al., 2004) and δ O in leaf water (Dongmann et al., 1974; Farquhar, Ehleringer and Hubick, 1989; Flanagan et al., 1991; Roden and
Ehleringer, 1999):
18 18 ∗ ∗ ea δ OCELL –[δ OXW + εc + εk+ε + (−ε − εk)( )] ei PEX = (2) ∗ ∗ ea −(εk+ε + (−ε − εk)( )) ei
Where PEX is the proportion of oxygen atoms in cellulose that exchange with oxygen
18 from local waters at the site of cell wall thickening, δ OXW is the oxygen isotopic composition of xylem water, εc is the equilibrium fractionation factor between water and carbonyl oxygen (27 ‰) , εk is the kinetic fractionation associated with the diffusion of water through the stomata and leaf boundary layer (28 ‰), ε* is the equilibrium fractionation associated with the transition of liquid water into water vapor, ea is the water vapor pressure for the ambient air surrounding the leaf-boundary layer, and ei is the intercellular water vapor pressure found within the leaf.
Dividing ea by ei relates the vapor pressure from ambient air to that found in the leaf (Dongmann et al., 1974; Farquhar et al., 1989; Flanagan et al., 1991; Roden and
Ehleringer, 1999). This theoretical relationship drives stomata to open and close allowing carbon dioxide (used in carboxylation during photosynthesis) to enter the leaf while 51
oxygen and water vapor exit the leaf into the atmosphere. However, measuring ea and ei is nearly impossible, especially due to the remoteness of the Montane site. We assumed that the leaf was saturated during the phenodate because theoretically a leaf should be saturated in an arid hydroclimate to be able to open stomata. Saturated vapor pressure was used to estimate ei and was calculated from subhourly measurements (relative humidity, barometric pressure, and air temperature) from a meteorological station at the
Montane site (within 100 m of the ponderosa pines) averaged over each phenodate
(Szejner et al., 2018). We used actual vapor pressure to represent ea and derived the value from subhourly measurements (saturated vapor pressure, relative humidity, and barometric pressure) from the same meteorological station averaged over each phenodate similar to ei. Refer to Truettner et al. (2019) for methods of calculating actual and saturated vapor pressure from the Montane site. We used the value of 28‰ for εk
(Buhay et al., 1996), 27‰ as the value for εc based on Sternberg et al. (1986), and ε* was calculated for each phenodate using the experimentally established equation from
Majoube (1971):
0.4156 1137 ε ∗ = 0.0020667 + ( ) – ( ) (3) T T2
Where T is air temperature in Kelvin measured from the Montane site and was averaged for each phenodate.
Intrinsic water use efficiency
Intrinsic water use efficiency (iWUE) was calculated for each phenodate using
13 the δ CCELL of the tree-ring growth quartiles to quantify the ratio of net carbon assimilation to stomatal conductance to water (A/gw). Carbon isotope discrimination
13 13 13 13 (∆ C) was calculated between the δ C of plant tissue (i.e., δ CCELL) and the δ C of 52
atmospheric CO2 used in photosynthesis and carboxylation to provide sugars for plant tissue formation (Farquhar and Richards, 1984):
13 13 13 δ CATM – δ CPLANT ∆ C = δ13C (4) 1 + PLANT 1000
13 13 13 13 whereas δ CATM is the δ C of atmospheric CO2, and δ CPLANT is the δ C of the plant
13 tissue. The ∆ C is related to the ratio of intercellular to atmospheric CO2 mole fractions
(Farquhar et al., 1982):
푐 ∆13C = 푎 + (푏 – 푎) 푖 (5) 푐푎
Whereas ci and ca are the mole fractions of intercellular and atmospheric CO2, respectively, a is the fractionation that occurs when CO2 enters the leaf through the stomata (4.4‰), and b is the fractionation associated with Rubisco and PEP carboxylase during photosynthesis and carboxylation (27‰, Farquhar and Richards, 1984). We
13 averaged weekly measurements of δ CATM and ca (measured in ppm) from the Mauna
Loa Observatory (www.esrl.noaa.gov/gmd/ccgg/trends/data.html) over the phenodate.
The stomatal conductance to CO2 (gc) and water vapor (gw) relates the leaf-gas exchange of water and carbon (i.e., plant-water interactions) and is a constant relationship (gw = 1.6*gc: Farquhar and Ehleringer, 1989). The iWUE can then be calculated (McCarroll and Loader, 2004):
푐 푏− ∆13퐶 푖푊푈퐸 = 푎 × (6) 1.6 푏−푎
In arid hydroclimates, where stomatal conductance tends to dominate (McCarroll et al.,
2009), high iWUE indicates lower transpiration rates and an increase of water-saving strategies in a tree. 53
Statistical analysis
18 We addressed our hypothesis of no differences in the subseasonal δ OCELL, iWUE, and PEX between small and large trees through an analysis of variance (ANOVA) and Wilcoxon rank sum test in the statistical language R (R Core Team, 2019).
18 Response variables in the ANOVA included the δ OCELL, iWUE, and PEX for each tree- ring growth quartile, and size of tree (small vs. large) and subseason were used as predictor variables. We used a Tukey’s Honestly Significant Difference test to do a
18 pairwise comparison of subseasonal δ OCELL, iWUE, PEX to test significant differences in
18 subseasonal means within large and small trees. Subseasonal δ OCELL, iWUE, and PEX measurements were not normally distributed when isolated to subseason (i.e., not large or small trees), therefore we used the Wilcoxon rank sum test to identify particular significant differences between subseasons in small vs. large trees.
RESULTS
Precipitation, stem water, and soil water isotopes
The LMWL for the 2015 and 2016 warm seasons was calculated using twenty- one precipitation samples from 31 October 2014 to 7 October 2016 (Figure 2a). The
LMWL falls on the GMWL. The δ18O and δ2H consistently fell below the GMWL for large
18 and small trees, indicating an enrichment of the δ OXW during the study period, while the
δ18O and δ2H of soil water was more consistent with precipitation.
18 18 The δ OXW was consistently enriched vs. synchronous δ OPPT during 2015 for both large and small trees (Figure 3a,3b). A gradual enrichment of 11.3‰ was measured
18 18 in δ OPPT during the EWS2015 as the δ OXW for large and small trees followed a similar
18 18 pattern as the δ OPPT (Figure 2b). Although, the δ OXW was consistently 3-4‰ more 54
18 18 enriched than δ OPPT during the EWS2015. A depletion in the δ OXW for large and small
18 trees coincided with a depletion of δ OPPT measured after the arrival of two large storms on 21 July 2015 (46.2 mm) and 2 August 2015 (57.4 mm) (Figure 3b). Few precipitation events (3.1 mm total precipitation) occurred for the remaining of MS2015 (129.8 mm total precipitation) from 9 August 2015 to 13 September 2015. Patterns in integrated water vapor transport (IVT) switched from subtropical during MS2015 to a mix of subtropical and
18 Pacific frontal sources during the PM2015 (90.2 mm total precipitation). The δ OXW in
18 small trees became enriched during the PM2015 (Figure 3b), while the δ OXW of large trees decreased (~2‰) over this time period.
Precipitation events were frequent during the EWS2016 (141.9 mm total precipitation) when compared to the drier EWS2015 (58.4 mm total precipitation). No
18 18 difference (1 s.d.) was found between δ OXW and δ OPPT for both large and small trees during the EWS2016 (Figure 4a, 4b). Four sparse precipitation events (> 2 mm) occurred for the MS2016 (44.2 mm total precipitation) and PM2016 (36.1 mm total precipitation),
18 indicative of weak or absent monsoonal precipitation. The δ OPPT was consistently
18 enriched vs. δ OXW in large and small trees during MS2016 (~4‰ and 3‰, respectively) and PM2016 (~3‰ and 2‰, respectively), which is in contrast to the 2015 warm season
18 18 when δ OPPT was depleted vs. δ OXW for large and small trees during MS2015 (~4‰) and
PM2015 (~3‰ and 5%) (Figure 10a,10b). A slight enrichment (~1 ‰) from MS2015 to
18 PM2015 was measured in the δ OXW of small trees while there was no enrichment in the
18 18 δ OXW from MS2015 to PM2015 in large trees. The δ OPPT was more depleted during the
18 MS2015 compared to MS2016 (~3 ‰). Similarly, the δ OPPT was more depleted during the
18 PM2015 compared to PM2016 (~5 ‰). In contrast, the δ OXW was more enriched in MS2015 compared to MS2016 (~6‰), as well as PM2016 compared to PM2016 (~5‰). 55
No soil water sample was present after the dry 2014-2015 cool season. Soil water at 10 cm depth was only present after the 2015 warm season and was the most depleted sample in δ18O for the entire study (-18.8‰). The δ18O of soil water at 10 cm depth and at 20 cm depth were ~-11‰ at the beginning of the EWS2016. By the end of the 2016 warm season, soil water at 10 cm depth was -6.2‰ while soil water at 20 cm depth was comparable to the sample collected in April 2016 with about a 0.1‰ difference (Figure 3b).
18 Subseasonal δ OCELL, iWUE, and PEX
18 The average δ OCELL was slightly more enriched for small trees (33.1 ± 1.2‰) than large trees (32 ± 1.2‰) (Figure 4c,d). Subseason (F = 10.2, p < 0.0001) and size of
18 18 tree (F = 18.7, p < 0.0001) had separate significant relationships on δ OCELL (δ OCELL
13 and δ CCELL raw tree-ring quartile measurements in Appendix D). The average iWUE was equivalent in small trees (137.2 ± 13.2 mmol) and large trees (135.2 ± 13.1‰)
(Figure 4e,f), and subseason (F = 2.9, p < 0.03) had a significant relationship on iWUE, while size of tree had no significant effect on iWUE.
18 The highest mean for subseasonal δ OCELL was for PT2015 in small trees (34.7 ±
18 1.3‰) and was significantly different than the δ OCELL for the other subseasons (Figure
18 4d). The lowest mean for subseasonal δ OCELL was for PM2016 (30.3 ± 1.2‰) and was significantly different from the other subseasons in large trees (Figure 4c). A more
18 enriched δ OCELL (~2‰) during MS2016 for large trees was significantly different than
MS2015 for large trees, while no significant difference was measured between monsoon seasons for small trees (Appendix E). Small trees were more enriched than large trees for PM2015 (~2‰), EWS2015 (~1‰), and PM2016 (~2.5‰) and significant differences in
18 subseasonal δ OCELL between large vs. small trees were found for PM2015, EWS2016, and 56
18 PM2016. No significant differences between the δ OCELL of large and small trees were found for the monsoon subseasons (Figure 4c,d).
The iWUE for MS2015 in small trees was higher than PM2015 by 22.4 mmols and was the only significantly difference for subseasonal iWUE measurements (Figure 4f).
Subseasonal PEX for all subseasons in large and small trees was within the range of other ponderosa pine studies. The average PEX for MS2015 (0.5 ± 0.05) and PM2016
(0.53 ± 0.05) for large trees were higher than PM2015 (0.43 ± 0.04), EWS2016 (0.42 ±
0.04), and MS2016 (0.45 ± 0.05) for large trees (Figure 5a). The PEX for MS2015 and PM2016 in large trees was significantly different for PM2015, EWS2016, and marginally significantly different for MS2016 in large trees (Appendix E). The subseasonal PEX for small trees varied more than large trees and significant differences were measured between different subseasons (Fig. 5a,b). The PEX for EWS2016 was lower and significantly different in small trees (0.36 ± 0.02) than large trees. Likewise, the PEX for PM2016 was lower and significantly different in small trees (0.43 ± 0.03) than large trees. Oppositely, the PEX for MS2016 was higher and significantly different in small trees (0.51 ± 0.04) than large trees.
DISCUSSION
Although our sample size was minimal for some subseasons, differences
18 between large vs. small ponderosa pine subseasonal δ OCELL and PEX were measured at the NevCAN Montane site in the Sheep Range of southern Nevada, USA. The most
18 18 noticeable difference in subseasonal δ OCELL was the enrichment of δ OCELL in small trees vs. large trees during PM2015 after two extreme precipitation events in MS2015. The first of these storms, a recurved tropical cyclone remnant from Hurricane Dolores on 21
July 2015, was the largest storm on record at the Montane site since its initial 57
measurements on 1 April 2011 (Truettner et al., 2019). That was until the second storm, characteristic of a convective afternoon thunderstorm common in the NAMS core region during the late warm season, precipitated at the Montane site on 2 August 2015. These
18 two extreme precipitation events were depleted in δ OPPT and initiated radial growth for
18 most trees in our study (Ziaco et al., 2018). The δ OXW for both large and small trees experienced a depletion during these two storm periods, likely in response to the influx
18 of precipitation depleted δ OPPT.
18 18 The δ OXW and δ OCELL remained consistent in large trees after the two extreme precipitation events through the remaining of the warm season until tree-ring growth
18 18 ceased around mid-October. An enrichment of δ OXW and δ OCELL was measured in small trees during the same period. The iWUE for small trees decreased significantly during the PM2015 suggesting a relief from water-saving strategies and an increase in transpiration. Belmecheri et al. (2018) found a similar enrichment pattern of post-
18 monsoon δ OCELL after a false ring in ponderosa pine in the NAMS core region of southern Arizona that corresponded with a phenological lag explained by the rate of cell
18 wall thickening. The δ OCELL is positively influenced by an increase in vapor-pressure deficit (i.e., drier air) (Song et al., 2014; Kerhlous et al., 2017) and the enriched isotopic
18 18 signature in the δ OCELL of small trees could be representative of the δ O in sugars formed during the drier August and early September after the two extreme precipitation events. Interestingly, the PEX significantly decreased in both large trees and small trees and may reflect an increase in lumen diameter when vapor-pressure deficit decreased with the arrival of a second set of Pacific frontal storms during the PM2015. The number of hexose molecules that stay in the leaf to undergo triose-phosphate cycling in large and small trees would have decreased and moved rapidly down the phloem to the site of 58
cellulose synthesis identified by a lower PEX and an increase in lumen diameter (Song et al., 2014; Szejner et al., 2020).
The tree-ring growth at the beginning of the 2015 tree ring under a microscope may appear to be earlywood followed by a false ring and then latewood. However, precise micrometeorological, xylogenesis and stable isotopes measurements suggest that the 2015 tree ring should be defined by two extreme precipitation events followed by a dry period followed by a series of Pacific frontal storms in late September and October.
The combination of evidence suggests that drier years during drought might be completely driven by late warm-season precipitation distorting the bimodal precipitation regime signal in dendroclimatological reconstructions. Further studies similar to Szejner et al. (2016) at a region scale could identify if this is the case for isotopic signature in trees in the core of monsoonal regions compared to those found on the boundary of the regional monsoon precipitation regime.
The most distinct difference between the 2015 and 2016 warm seasons was the
18 18 transition from enriched water in the xylem compared to precipitation (δ OXW > δ OPPT)
18 during MS2015 and PM2015 to more enriched water in precipitation than the xylem (δ OXW
18 < δ OPPT) in MS2016 and PM2016. This reversal is likely linked to a shift in the bimodal precipitation regime. A larger snowpack and consistent spring precipitation from Pacific frontal storms established a less water-stressed hydroclimate during the EWS2016 compared to EWS2015 (Truettner et al., 2019). Precipitation was scarce during the MS2016 and PM2016 with little to no association with the weak NAMS of 2016 (Sierks et al., 2020).
The soil water at 10 cm depth was indicative of this shift with no soil water being
18 measured after the 2014 – 2015 cool season to the most depleted δ O value after the wetter 2015 late warm season to the most enriched δ18O value at the end of the drier 59
2016 late warm season. No difference was measured in the subseasonal iWUE of small and large trees throughout the 2016 warm season. Although, PEX was significantly
18 different between large and small trees for every subseason. The δ OXW was more enriched in small trees than large trees during the MS2015, which coincided with a higher
18 PEX value in small trees. This difference with an enrichment of δ OXW and increase in
PEX was different than what we measured in PM2015 when small trees switched to a less water-stressed signal. This indicates that the hexophosphates turnover time would have decreased and there was more oxygen exchange during triose-phosphate cycling leading to enriched leaf water during the weak monsoon. Therefore, small trees respond to a drier hydroclimate during MS2016 even though it is not recorded in the subseasonal
18 δ OCELL.
18 The most depleted subseasonal δ OCELL of the study was for PM2016 in large trees and was significantly lower than small trees. It is likely that both large and small trees would be more water-stressed by the end of August and September during the
PM2016 after a drier late warm season. An increase in iWUE was measured, although not significant, and the non-significant relationship could be due to a lower sample size.
18 Nonetheless, the significant depletion in subseasonal δ OCELL in large trees suggests that large trees were using a different water source than small trees. Rooting depth may be the best explanation for this difference between large vs. small trees because large trees may have used tightly bound soil water in deeper soils leftover from storms
18 depleted in δ OPPT at the end of the 2015 warm season after soil water in wider pores had been used during the drier MS2016 (Brooks et al., 2010; Kerhoulas et al., 2013).
Further studies with more frequent samples at different soil depths would provide better insight on these plant-water-soil interactions. 60
CONCLUSION
Ponderosa pine at the Montane site display high phenological plasticity for both large and small trees, which could explain the quick radial growth response to the two extreme precipitation events in 2015 during a multi-year severe drought (Ziaco et al.,
2018; Hatchett et al., 2015). Increased extreme precipitation events and shifts in hydroclimate, like the NAMS, may be a result of 21st century abrupt climate change
(Cook and Seager, 2013; Dominguez et al., 2012). The ponderosa pine haplotype in the
Sheep Range of Nevada has the highest heterozygosity of any measured ponderosa pine haplotype (Potter et al., 2015) and is influenced by both cool-season and warm- season precipitation climatic niches (Shinneman et al., 2016). Further studies investigating subseasonal ponderosa pine radial growth to shifts in hydroclimate in common gardens like the Southwest Experimental Garden Array (Cooper et al., 2019) could compare ecophysiological responses of ponderosa pine haplotypes throughout the deserts of North America. If the Sheep Range ponderosa pine haplotype has a higher survival and fecundity rate than other haplotypes, then government agencies and public stakeholders should consider the Sheep Range ponderosa pine haplotype as a candidate for assisted migration strategies.
Phenodates in combination with precise measurements of ecophysiological
18 parameters that affect the δ OCELL in tree rings provided a unique opportunity to measure subseasonal patterns in the isotopic signature and water-use strategies of an old-growth ponderosa pine grove during drought. Our results suggest that small trees are more responsive to variation in subseasonal hydroclimate than large trees in an arid hydroclimate. However, we have evidence that large trees have the ability to use water reserves deep in the soil that small trees cannot access. Further studies using isotopes 61
and quantitative wood anatomy that investigate differences in drought legacy effect
(Anderegg et al., 2015) of small vs. large trees after a drought-relief period could further explain mechanisms driving high vs. low phenological plasticity in different sizes and populations of trees in water-stressed hydroclimates. Worldwide, large trees are thought to be more susceptible to drought with relatively smaller water reserves (Bennett et al.,
2015). Extreme precipitation events measured during the 2015 warm season provided drought relief for large and small trees at the end of the severe drought, but extreme precipitation events are hard to predict and less consistent than monsoon-associated precipitation. Evaporative demand continues to increase in the deserts of North America and affects deep water soil recharge in hydroclimates that large trees established in previous centuries. Large trees in old-growth groves in southern Nevada are most threatened by a more variable hydroclimate.
Acknowledgments
We would like to thank our funding source from the P2C2 program at the National
Science Foundation. We express our gratitude to F. Biondi, M. Dettinger, and M. Sierks for their helpful insights and reviews during the development and writing of the manuscript. We would also like to thank the MtnClim community for their discussions, interpretations, and enthusiasm during the study.
References
Adams, D. K. and Comrie, A. C. (1997) The north American monsoon. Bulletin of the American Meteorological Society, 78(10), pp. 2197-2214.
Anderegg, W. R., Schwalm, C., Biondi, F., Camarero, J. J., Koch, G., Litvak, M., ... & Wolf, A. (2015). Pervasive drought legacies in forest ecosystems and their implications for carbon cycle models. Science, 349(6247), 528-532. 62
Anchukaitis, K. J., Evans, M. N., Wheelwright, N. T. and Schrag, D. P. (2008) Stable isotope chronology and climate signal calibration in neotropical montane cloud forest trees, Journal of Geophysical Research: Biogeosciences, 113(G3).
Barbour, M., Farquhar, G. and Buckley, T. (2017) Leaf water stable isotopes and water transport outside the xylem, Plant, Cell & Environment, 40(6), pp. 914-920.
Barbour, M. M., Roden, J. S., Farquhar, G. D. and Ehleringer, J. R. (2004) Expressing leaf water and cellulose oxygen isotope ratios as enrichment above source water reveals evidence of a Péclet effect, Oecologia, 138(3), pp. 426-435.
Bates, D., Maechler, M., Bolker, B., Walker, S., Christensen, R. H. B., Singmann, H., Dai, B., Grothendieck, G., Green, P. and Bolker, M. B. (2015) Package lme4, Convergence, 12(1), pp. 2.
Battipaglia, G., Campelo, F., Vieira, J., Grabner, M., De Micco, V., Nabais, C., Cherubini, P., Carrer, M., Bräuning, A. and Čufar, K. (2016) Structure and function of intra–annual density fluctuations: mind the gaps, Frontiers in plant science, 7, pp. 595.
Battipaglia, G., De Micco, V., Brand, W. A., Saurer, M., Aronne, G., Linke, P. and Cherubini, P. (2014) Drought impact on water use efficiency and intra‐annual density fluctuations in E rica arborea on E lba (I taly), Plant, cell & environment, 37(2), pp. 382- 391.
Belmecheri, S., Wright, W. E., Szejner, P., Morino, K. A. and Monson, R. K. (2018) Carbon and oxygen isotope fractionations in tree rings reveal interactions between cambial phenology and seasonal climate, Plant, cell & environment, 41(12), pp. 2758- 2772.
Bennett, A. C., McDowell, N. G., Allen, C. D. and Anderson-Teixeira, K. J. (2015) Larger trees suffer most during drought in forests worldwide, Nature Plants, 1(10), pp. 15139.
Berkelhammer, M. B. and Stott, L. D. (2008) Recent and dramatic changes in Pacific storm trajectories recorded in δ18O from Bristlecone Pine tree ring cellulose, Geochemistry, Geophysics, Geosystems, 9(4). 63
Brooks, J. R., Barnard, H. R., Coulombe, R. and McDonnell, J. J. (2010) Ecohydrologic separation of water between trees and streams in a Mediterranean climate, Nature Geoscience, 3(2), pp. 100-104.
Buhay, W., Edwards, T. and Aravena, R. (1996) Evaluating kinetic fractionation factors used for ecologic and paleoclimatic reconstructions from oxygen and hydrogen isotope ratios in plant water and cellulose, Geochimica et Cosmochimica Acta, 60(12), pp. 2209- 2218.
Carrillo, C. M., Castro, C. L., Woodhouse, C. A. and Griffin, D. (2016) Low‐frequency variability of precipitation in the North American monsoon region as diagnosed through earlywood and latewood tree‐ring chronologies in the southwestern US, International Journal of Climatology, 36(5), pp. 2254-2272.
Cernusak, L. A., Barbour, M. M., Arndt, S. K., Cheesman, A. W., English, N. B., Feild, T. S., Helliker, B. R., Holloway‐Phillips, M. M., Holtum, J. A. and Kahmen, A. (2016) Stable isotopes in leaf water of terrestrial plants, Plant, Cell & Environment, 39(5), pp. 1087- 1102.
Cernusak, L. A., Farquhar, G. D. and Pate, J. S. (2005) Environmental and physiological controls over oxygen and carbon isotope composition of Tasmanian blue gum, Eucalyptus globulus, Tree physiology, 25(2), pp. 129-146.
Cook, B. and Seager, R. (2013) The response of the North American Monsoon to increased greenhouse gas forcing, Journal of Geophysical Research: Atmospheres, 118(4), pp. 1690-1699.
Cooper, H. F., Grady, K. C., Cowan, J. A., Best, R. J., Allan, G. J. and Whitham, T. G. (2019) Genotypic variation in phenological plasticity: Reciprocal common gardens reveal adaptive responses to warmer springs but not to fall frost, Global change biology, 25(1), pp. 187-200.
Cuny, H. E. and Rathgeber, C. B. (2016) Xylogenesis: coniferous trees of temperate forests are listening to the climate tale during the growing season but only remember the last words!, Plant physiology, 171(1), pp. 306-317. 64
Dominguez, F., Rivera, E., Lettenmaier, D. and Castro, C. (2012) Changes in winter precipitation extremes for the western United States under a warmer climate as simulated by regional climate models, Geophysical Research Letters, 39(5).
Dongmann, G., Nürnberg, H., Förstel, H. and Wagener, K. (1974) On the enrichment of H 2 18 O in the leaves of transpiring plants, Radiation and environmental biophysics, 11(1), pp. 41-52.
Douglas, M. W., Maddox, R. A., Howard, K. and Reyes, S. (1993) The mexican monsoon, Journal of Climate, 6(8), pp. 1665-1677.
Ehleringer, J. and Dawson, T. (1992) Water uptake by plants: perspectives from stable isotope composition, Plant, cell & environment, 15(9), pp. 1073-1082.
Farquhar, Graham D., Marion H. O'Leary, and Joe A. Berry. "On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves." Functional Plant Biology 9.2 (1982): 121-137.
Farquhar, G. D., & Richards, R. A. (1984). Isotopic composition of plant carbon correlates with water-use efficiency of wheat genotypes. Functional Plant Biology, 11(6), 539-552.
Farquhar, G. D., Ehleringer, J. R. and Hubick, K. T. (1989) Carbon isotope discrimination and photosynthesis, Annual review of plant biology, 40(1), pp. 503-537.
Farquhar, G. D. and Lloyd, J. (1993) Carbon and oxygen isotope effects in the exchange of carbon dioxide between terrestrial plants and the atmosphere, Stable isotopes and plant carbon-water relations: Elsevier, pp. 47-70.
Flanagan, L. B., Comstock, J. P. and Ehleringer, J. R. (1991) Comparison of modeled and observed environmental influences on the stable oxygen and hydrogen isotope composition of leaf water in Phaseolus vulgaris L, Plant Physiology, 96(2), pp. 588-596.
Friedman, I., Smith, G. I., Gleason, J. D., Warden, A. and Harris, J. M. (1992) Stable isotope composition of waters in southeastern California 1. Modern precipitation, Journal of Geophysical Research: Atmospheres, 97(D5), pp. 5795-5812.
Frisbee, M. D., Phillips, F. M., Campbell, A. R. and Hendrickx, J. M. (2010) Modified passive capillary samplers for collecting samples of snowmelt infiltration for stable 65 isotope analysis in remote, seasonally inaccessible watersheds 1: laboratory evaluation, Hydrological Processes: An International Journal, 24(7), pp. 825-833.
Fukuda, H. (1996) Xylogenesis: initiation, progression, and cell death, Annual review of plant biology, 47(1), pp. 299-325.
Fuller, R. D. and Stensrud, D. J. (2000) The relationship between tropical easterly waves and surges over the Gulf of California during the North American monsoon, Monthly weather review, 128(8), pp. 2983-2989.
Gebrekirstos, A., Bräuning, A., Sass-Klassen, U. and Mbow, C. (2014) Opportunities and applications of dendrochronology in Africa, Current Opinion in Environmental Sustainability, 6, pp. 48-53.
Gessler, A., Brandes, E., Buchmann, N., Helle, G., Rennenberg, H. and Barnard, R. L. (2009) Tracing carbon and oxygen isotope signals from newly assimilated sugars in the leaves to the tree‐ring archive, Plant, Cell & Environment, 32(7), pp. 780-795.
Graham, R. T. and Jain, T. B. Ponderosa pine ecosystems. In: Ritchie, Martin W.; Maguire, Douglas A.; Youngblood, Andrew, tech. coordinators. Proceedings of the Symposium on Ponderosa Pine: Issues, Trends, and Management, 2004 October 18-21, Klamath Falls, OR. Gen. Tech. Rep PSW-GTR-198. Albany, CA: Pacific Southwest Research Station, Forest Service, US Department of Agriculture: 1-32., 1-32.
Grießinger, J., Bräuning, A., Helle, G., Thomas, A. and Schleser, G. (2011) Late Holocene Asian summer monsoon variability reflected by δ18O in tree‐rings from Tibetan junipers, Geophysical Research Letters, 38(3).
Griffin, D., Meko, D. M., Touchan, R., Leavitt, S. W. and Woodhouse, C. A. (2011) Latewood Chronology Development for Summer-Moisture Reconstruction In the US Southwest, Tree-Ring Research, 67(2), pp. 87-101, 15.
Griffin, D., Woodhouse, C. A., Meko, D. M., Stahle, D. W., Faulstich, H. L., Carrillo, C., Touchan, R., Castro, C. L. and Leavitt, S. W. (2013) North American monsoon precipitation reconstructed from tree‐ring latewood, Geophysical Research Letters, 40(5), pp. 954-958. 66
Hatchett, B. J., Boyle, D. P., Putnam, A. E. and Bassett, S. D. (2015) Placing the 2012– 2015 California‐Nevada drought into a paleoclimatic context: insights from Walker Lake, California‐Nevada, USA, Geophysical Research Letters, 42(20), pp. 8632-8640.
Higgins, R., Yao, Y. and Wang, X. (1997) Influence of the North American monsoon system on the US summer precipitation regime, Journal of Climate, 10(10), pp. 2600- 2622.
Higgins, W. and Gochis, D. (2007) Synthesis of results from the North American Monsoon Experiment (NAME) process study, Journal of Climate, 20(9), pp. 1601-1607.
Johnson, B. G., Verburg, P. S. and Arnone, J. A. (2014) Effects of climate and vegetation on soil nutrients and chemistry in the Great Basin studied along a latitudinal- elevational climate gradient, Plant and soil, 382(1-2), pp. 151-163.
Kerhoulas, L. P., Kolb, T. E. and Koch, G. W. (2013) Tree size, stand density, and the source of water used across seasons by ponderosa pine in northern Arizona, Forest Ecology and Management, 289, pp. 425-433.
Kerhoulas, L. P., Kolb, T. E. and Koch, G. W. (2017) The influence of monsoon climate on latewood growth of southwestern ponderosa pine, Forests, 8(5), pp. 140.
Koziet, J. (1997) Isotope ratio mass spectrometric method for the on-line determination of oxygen-18 in organic matter. Journal of Mass Spectrometry, 32: 103-108.
Leavitt, S. W. and Danzer, S. R. (1993) Method for batch processing small wood samples to holocellulose for stable-carbon isotope analysis, Analytical Chemistry, 65(1), pp. 87-89.
Leavitt, S. W., Woodhouse, C. A., Castro, C. L., Wright, W. E., Meko, D. M., Touchan, R., Griffin, D. and Ciancarelli, B. (2011) The North American monsoon in the US Southwest: Potential for investigation with tree-ring carbon isotopes, Quaternary International, 235(1-2), pp. 101-107.
Leavitt, S. W., Wright, W. E. and Long, A. (2002) Spatial expression of ENSO, drought, and summer monsoon in seasonal δ13C of ponderosa pine tree rings in southern Arizona and New Mexico, Journal of Geophysical Research: Atmospheres, 107(D18), pp. ACL 3-1-ACL 3-10. 67
Liu, Y., Cai, Q., Liu, W., Yang, Y., Sun, J., Song, H. and Li, X. (2008) Monsoon precipitation variation recorded by tree-ring δ18O in arid Northwest China since AD 1878, Chemical Geology, 252(1-2), pp. 56-61.
Maherali, H. and DeLucia, E. H. (2000) Xylem conductivity and vulnerability to cavitation of ponderosa pine growing in contrasting climates, Tree Physiology, 20(13), pp. 859- 867.
Majoube, M. (1971) Fractionnement en oxygene 18 et en deuterium entre leau et sa vapeur, Journal de Chimie Physique, 68, pp. 1423-1436.
Martin-Benito, D., Anchukaitis, K. J., Evans, M. N., Del Río, M., Beeckman, H. and Cañellas, I. (2017) Effects of drought on xylem anatomy and water-use efficiency of two co-occurring pine species, Forests, 8(9), pp. 332.
McCarroll, D. and Loader, N. J. (2004) Stable isotopes in tree rings, Quaternary Science Reviews, 23(7-8), pp. 771-801.
Meko, D. M. and Baisan, C. H. (2001) Pilot study of latewood‐width of conifers as an indicator of variability of summer rainfall in the North American monsoon region, International Journal of Climatology: A Journal of the Royal Meteorological Society, 21(6), pp. 697-708.
Mensing, S., Strachan, S., Arnone, J., Fenstermaker, L., Biondi, F., Devitt, D., Johnson, B., Bird, B. and Fritzinger, E. (2013) A network for observing Great Basin climate change, Eos, Transactions American Geophysical Union, 94(11), pp. 105-106.
Nakagawa, S. and Schielzeth, H. (2013) A general and simple method for obtaining R2 from generalized linear mixed‐effects models, Methods in ecology and evolution, 4(2), pp. 133-142.
Pacheco, A., Camarero, J. J., Pompa-García, M., Battipaglia, G., Voltas, J. and Carrer, M. (2020) Growth, wood anatomy and stable isotopes show species-specific couplings in three Mexican conifers inhabiting drought-prone areas, Science of The Total Environment, 698, pp. 134055.
Peltier, D. M. and Ogle, K. (2019) Legacies of more frequent drought in ponderosa pine across the western United States, Global change biology, 25(11), pp. 3803-3816. 68
Potter, K. M., Hipkins, V. D., Mahalovich, M. F. and Means, R. E. (2015) Nuclear genetic variation across the range of ponderosa pine (Pinus ponderosa): Phylogeographic, taxonomic and conservation implications, Tree Genetics & Genomes, 11(3), pp. 38.
Pumijumnong, N., Muangsong, C., Buajan, S., Sano, M. and Nakatsuka, T. (2020) Climate variability over the past 100 years in Myanmar derived from tree-ring stable oxygen isotope variations in Teak, Theoretical and Applied Climatology, 139(3-4), pp. 1401-1414.
Remke, M. J., Hoang, T., Kolb, T., Gehring, C., Johnson, N. C. and Bowker, M. A. (2020) Familiar soil conditions help Pinus ponderosa seedlings cope with warming and drying climate, Restoration Ecology.
Rinne, K. T., Boettger, T., Loader, N. J., Robertson, I., Switsur, V. R. and Waterhouse, J. S. (2005) On the purification of α-cellulose from resinous wood for stable isotope (H, C and O) analysis, Chemical Geology, 222(1-2), pp. 75-82.
Roden, J. S. and Ehleringer, J. R. (1999) Observations of hydrogen and oxygen isotopes in leaf water confirm the Craig-Gordon model under wide-ranging environmental conditions, Plant physiology, 120(4), pp. 1165-1174.
Roden, J. S., Lin, G. and Ehleringer, J. R. (2000) A mechanistic model for interpretation of hydrogen and oxygen isotope ratios in tree-ring cellulose, Geochimica et Cosmochimica Acta, 64(1), pp. 21-35.
Rossi, S., Anfodillo, T. and Menardi, R. (2006) Trephor: a new tool for sampling microcores from tree stems, Iawa Journal, 27(1), pp. 89-97.
Rozanski, K., Araguás‐Araguás, L. and Gonfiantini, R. (1993) Isotopic patterns in modern global precipitation, Climate change in continental isotopic records, 78, pp. 1-36.
Sano, M., Dimri, A., Ramesh, R., Xu, C., Li, Z. and Nakatsuka, T. (2017) Moisture source signals preserved in a 242-year tree-ring δ18O chronology in the western Himalaya, Global and Planetary Change, 157, pp. 73-82.
Shinneman, D. J., Means, R. E., Potter, K. M. and Hipkins, V. D. (2016) Exploring climate niches of ponderosa pine (Pinus ponderosa Douglas ex Lawson) haplotypes in 69 the western United States: Implications for evolutionary history and conservation, PloS one, 11(3).
Sierks, M. D., Kalansky, J., Cannon, F. and Ralph, F. (2020) Characteristics, Origins, and Impacts of Summertime Extreme Precipitation in the Lake Mead Watershed, Journal of Climate, (2019).
Song, X., Farquhar, G. D., Gessler, A. and Barbour, M. M. (2014) Turnover time of the non‐structural carbohydrate pool influences δ18 O of leaf cellulose, Plant, cell & environment, 37(11), pp. 2500-2507.
Stahle, D. W., Cleaveland, M. K., Grissino-Mayer, H. D., Griffin, R. D., Fye, F. K., Therrell, M. D., Burnette, D. J., Meko, D. M. and Villanueva Diaz, J. (2009) Cool-and warm-season precipitation reconstructions over western New Mexico, Journal of Climate, 22(13), pp. 3729-3750.
Sternberg, L., Pinzon, M. C., Anderson, W. T. and Jahren, A. H. (2006) Variation in oxygen isotope fractionation during cellulose synthesis: intramolecular and biosynthetic effects, Plant, Cell & Environment, 29(10), pp. 1881-1889.
Sternberg, L. D. S., Deniro, M. J. and Savidge, R. A. (1986) Oxygen isotope exchange between metabolites and water during biochemical reactions leading to cellulose synthesis, Plant Physiology, 82(2), pp. 423-427.
Szejner, P., Clute, T., Anderson, E., Evans, M. N. and Hu, J. (2020) Reduction in lumen area is associated with the δ18O exchange between sugars and source water during cellulose synthesis, New Phytologist, 226(6), pp. 1583-1593.
Szejner, P., Wright, W. E., Babst, F., Belmecheri, S., Trouet, V., Leavitt, S. W., Ehleringer, J. R. and Monson, R. K. (2016) Latitudinal gradients in tree ring stable carbon and oxygen isotopes reveal differential climate influences of the North American Monsoon System, Journal of Geophysical Research: Biogeosciences, 121(7), pp. 1978- 1991.
Szejner, P., Wright, W. E., Belmecheri, S., Meko, D., Leavitt, S. W., Ehleringer, J. R. and Monson, R. K. (2018) Disentangling seasonal and interannual legacies from inferred patterns of forest water and carbon cycling using tree‐ring stable isotopes, Global change biology, 24(11), pp. 5332-5347. 70
Treydte, K., Boda, S., Graf Pannatier, E., Fonti, P., Frank, D., Ullrich, B., Saurer, M., Siegwolf, R., Battipaglia, G. and Werner, W. (2014) Seasonal transfer of oxygen isotopes from precipitation and soil to the tree ring: source water vs. needle water enrichment, New Phytologist, 202(3), pp. 772-783.
Truettner, C., Anderegg, W. R., Biondi, F., Koch, G. W., Ogle, K., Schwalm, C., Litvak, M. E., Shaw, J. D. and Ziaco, E. (2018) Conifer radial growth response to recent seasonal warming and drought from the southwestern USA, Forest ecology and management, 418, pp. 55-62.
Truettner, C., Dettinger, M. D., Ziaco, E. and Biondi, F. (2019) Seasonal Analysis of the 2011–2017 North American Monsoon near its Northwest Boundary, Atmosphere, 10(7), pp. 420.
Vankat, John. Vegetation dynamics on the mountains and plateaus of the American Southwest. Vol. 8. Springer Science & Business Media, 2013.
Vera, C., Higgins, W., Amador, J., Ambrizzi, T., Garreaud, R., Gochis, D., Gutzler, D., Lettenmaier, D., Marengo, J. and Mechoso, C. (2006) Toward a unified view of the American monsoon systems, Journal of climate, 19(20), pp. 4977-5000.
Verheyden, A., Helle, G., Schleser, G., Dehairs, F., Beeckman, H. and Koedam, N. (2004) Annual cyclicity in high‐resolution stable carbon and oxygen isotope ratios in the wood of the mangrove tree Rhizophora mucronata, Plant, Cell & Environment, 27(12), pp. 1525-1536.
Williams, A. P., Allen, C. D., Macalady, A. K., Griffin, D., Woodhouse, C. A., Meko, D. M., Swetnam, T. W., Rauscher, S. A., Seager, R. and Grissino-Mayer, H. D. (2013) Temperature as a potent driver of regional forest drought stress and tree mortality, Nature climate change, 3(3), pp. 292-297.
Ziaco, E. (2020) A phenology-based approach to the analysis of conifers intra-annual xylem anatomy in water-limited environments, Dendrochronologia, pp. 125662.
Ziaco, E., Truettner, C., Biondi, F. and Bullock, S. (2018) Moisture‐driven xylogenesis in Pinus ponderosa from a Mojave Desert mountain reveals high phenological plasticity, Plant, cell & environment, 41(4), pp. 823-836. 71
Figure 1. Location of the study site (2320 m) in the Sheep Range of southern Nevada,
U.S.A. 35 km north-northwest of Las Vegas. The study site is included in the Nevada
Climate-Ecohydrological Assessment Network (NevCAN) located in the Desert National
Wildlife Refuge.
72
Figure 2. Microscopic image of the 2015 and 2016 tree rings from a sampled ponderosa pine at the NevCAN Montane site highlighting the variation in wood anatomy including lumen diameter (LD) and cell-wall thickness (CWT). Tree rings were sliced into quartiles
(Q12015 - Q42016) and classified into subseasons depending on the tree-ring phenodate.
Subseasons include: 2015 Monsoon Season (MS2015), 2015 Post-Monsoon Season
(PM2015), 2016 Early Warm Season (EWS2016), 2016 Monsoon Season (MS2016), and
2016 Post-Monsoon Season (PM2016). A false ring is present in MS2015 and separates the subseasons.
73
Figure 3. The δ18O VSMOW (‰) and δ2H VSMOW (‰) measurements for precipitation, xylem water (large and small trees), and soil water (10 cm depth and 20 cm depth) collected during the 2015 and 2016 growing season at the NevCAN Montane Site in the
Sheep Range of southern Nevada. Precipitation was classified into five subseasons (A) and were used to plot the local meteoric water line (δ2H = 8.27*δ18O + 9.7) compared to the global meteoric water line from Rozanski et al. (1993) (δ2H = 8.13*δ18O + 10.8). The
δ18O VSMOW (‰) of precipitation, stem water, and soil water δ18O VSMOW (‰) are plotted with daily precipitation for the study time period (B). 74
Figure 4. Subseasonal measurements of ponderosa pine radial growth during the 2015 and 2016 growing season at the NevCAN Sheep Range Montane Site. Subseasons were classified as the 2015 monsoon season (MS2015), 2015 post-monsoon season
(PM2015), 2016 early warm season (EWS2016), 2016 monsoon season (MS2016), and 2016
18 18 post-monsoon season (PM2016). Precipitation (δ OPPT) and xylem water (δ OXW) were collected at the field site for large (A) and small (B) trees; the δ18O in α-cellulose (δ18O-
CELL ) was measured from high resolution microtome slices of the 2015 and 2016 tree for
13 18 large (C) and small (D) trees; and δ C in α-cellulose (δ OCELL ) for large (E) and small
(F) trees. Subseasons with the same letter are not significantly different (p < 0.05) using the Tukey Honest Significant Differences (HSD) test for large or small trees. Asterix (*)
18 denotes significantly different (p < 0.05) subseasonal δ OCELL between large vs. small trees using Wilcox rank sum test. 75
Figure 5. Proportion of oxygen atoms from leaf water that have exchanged with xylem water at the site of cellulose synthesis (Pex) for large (a) and small (b) trees. Subseasons that have the same letter are not significantly different (p < 0.05) using the Tukey Honest
Significant Differences (HSD) test. The HSD test was restricted to differences in subseason for large or small trees. Asterix (*) denotes significantly different (p < 0.05)
18 subseasonal δ OCELL between large vs. small trees using Wilcox rank sum test.
76
Table 1. Diameter at Breast Height (DBH), height, and final date of lignification for tree- ring growth quartiles for the 2015 and 2016 tree rings from the twelve ponderosa pines sampled
DBH Height Tree (cm) (m) Q12015 Q22015 Q32015 Q42015 Q12016 Q22016 Q32016 Q42016 T1 47.5 11.0 17-Jul 29-Jul 12-Aug 9-Sep 10-Jun 28-Jun 19-Jul 7-Sep T2 34.5 7.0 5-Aug 11-Aug 19-Aug 3-Sep 22-Jun 27-Jun 2-Jul 14-Jul T3 49.0 10.0 10-Jul 28-Jul 18-Aug 4-Oct 10-Jun 20-Jun 2-Jul 31-Jul T4 49.0 10.0 23-Jul 5-Aug 20-Aug 27-Sep 13-Jun 20-Jun 29-Jun 19-Jul T5 56.5 10.9 30-Jun 17-Jul 6-Aug 15-Sep 10-Jun 20-Jun 3-Jul 30-Jul T6 76.0 13.4 29-Jul 9-Aug 21-Aug 20-Sep 14-Jun 26-Jun 9-Jul 9-Aug T7 53.5 10.5 16-Aug 21-Aug 27-Aug 8-Sep 7-Jun 25-Jun 16-Jul 5-Sep T8 66.0 12.1 13-Aug 17-Aug 22-Aug 31-Aug 14-Jun 23-Jun 5-Jul 3-Aug T9 39.5 9.0 14-Aug 21-Aug 30-Aug 19-Sep 13-Jun 19-Jun 26-Jun 12-Jul T10 33.0 8.2 24-Aug 28-Aug 3-Sep 15-Sep 16-Jun 9-Jul 6-Aug 16-Oct T11* 32.0 8.1 - - - - 11-Jun 22-Jun 5-Jul 2-Aug T12 38.5 8.8 20-Aug 24-Aug 29-Aug 6-Sep 2-Jun 24-Jun 22-Jul 11-Oct 11- Large 55.3 11.0 21-Jul 3-Aug 18-Aug 18-Sep 11-Jun 23-Jun 8-Jul Aug 23- Small 35.5 8.2 16-Aug 21-Aug 28-Aug 10-Sep 13-Jun 26-Jun 12-Jul Aug
*No tree-ring growth for T19 in 2015 **Bolded = False Ring or Intra-Annual Density Fluctuation (IADF) present 77
Bimodal Precipitation Variation Measured in Intra-Annual Tree-Ring Isotope
Chronologies from Southern Nevada, USA
1 2 3,4 5 1 4 Truettner, C. , A. Z. Csank , M.D. Dettinger , S.R. Poulson , E. Ziaco , M.D. Sierks ,
C.M. Albano6, F. Biondi1
1Department of Natural Resources and Environmental Science, University of Nevada, Reno, NV
2Department of Geography, University of Nevada, Reno, NV
3U.S. Geological Survey (Retired), Carson City, NV, USA
4Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA, USA
5Department of Geological Sciences and Engineering, University of Nevada, Reno, NV
6Division of Hydrological Science, Desert Research Institute, Reno, NV
Keywords: Dendroclimatology, climate change, NevCAN, North American monsoon, precipitation extremes, floods, Lake Mead reservoir
Abstract The North American Monsoon System (NAMS) is a climatic phenomenon during the late warm-season (June-September) that delivers vital summertime precipitation to the deserts of the southwestern USA. The onset and strength of the NAMS is highly variable in the paleoclimatological record, especially on the boundaries of the NAMS in the
Mojave and Great Basin deserts. A 377-year old earlywood and latewood isotope chronology was constructed in southern Nevada near the northwest boundary of the
NAMS. Past studies have used annual or latewood δ13C isotopic signatures to reconstruct variability in late warm-season precipitation in the NAMS-core region. We found that an index calculated as the difference between the δ13C of α-cellulose in the 78
13 13 latewood and the δ C value of earlywood (δ CDIFF = latewood – earlywood) was highly influenced by the fraction (fSUMMER) of summer precipitation (June-September) to previous cool-season precipitation (October – April). We compared the δ13C value of
13 earlywood, latewood, and δ CDIFF to monthly combinations of late warm-season precipitation, cool-season precipitation, and fSUMMER. The strongest relationship was
13 13 between δ CDIFF and fSUMMER. We recommend investigating the simple index of δ CDIFF along with earlywood and latewood δ13C values in isotope chronologies near the boundaries of monsoonal regions to investigate variations in bimodal precipitation regimes.
Introduction Dendroclimatology is the study of variations in tree ring series as a function of climate. It has been studied for over a century, beginning with foundational research in
Arizona, USA (Douglass, 1914). The arid southwestern region of the USA (including
Arizona, and southern Nevada) is influenced by a bimodal precipitation regime with the majority of precipitation in the region coming during the cool-season (previous October-
April) but with a secondary pulse of precipitation arriving during the late warm-season
(mid-June-September) (Douglas et al., 1993; Adams and Comrie, 1997; Truettner et al.,
2019). Cool-season precipitation is associated with mid-latitude storms originating off the eastern Pacific coast that move westward inland to the deserts of North America.
Summer precipitation is often associated with the North American monsoon system
(NAMS) and precipitation extremes like remnant storms from hurricanes (Vera et al.,
2006; Sierks et al., 2020). Precipitation events associated with the NAMS is driven by shifts in wind direction and pulses of water vapor from subtropical sources that include the Gulf of California, Gulf of Mexico, vegetation transpiration, and the Pacific Ocean (Hu and Dominguez, 2015; Jana et al., 2018). 79
Interannual variability in NAMS moisture is of critical interest to water and agricultural resource managers in the region. NAMS moisture becomes particularly important to maintain water resources and ecosystems during years with below normal winter precipitation (Ray et al. 2007). For this reason, the NAMS has been the focus of numerous studies using instrumental observations (e.g. Seager et al., 2009) and paleoclimate data, including tree rings, to reconstruct NAMS variability (e.g. Barron et al.
2012; Griffin et al. 2013). Prior dendroclimatological studies have focused on reconstructing the NAMS in the NAMS core region located in Arizona and New Mexico,
USA and northwestern Mexico (Stahle et al, 2009; Griffin et al., 2013). In order to understand how the NAMS may change throughout time, it is important to understand how interannual monsoon moisture varies at the edges of the NAMS core region.
Arguably, the boundary regions are where variability will be most pronounced. Here we investigate interannual variability in the bimodal precipitation regime near the northwest boundary of the NAMS core region in the arid Mojave desert of southern Nevada, USA.
Although the southwestern USA has a long history of dendroclimatic studies, the majority of dendroclimatic reconstructions produced from the region reflect cool-season or annual-scale moisture variability. Researchers have used subannual chronologies produced using the spring-forming earlywood and summer-forming latewood portions of annual tree-rings to reconstruct seasonal variations of the bimodal precipitation regime in the NAMS-core region (e.g. Griffin et al. 2013). The earlywood portion appears lighter in color because of wider lumen area in xylem cells and is associated with cool-season precipitation in western North America. The latewood portion appears darker in color because of narrower lumen areas and thicker cell walls. Latewood is associated with late warm-season precipitation in the NAMS-core region as convective monsoonal 80
thunderstorms become more frequent towards the end of June and early July (Stahle et al, 2009; Griffin et al., 2013; Higgins et al., 1999).
13 13 More recently, the δ C composition of tree-ring cellulose (δ CCELL) has been measured to investigate the degree of water stress in trees which is related to differences in cool-season and late warm-season moisture delivery. This could explain
13 the differences between δ CCELL measured in the earlywood and latewood (Belmecheri et al., 2018; Szejner et al., 2020). In addition, dendroclimatologists have tested hypotheses as to whether interannual variability in monsoon moisture delivery in the
13 NAMS core region is reflected in isotopic signatures of latewood δ CCELL (Leavitt et al.,
2011;; Szejner et al., 2018). However, the NAMS-associated precipitation signal (i.e.,
July-September) in their isotope chronologies is weaker the farther north from the
NAMS-core region the isotope chronology was sampled (Szejner et al., 2016). The present study builds on the results of these prior studies by producing the longest record
13 13 13 of δ CCELL measured in both earlywood (δ CEW) and latewood (δ CLW) from the southwestern USA. This study presents a new method of investigating the bimodal
13 13 precipitation hydroclimate of the southwestern USA in relation to the δ CEW and δ CLW through measuring the fraction (fSUMMER) of late warm-season precipitation that falls from
June-September compared to precipitation that fell from the previous cool-season
(October-April). The fSUMMER should more accurately reflect the balance between late warm-season precipitation and cool-season precipitation on the boundaries of the
NAMS-core region where cool-season precipitation may be greater than late warm- season precipitation. We do this by investigating a new index derived as the difference
13 13 13 13 between the δ CLW and δ CEW (δ CDIFF). Theoretically, a more positive δ CCELL value
13 reflects more water stress in ponderosa pine compared to a negative δ CCELL value 81
13 (Farquhar et al., 1982 and thereon). The δ CEW is a function of moisture stress in the early part of the growing season primarily influenced by the amount of cool-season
13 precipitation. The δ CLW is a function of water-stress during the late warm-season when monsoonal precipitation may reach southern Nevada. Therefore, if there is a dry winter
13 followed by a wet summer, the δ CDIFF would be negative because the trees would become less water-stressed throughout the growing season. Conversely, if there was a
13 wet winter followed by a dry summer, then δ CDIFF would be more positive because trees would become progressively more water-stressed during the growing season. Our
13 main question was to test if the δ CDIFF is reflective of variation in bimodal precipitation of southern Nevada?
Methods
Tree-ring isotope chronologies
13 The width and δ CCELL of earlywood and latewood portions of tree rings were extracted and measured in tree-ring series from an old-growth ponderosa pine (Pinus ponderosa Lawson & C. Lawson) stand located in the Sheep Range of southern
Nevada, USA, from October 2014 until November 2017. Two tree cores from each of 30 ponderosa pine individuals of variable ages (70-462 years old) were collected from an old-growth ponderosa pine stand at the Nevada Climate-ecohydrological Assessment
Network (NevCAN) Sheep Range Montane site. NevCAN is a valley-to-mountain top elevation transect with meteorological stations in each vegetation type from the Mojave desert to subalpine forests (Mensing et al., 2013). The study site (2730 m a.s.l) is located on a desert “sky-island” where mid-to-upper-elevations consist of a conifer dominated ecosystem, with dominant ponderosa pine (Pinus ponderosa) and minor pinon pine (Pinus monophylla). The study site is in the Desert National Wildlife Refuge 82
(DNWR) located ~35 km north-northwest of Las Vegas, Nevada, U.S.A. near the northwest boundary of the NAMS (Figure 1).
Tree core samples were sanded, mounted, and cross-dated (Stokes, 1996) with assistance from COFECHA (Grissino-Mayer, 2001) and dplR (Bunn, 2008). The earlywood and latewood of each tree ring was measured using a Velmex system with
0.001 mm precision. Tree-ring chronologies of the width of earlywood and latewood were successfully cross-dated and measured for 60 tree cores from 30 ponderosa pine individuals. The earlywood and latewood tree-ring width chronologies were detrended using the auto-regressive model in dendrochronology program library in R (Bunn, 2008).
The earlywood (0.98) and latewood (0.96) estimated population signal (EPS) statistics were higher than the commonly used standard (>0.85) from Wigley et al. (1984). The signal-to-noise ratio was high for the earlywood (46.3) and latewood (23.6) ring width
13 chronologies. The coefficient of variation for inter-tree variability was low for δ CCELL (-
3.2 ± 1.1%). Fifteen trees were selected for isotopic analysis with a minimum overlap of five trees for each earlywood and latewood subring from 1640-2017 AD. The earlywood and latewood for each annual tree ring were drilled using a Dremel rotary drill with a dental drill bit fitted under a microscope. We did not use the first 25 annual tree rings
13 because of the “juvenile effect” which is an enrichment of δ CCELL likely from an increase in the δ13C in ambient air around the tree possibly from the byproduct of cellular respiration from soil microbes (Loader et al., 2007; Duffy et al., 2017). Two separate isotope chronologies were measured due to narrow tree rings in the last 150 years of older trees because older trees form narrower tree rings as they grow older and larger.
Therefore, pre-Industrial Revolution (1640-1862 CE; pre-IR) and post-Industrial
Revolution (1830-2017 CE; post-IR) isotope chronologies were measured from the earlywood and latewood portions of an annual tree ring. Sample sizes were robust for 83
13 the δ CCELL of earlywood (n = 827) and latewood (n = 787). The isotope chronologies overlapped for a 32-year period from 1830-1862. Samples were pooled together with the exception of approximately every 10th ring where each individual tree was sampled separately to allow for an assessment of inter-tree variability (Leavitt and Long, 1989).
Samples were processed to α-cellulose using a modified version of the Leavitt and Danzer (1993) cellulose processing method. We followed the suggestion of Rinne et al. (2005) and did not extract resins prior to delignification in an acidified sodium chlorite solution. Following delignification, α-cellulose was purified from holocellulose by placing samples in a 17% NaOH solution at 80˚C for 45 minutes followed by 10% HCl solution
(Rinne et al., 2005; Anchukaitis et al., 2008). The δ13C values were measured using a
Eurovector elemental analyzer interfaced to a Micromass IsoPrime stable isotope ratio
13 mass spectrometer. The δ CCELL analyses were performed using a combustion method, were calibrated using an internal laboratory standard of acetanilide (previously calibrated vs. IAEA reference standards) and standardized to Vienna Pee Dee Belemnite (VPDB).
13 The resultant δ CCELL chronologies were corrected for both the post-industrial
13 depletion in the δ C value of atmospheric CO2 as a result of fossil fuel emissions (i.e., the Suess effect) and changes in the ca/ci ratio as a result of increasing atmospheric
13 CO2 levels. This was done by first correcting the δ CCELL chronologies for the Suess effect using the values for atmospheric δ13C from McCarroll and Loader (2004) with values from 2004-2016 derived from data collected from the Mauna Loa observatory
(Keeling et al., 1976; https://www.esrl.noaa.gov/gmd/ccgg/trends/data.html) and subsequently applying the pre-industrial (PIN) correction method to correct for the influence of changing atmospheric CO2 on ci/ca ratios (McCarroll et al., 2009).
Seasonal Climate Data 84
Total precipitation from June-September (JJAS), June-August (JJA), and July-
September (JAS) were derived from the Parameter Regression for Independent Slopes
Model (PRISM; Daly et al., 2008) for the instrumental period (1948-2017). In addition, total precipitation during the cool-season (previous October-April) and fraction of JJAS to cool-season precipitation (fSUMMER) were derived from the PRISM dataset.
Statistical Analysis
We initially tested for correlations in the overlap period (1830-1862) between the pre-Industrial Revolution (pre-IR) and post-Industrial Revolution (post-IR) isotope
13 13 13 chronologies. When then compared the δ CEW and δ CLW to the annual δ C isotope chronology from Bale et al. (2011) in the White Mountains of southeastern, California,
USA. Moving correlations of the seasonal precipitation variables were tested for
13 significant relationships (p < 0.05) to the δ CCELL in earlywood, latewood, and difference
13 between latewood and earlywood (δ CDIFF = latewood – earlywood) using treeclim in R
(Zang and Biondi, 2015). We used a 20-year window offset by one year to test the temporal stability of the seasonal precipitation variables to the isotope chronologies, similar to Csank et al., (2016). Linear models were calculated for the instrumental period
13 13 (1948-2017 CE) to test the relationship of δ CCELL in earlywood, latewood, and δ CDIFF to the seasonal precipitation variables.
Results
Tree-ring isotope chronologies
13 13 The pre-IR δ CEW and δ CLW isotope chronologies were ~ 1 ‰ more enriched than the post-IR isotope chronologies (Figure 2, Appendix G). Older trees are taller and have larger canopy sizes, which would decrease hydraulic connectivity and enrich the 85
13 δ CCELL in tree rings as the tree grows older (Klesse et al., 2018). Also, the post-IR isotope chronologies had a general enrichment from the beginning of the isotope chronologies from 1830-1845 CE. This may be due to the juvenile effect, even though we attempted to account for it. Besides these possible explanations, we are not
13 13 completely sure why the pre-IR δ CEW and δ CLW isotope chronologies were more enriched. The Suess effect was detected in the post-IR isotope chronology (Figure 2).
13 The PIN correction successfully adjusted the depletion in δ CCELL. Although, a depletion
13 of δ CCELL was present from ~1860-1930 CE. This depletion might be because trees were younger, or it could be a time period when the trees were less water-stressed. The
13 correlations for the overlap period (1830-1862) were significant for the δ CCELL in
13 earlywood (0.61) and latewood (0.51). Therefore, we averaged the δ CCELL of earlywood and latewood values during the overlap period for comparison to the isotope chronology from Bale et al. (2011).
13 The correlation for the δ CCELL of latewood (0.46) in the Sheep Range chronology was higher than that of earlywood (0.38) when compared to the annual resolved isotope chronology from Bale et al. (2011). Therefore, a summer precipitation
13 signal from the δ CLW isotope chronology could be associated with the summer precipitation signal from the Bale et al. (2011). Similarities and differences in the
13 13 variations of the δ CEW and δ CLW isotope chronologies where present for different time periods for the three isotope chronologies (Figure 3a). A similar pattern was measured
13 13 between the δ CEW, δ CLW, and Bale et al. (2011) isotope chronologies form ~1720 to
13 13 1790 CE. A depletion was measured in the δ CEW and δ CLW isotope chronologies after
~1800 CE (Figure 3a). Again, a possible explanation for this depletion may be from isotopic values from younger trees. Inter-annual variation does appear to increase after
13 13 ~1890 in the δ CEW and δ CLW isotope chronologies in comparison to Bale et al. (2011). 86
A general enrichment occurred from 1920-2017 CE, although with great inter-annual variation.
13 The intra-annual signal that δ CDIFF measured may be a simple approach to help
13 13 avoid these possible age-related trends in the δ CEW and δ CLW isotope chronologies
13 (Figure 3b). An overall enrichment was measured in δ CDIFF from 1830 CE, albeit strong inter-annual variation. This could be related to smaller trees becoming progressively more enriched as they grow older and taller.
13 st nd Autocorrelation was low for both δ CEW (1 -order = -0.002, 2 -order = -0.045)
13 st nd st and δ CDIFF (1 -order = 0.004, 2 -order = 0.085). Autocorrelation was higher for 1 -
13 nd order autocorrelation in the δ CLW (0.141) isotope chronology but was low again for 2 -
13 order autocorrelation (-0.021). Autocorrelation in δ CCELL isotope chronologies is associated with photosynthates (i.e., non-structural carbohydrates) formed in previous years that carry-over to the next year of growth (Song et al., 2014; Szejner et al., 2020).
This could explain the low autocorrelation measured in our isotope chronologies.
Instrumental period (1948-2017 AD)
Results from the moving correlation analysis highlight a significant negative
13 correlation with cool-season precipitation and the δ CEW starting ~1975 CE (Figure 4a).
A similar, but not as consistent, negative significant relationship was measured between
13 cool-season precipitation and the δ CLW. A significant relationship between JJAS, JAS,
13 and JJA was found in the δ CLW between ~1965-2003 CE. The strongest correlations
13 were for JJA and δ CLW during this time period. JJA was the strongest correlation found in the Bale et al. (2011) isotope chronologies providing more evidence that a similar summer precipitation signal may be measured between the two isotope chronologies.
13 Most apparent was the strong negative moving correlation relationship between δ CDIFF 87
and JJAS, JAS, and JJA from ~1965 to 2017 CE (Figure 3c). A weaker positive
13 significant relationship was present between cool-season precipitation and δ CDIFF.
13 Therefore, δ CDIFF has a temporally stable significant relationship with late warm-season
13 and cool-season precipitation where a negative δ CDIFF is related to a wet late warm-
13 season, whereas a positive δ CDIFF is related to a wet cool-season.
13 2 The negative relationship between fSUMMER and δ CDIFF (adjusted-R = 0.42, p <
13 13 13 0.0001) is the most significant linear relationship among δ CEW, δ CLW, and δ CDIFF and the seasonal precipitation variables derived from the PRISM dataset (Figure 5o,
13 Appendix H). The negative relationship between cool-season precipitation and δ CEW was the second strongest relationship (Figure 5j). JJAS, JJA, and JAS had significant
13 13 but weaker relationships to δ CLW (Figure 5b,e,h) when compared to δ CDIFF (Figure
5c,f,i). Therefore, the cool-season precipitation signal is likely the dominant driver of the
13 δ CDIFF and fSUMMER relationship, yet the summer precipitation signal likely improves the
13 δ CDIFF and fSUMMER relationship.
13 The relationship between δ CDIFF and fSUMMER from the PRISM dataset was significantly correlated (-0.65, p < 0.0001) during the instrumental period (Figure 6).
13 Years when fSUMMER was high, like 1984 CE, corresponded to a negative δ CDIFF. Years
13 when fSUMMER was low, like in the late 1970s and early 1990s, δ CDIFF was high.
Therefore, there is evidence that trees become less water-stressed throughout the growing season when there was a greater amount of late warm-season precipitation than cool-season precipitation, while trees became more water-stressed through the growing season when more precipitation fell during the cool-season than during the late warm-season.
88
Discussion
Isotope Chronologies
Bias in age related trends in isotope tree-ring chronologies can distort long-term signals including dendroclimatic reconstructions (Klesse et al., 2018). One of the major challenges in constructing long isotope chronologies is the precision and available material in the smaller tree rings towards the outer edge of older trees (McCarroll and
Loader, 2004). Our attempt to combine the pre-IR and post-IR isotope chronologies was challenging. Issues included the presence of the juvenile effect for a longer time period than we expected in the post-IR isotope chronologies. The ponderosa pine at our arid
13 study site may have slow growth rates and would therefore use CO2 depleted in δ C in the ambient air closer to the ground for a longer period of time. A general enrichment of
13 δ CCELL was also present in pre-IR and post-IR isotope chronologies and is likely due to an increase in canopy size and height of a tree (Klesse et al., 2018). However, since the
13 δ CDIFF measures the intra-annual variation of an increase or decrease in water-stress throughout the growing season, some of these age-related trends can be accounted for
13 13 by simply subtracting δ CEW from δ CLW. Further comparisons to other intra-annual
13 isotope chronologies could further test how δ CDIFF accounts for age-related trends.
13 13 The δ CEW from δ CLW isotope chronologies had some similarities to the Bale et al. (2011) annual isotope chronologies in the White Mountains, CA, USA. For instance, all three isotope chronologies had similar values from ~1720 to 1790 CE. This indicates that the hydroclimate in southern Nevada may have been more similar to the hydroclimate east of the Sierra Nevada during this time period. This relationship departed starting in the early 19th century and could indicate a shift in hydroclimates, or it could be simply that smaller trees are more sensitive to variations in late warm-season 89
precipitation (Truettner et al., in prep). A detection of a shift in hydroclimate in southern
Nevada compared to southeastern California may be present. However, we are not completely confident about this shift in hydroclimate without further investigation into paleoclimate with age-related trends removed.
13 δ CDIFF and fSUMMER
13 We demonstrated that the δ CDIFF was a significant biological indicator of the fraction (fSUMMER) of late warm-season to cool-season precipitation at our study site.
Other studies that investigated intra-annual variations in isotope chronologies have seen a decrease in the variation of bimodal precipitation the farther north from the NAMS-core region a tree-ring isotope chronology was extracted (Szejner et al., 2016; Szejner et al.,
2018). This may be why the different monthly combinations of late warm-season
13 precipitation had a weak relationship to the δ CLW compared to previous studies. Leavitt
13 et al. (2011) found that the δ CCELL of latewood in ponderosa pine was strongly related to July-September (JAS) precipitation in Arizona and New Mexico in the NAMS-core region. Precipitation that fell during JAS at our site had a weak relationship. Bale et al.
13 (2011) used the annual tree-ring value of δ CCELL in the tree rings of bristlecone pine
(Pinus longevea) to reconstruct June-August (JJA) precipitation. We found that JJA was
13 13 not the strongest relationship to earlywood or latewood δ CCELL, but that δ CDIFF best estimated late warm-season precipitation in southern Nevada, USA.
13 The δ CEW had a strong relationship to cool-season precipitation and is likely the
13 primary driver of δ CDIFF. However, the variations between cool-season precipitation and late warm-season precipitation represented by fSUMMER was strongly related to the
13 variation in δ CDIFF. Although, reconstructing paleoclimate from the water-stress of a tree over a growing season has been difficult to comprehend. Combining a dendroclimatic 90
13 reconstruction of fSUMMER using δ CDIFF with other dendroclimatic proxies, like a cool- season precipitation reconstruction from earlywood, could yield a more precise dendroclimatic reconstruction of summer precipitation at the boundary of the NAMS-core region. Further investigations using independent proxies to reconstruct variability in summer precipitation is an exciting avenue for further research.
Conclusion
13 The δ CDIFF in tree-ring isotopes is a simple index that has potential to reconstruct trends in the bimodal precipitation regime of the southwestern USA. Trees that are water-stressed from dry winters that receive a pulse of precipitation during the late warm-season become less water-stressed during the growing season, thus a more
13 negative δ CDIFF. If there is a wet cool-season followed by a dry summer, trees become
13 more water-stressed and have a more positive δ CDIFF. Relating this to the bimodal precipitation of southern Nevada has been difficult to translate and has limited us from reconstructing fSUMMER. However, if large precipitation events from remnant hurricanes or
13 monsoonal thunderstorms can be associated to δ CDIFF, then understanding the past climate of southern Nevada could lead to better risk assessments of the likelihood of detrimental floods in southern Nevada where more than a million people live in the Las
Vegas region. Further dendroclimatological studies investigating NAMS-associated and extreme precipitation events at a regional scale are needed to understand the history, causes, and effects of long-term shifts in subseasonal hydroclimate and flood risks in this region. High-resolution dendrochronology studies show promise in using the isotopic signature of α-cellulose to investigate past subseasonal climate (Belmecheri et al., 2018,
Truettner et al., in prep) and could provide more precise reconstructions of extreme 91
precipitation events similar to tropical cyclone reconstructions in southeastern USA
(Miller et al., 2006) and elsewhere.
Large precipitation events in subtropical, desert regions are projected to be more variable and extreme with increased global temperatures from greenhouse gas emissions. A new generation of subseasonal dendroclimatological studies with novel
13 indices, like δ CDIFF pioneered in this dissertation, is needed to help understand past and current trends of hydroclimate. Subseasonal approaches used to reconstruct changes in hydroclimate set the stage for recognizing past and predicting future geographic shifts in the precipitation regimes, flood risks, and ecological challenges that will face the deserts of North America.
References
Adams, D. K. and Comrie, A. C. (1997) The north American monsoon. Bulletin of the American Meteorological Society, 78(10), pp. 2197-2214.
Anchukaitis, K. J., Evans, M. N., Wheelwright, N. T. and Schrag, D. P. (2008) Stable isotope chronology and climate signal calibration in neotropical montane cloud forest trees, Journal of Geophysical Research: Biogeosciences, 113(G3).
Bale, R. J., Robertson, I., Salzer, M. W., Loader, N. J., Leavitt, S. W., Gagen, M., ... & McCarroll, D. (2011). An annually resolved bristlecone pine carbon isotope chronology for the last millennium. Quaternary Research, 76(1), 22-29.
Barron, J. A., Metcalfe, S. E., & Addison, J. A. (2012). Response of the North American monsoon to regional changes in ocean surface temperature. Paleoceanography, 27(3).
Battipaglia, G., Campelo, F., Vieira, J., Grabner, M., De Micco, V., Nabais, C., Cherubini, P., Carrer, M., Bräuning, A. and Čufar, K. (2016) Structure and function of intra–annual density fluctuations: mind the gaps, Frontiers in plant science, 7, pp. 595.
Belmecheri, S., Wright, W. E., Szejner, P., Morino, K. A. and Monson, R. K. (2018) Carbon and oxygen isotope fractionations in tree rings reveal interactions between 92 cambial phenology and seasonal climate, Plant, cell & environment, 41(12), pp. 2758- 2772.
Breshears, D. D., Cobb, N. S., Rich, P. M., Price, K. P., Allen, C. D., Balice, R. G., ... & Anderson, J. J. (2005). Regional vegetation die-off in response to global-change-type drought. Proceedings of the National Academy of Sciences, 102(42), 15144-15148.
Bunn, A. G. (2008). A dendrochronology program library in R (dplR). Dendrochronologia, 26(2), 115-124.
Cook, B. I., & Seager, R. (2013). The response of the North American Monsoon to increased greenhouse gas forcing. Journal of Geophysical Research: Atmospheres, 118(4), 1690-1699.
Csank, A. Z., Miller, A. E., Sherriff, R. L., Berg, E. E., & Welker, J. M. (2016). Tree‐ring isotopes reveal drought sensitivity in trees killed by spruce beetle outbreaks in south‐ central Alaska. Ecological Applications, 26(7), 2001-2020.
Daly, C., Halbleib, M., Smith, J. I., Gibson, W. P., Doggett, M. K., Taylor, G. H., ... & Pasteris, P. P. (2008). Physiographically sensitive mapping of climatological temperature and precipitation across the conterminous United States. International Journal of Climatology: a Journal of the Royal Meteorological Society, 28(15), 2031-2064.
Demko, J.C.; Geerts, B.; Miao, Q.; Zehnder, J.A. Boundary layer energy transport and cumulus development over a heated mountain: An observational study. Mon. Weather Rev. 2009, 137, 447–468.
Douglass, A. E. (1914). A method of estimating rainfall by the growth of trees. Bulletin of the American Geographical Society, 46(5), 321-335.
Douglas, M. W., Maddox, R. A., Howard, K., & Reyes, S. (1993). The mexican monsoon. Journal of Climate, 6(8), 1665-1677.
Duffy, J. E., McCarroll, D., Barnes, A., Ramsey, C. B., Davies, D., Loader, N. J., ... & Young, G. H. (2017). Short-lived juvenile effects observed in stable carbon and oxygen isotopes of UK oak trees and historic building timbers. Chemical Geology, 472, 1-7. 93
Farquhar, Graham D., Marion H. O'Leary, and Joe A. Berry. "On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves." Functional Plant Biology 9.2 (1982): 121-137.
Gershunov, A., Shulgina, T., Ralph, F. M., Lavers, D. A., & Rutz, J. J. (2017). Assessing the climate‐scale variability of atmospheric rivers affecting western North America. Geophysical Research Letters, 44(15), 7900-7908.
Griffin, D., Woodhouse, C. A., Meko, D. M., Stahle, D. W., Faulstich, H. L., Carrillo, C., Touchan, R., Castro, C. L. and Leavitt, S. W. (2013) North American monsoon precipitation reconstructed from tree‐ring latewood, Geophysical Research Letters, 40(5), pp. 954-958.
Griffin, D., Meko, D. M., Touchan, R., Leavitt, S. W. and Woodhouse, C. A. (2011) Latewood Chronology Development for Summer-Moisture Reconstruction In the US Southwest, Tree-Ring Research, 67(2), pp. 87-101, 15.
Grissino-Mayer, H. D. (2001). Evaluating crossdating accuracy: a manual and tutorial for the computer program COFECHA.
Hatchett, B. J., Boyle, D. P., Putnam, A. E., & Bassett, S. D. (2015). Placing the 2012– 2015 California‐Nevada drought into a paleoclimatic context: insights from Walker Lake, California‐Nevada, USA. Geophysical Research Letters, 42(20), 8632-8640.
Higgins, R. W., Chen, Y., & Douglas, A. V. (1999). Interannual variability of the North American warm season precipitation regime. Journal of Climate, 12(3), 653-680.
Hu, H.; Dominguez, F. Evaluation of oceanic and terrestrial sources of moisture for the North American Monsoon using numerical models and precipitation stable isotopes. J. Hydrometeorol. 2015, 16, 19–35.
Jana, S., Rajagopalan, B., Alexander, M. A., & Ray, A. J. (2018). Understanding the dominant sources and tracks of moisture for summer rainfall in the southwest United States. Journal of Geophysical Research: Atmospheres, 123(10), 4850-4870.
Keeling, C. D., Bacastow, R. B., Bainbridge, A. E., Ekdahl Jr, C. A., Guenther, P. R., Waterman, L. S., & Chin, J. F. (1976). Atmospheric carbon dioxide variations at Mauna Loa observatory, Hawaii. Tellus, 28(6), 538-551. 94
Klesse, S., Weigt, R., Treydte, K., Saurer, M., Schmid, L., Siegwolf, R. T., & Frank, D. C. (2018). Oxygen isotopes in tree rings are less sensitive to changes in tree size and relative canopy position than carbon isotopes. Plant, cell & environment, 41(12), 2899- 2914.
Leavitt, S. W., Woodhouse, C. A., Castro, C. L., Wright, W. E., Meko, D. M., Touchan, R., Griffin, D. and Ciancarelli, B. (2011) The North American monsoon in the US Southwest: Potential for investigation with tree-ring carbon isotopes, Quaternary International, 235(1-2), pp. 101-107.
Leavitt, S. W. and Danzer, S. R. (1993) Method for batch processing small wood samples to holocellulose for stable-carbon isotope analysis, Analytical Chemistry, 65(1), pp. 87-89.
Leavitt, S. W., & Long, A. (1989). Intertree variability of δ 13 C in tree rings. In Stable isotopes in ecological research (pp. 95-104). Springer, New York, NY.
Leavitt, S. W., Woodhouse, C. A., Castro, C. L., Wright, W. E., Meko, D. M., Touchan, R., Griffin, D. and Ciancarelli, B. (2011) The North American monsoon in the US Southwest: Potential for investigation with tree-ring carbon isotopes, Quaternary International, 235(1-2), pp. 101-107.
Loader, N. J., McCarroll, D., Gagen, M., Robertson, I., & Jalkanen, R. (2007). Extracting climatic information from stable isotopes in tree rings. Terrestrial Ecology, 1, 25-48.
Loader, N. J., McCarroll, D., Barker, S., Jalkanen, R., & Grudd, H. (2017). Inter-annual carbon isotope analysis of tree-rings by laser ablation. Chemical Geology, 466, 323-326.
Marvel, K., Cook, B. I., Bonfils, C. J., Durack, P. J., Smerdon, J. E., & Williams, A. P. (2019). Twentieth-century hydroclimate changes consistent with human influence. Nature, 569(7754), 59-65.
McCarroll, D., Gagen, M. H., Loader, N. J., Robertson, I., Anchukaitis, K. J., Los, S., ... & Waterhouse, J. S. (2009). Correction of tree ring stable carbon isotope chronologies for changes in the carbon dioxide content of the atmosphere. Geochimica et Cosmochimica Acta, 73(6), 1539-1547. 95
McCarroll, D., & Loader, N. J. (2004). Stable isotopes in tree rings. Quaternary Science Reviews, 23(7-8), 771-801.
Miller, D. L., Mora, C. I., Grissino-Mayer, H. D., Mock, C. J., Uhle, M. E., & Sharp, Z. (2006). Tree-ring isotope records of tropical cyclone activity. Proceedings of the National Academy of Sciences, 103(39), 14294-14297.
Pascale, S., Boos, W. R., Bordoni, S., Delworth, T. L., Kapnick, S. B., Murakami, H., ... & Zhang, W. (2017). Weakening of the North American monsoon with global warming. Nature Climate Change, 7(11), 806-812.
Pumijumnong, N., Muangsong, C., Buajan, S., Sano, M. and Nakatsuka, T. (2020) Climate variability over the past 100 years in Myanmar derived from tree-ring stable oxygen isotope variations in Teak, Theoretical and Applied Climatology, 139(3-4), pp. 1401-1414.
Ray, A. J., Garfin, G. M., Wilder, M., Vásquez-León, M., Lenart, M., & Comrie, A. C. (2007). Applications of monsoon research: Opportunities to inform decision making and reduce regional vulnerability. Journal of Climate, 20(9), 1608-1627.
Rinne, K. T., Boettger, T., Loader, N. J., Robertson, I., Switsur, V. R. and Waterhouse, J. S. (2005) On the purification of α-cellulose from resinous wood for stable isotope (H, C and O) analysis, Chemical Geology, 222(1-2), pp. 75-82.
Rutz, J. J., Steenburgh, W. J., & Ralph, F. M. (2014). Climatological characteristics of atmospheric rivers and their inland penetration over the western United States. Monthly Weather Review, 142(2), 905-921.
Sano, M., Dimri, A., Ramesh, R., Xu, C., Li, Z. and Nakatsuka, T. (2017) Moisture source signals preserved in a 242-year tree-ring δ18O chronology in the western Himalaya, Global and Planetary Change, 157, pp. 73-82.
Sierks, M. D., Kalansky, J., Cannon, F. and Ralph, F. (2019) Characteristics, Origins, and Impacts of Summertime Extreme Precipitation in the Lake Mead Watershed, Journal of Climate, (2019). 96
Song, X., Farquhar, G. D., Gessler, A. and Barbour, M. M. (2014) Turnover time of the non‐structural carbohydrate pool influences δ18 O of leaf cellulose, Plant, cell & environment, 37(11), pp. 2500-2507.
Stahle, D. W., Cleaveland, M. K., Grissino-Mayer, H. D., Griffin, R. D., Fye, F. K., Therrell, M. D., Burnette, D. J., Meko, D. M. and Villanueva Diaz, J. (2009) Cool-and warm-season precipitation reconstructions over western New Mexico, Journal of Climate, 22(13), pp. 3729-3750.
Szejner, P., Clute, T., Anderson, E., Evans, M. N. and Hu, J. (2020) Reduction in lumen area is associated with the δ18O exchange between sugars and source water during cellulose synthesis, New Phytologist, 226(6), pp. 1583-1593.
Szejner, P., Wright, W. E., Babst, F., Belmecheri, S., Trouet, V., Leavitt, S. W., Ehleringer, J. R. and Monson, R. K. (2016) Latitudinal gradients in tree ring stable carbon and oxygen isotopes reveal differential climate influences of the North American Monsoon System, Journal of Geophysical Research: Biogeosciences, 121(7), pp. 1978- 1991.
Szejner, P., Wright, W. E., Belmecheri, S., Meko, D., Leavitt, S. W., Ehleringer, J. R. and Monson, R. K. (2018) Disentangling seasonal and interannual legacies from inferred patterns of forest water and
Stokes, M. A. (1996). An introduction to tree-ring dating. University of Arizona Press.
Truettner, C., Dettinger, M. D., Ziaco, E. and Biondi, F. (2019) Seasonal Analysis of the 2011–2017 North American Monsoon near its Northwest Boundary, Atmosphere, 10(7), pp. 420.
Vera, C., Higgins, W., Amador, J., Ambrizzi, T., Garreaud, R., Gochis, D., ... & Nogues- Paegle, J. (2006). Toward a unified view of the American monsoon systems. Journal of climate, 19(20), 4977-5000.
Wigley, T. M., Briffa, K. R., & Jones, P. D. (1984). On the average value of correlated time series, with applications in dendroclimatology and hydrometeorology. Journal of climate and Applied Meteorology, 23(2), 201-213. 97
Williams, A. P., Allen, C. D., Macalady, A. K., Griffin, D., Woodhouse, C. A., Meko, D. M., ... & Dean, J. S. (2013). Temperature as a potent driver of regional forest drought stress and tree mortality. Nature climate change, 3(3), 292-297.
Zang, C., & Biondi, F. (2015). treeclim: an R package for the numerical calibration of proxy‐climate relationships. Ecography, 38(4), 431-436.
Ziaco, E., Truettner, C., Biondi, F., & Bullock, S. (2018). Moisture‐driven xylogenesis in Pinus ponderosa from a Mojave Desert mountain reveals high phenological plasticity. Plant, cell & environment, 41(4), 823-836. 98
Figure 1. Location of the Sheep Range Montane site in southern Nevada, USA compared to the fraction (fSUMMER) of summer precipitation (June-September) to non- summer precipitation (previous October-May) on a regional scale. Data from PRISM dataset (1980-2010 normals).
99
13 Figure 2. The δ CCELL earlywood and latewood isotope chronologies for the pre-
Industrial Revolution (1640 – 1862 CE) and post-Industrial revolution (1820 – 2017 CE).
The pre-industrial corrections method (PIN) was applied to the post-Industrial Revolution isotope chronologies. 100
Figure 3. Tree-ring isotope chronologies from the Sheep Range, NV, USA: (a) the
13 13 δ CCELL for earlywood and latewood compared to the δ CCELL of annual tree rings from
Bale et al., (2011), (b) Isotopic reconstruction of the fraction of summer precipitation to
13 non-summer precipitation (fSUMMER*) using δ CDIFF as a paleoclimate proxy, (c) dendroclimatic reconstruction of non-summer precipitation (*non-summer) using the earlywood ring width index as a paleoclimate proxy. The summer precipitation reconstruction (*summer) was calculated by multiplying the fSUMMER* by non-summer*.
Five-year moving averages were used to show variations in the precipitation reconstructions.
101
Figure 4. Results from treeclim that shows the moving correlation coefficient for June-
September (JJAS), July-September (JAS), June-August (JJA), and cool-season
13 13 13 precipitation in relation to a) δ CLW, b) δ CLW, and c) δ CDIFF for the instrumental period
(1948-2017 CE).
102
13 Figure 5. Comparison of season precipitation variable to the δ CCELL proxies that
13 13 13 included earlywood δ CCELL, latewood δ CCELL, and δ CDIFF compared to (a-c) June-
September precipitation (JJAS), (d-f) June-August (JJA) precipitation, (g-i) July-
September precipitation (JAS), (j-l) non-summer (previous October - May) precipitation,
13 (m-o) and fSUMMER precipitation. The strongest relationship is between δ CDIFF and fSUMMER (o). 103
13 Figure 6. Times series of the proxy of δ CDIFF was significantly correlated with fSUMMER
13 indicating that lower δ CDIFF are associated with more summer precipitation, and vice versa.
104
Appendix A
Vapor pressure deficit (VPD) was calculated as the difference between saturated vapor pressure and actual vapor pressure. Saturated vapor pressure (SVP, hPa) was computed from air temperature (AT, °C) using Equation 6 in (Brutsaert, 1982), as follows:
17.269 퐴푇 푆푉푃 = 6.11푒퐴푇+237.3
Actual vapor pressure (VP, hPa) was computed using a formula derived from (Biondi and
Hartsough, 2010) and reported as Equation 2 in [27], as follows:
푅퐻 × 푆푉푃 × 푃 푉푃 = 푅퐻 × 푆푉푃 + 100(푃 − 푆푉푃) with RH (%) = relative humidity, and P (hPa) = barometric pressure.
Dewpoint temperature (DT, °C) was calculated from water vapor pressure (VP, hPa) using Equation A.4 in Ellis et al. (2004), as follows:
푉푃 푙푛 퐷푇 = 240.97 6.11 푉푃 17.502 − 푙푛 6.11
105
Appendix B
Same as Figure 5, but also showing daily and hourly values. The x- and y-axis limits were kept constant in the three graphs to facilitate comparison. As is typical of meteorological variables, the amount of noise increases at the shorter time scales, but the relationship is still visible.
70.0 Early Warm Season 60.0 Late Warm Season 50.0
40.0
30.0
20.0
10.0 Total Weekly Precipitation (mm) Precipitation Weekly Total 0.0 -40.0 -30.0 -20.0 -10.0 0.0 10.0 20.0 Average Weekly Dewpoint Temperature (°C)
70.0 Early Warm Season 60.0 Late Warm Season 50.0
40.0
30.0
20.0
10.0 Total Daily Precipitation (mm) Precipitation Daily Total 0.0 -40.0 -30.0 -20.0 -10.0 0.0 10.0 20.0 Average Daily Dewpoint Temperature (°C)
106
70.0 Early Warm Season 60.0 Late Warm Season 50.0
40.0
30.0
20.0
10.0 Total Hourly Precipitation (mm) Precipitation Hourly Total 0.0 -40.0 -30.0 -20.0 -10.0 0.0 10.0 20.0 Average Hourly Dewpoint Temperature (°C)
107
Appendix C
Storm Track Analysis for Late-Warm Season Integrated Water Vapor Transport
The following figures display regional scale measurements of integrated water vapor transport (IVT) from NOAA Earth System Research Laboratory’s NCEP/NCAR Reanalysis portal (www.esrl.noaa.gov/psd/data/composites/day/derived.html) for two days prior, one day prior, and day of storm (> 2 mm) during the 2015 and 2016 late warm season. 7/4/2015 – 7/6/2015 (3.8 mm)
7/6/2015 – 7/8/2015 (2.8 mm)
7/19/2015 – 7/21/2015 (46.23 mm) Hurricane Dolores
108
7/31/2015 – 8/2/2015 (57.4 mm)
8/6/2015 – 8/8/2015 (3.05 mm)
9/14/2015 – 9/16/2015 (10.42 mm)
9/21/2015 – 9/23/2015 (4.06 mm)
109
10/4/2015 – 10/6/2015 (26.67 mm)
10/17/2015 – 10/19/2015 (56.90 mm)
7/1/2016 – 7/3/2016 (17.78 mm)
8/2/2016 – 8/4/2016 (10.67 mm)
110
8/22/2016 – 8/24/2016 (12.7 mm)
10/23/2016 – 10/25/2016 (28.70 mm)
10/27/2016 – 10/29/2016 (6.35 mm)
111
Appendix D
13 18 The two following tables are the raw measurements for δ CCELL and δ OCELL for the tree- ring quartiles
13 Table 1: Raw δ CCELL measurements for tree-ring quartiles
Tree Q12015 Q22015 Q32015 Q42015 Q12016 Q22016 Q32016 Q42016 T1 -21.80 -21.10 -21.18 -20.35 -22.41 -22.61 -22.11 -21.64 T2 -21.41 -21.51 -21.51 -21.34 -21.49 -21.44 -21.30 -21.27 T3 -21.71 -21.78 -22.76 -22.24 -22.74 -23.78 -24.19 -22.55 T4 -23.60 -24.57 -24.28 -23.85 -24.64 -24.86 -24.99 -24.87 T5 -23.16 -23.45 -23.13 -23.48 -23.22 -23.37 -22.81 -23.10 T6 - -21.78 -22.49 -22.11 -22.57 -22.43 -23.42 -22.18 T7 -21.21 -22.31 -21.29 -22.38 -22.14 -22.55 -22.30 -21.74 T8 -22.25 -23.26 -22.72 -22.17 -23.24 -22.93 -22.59 -22.64 T17 -20.94 -21.55 -22.67 -23.36 -22.08 -23.02 -21.93 -22.73 T18 -21.55 -21.58 -21.60 -21.52 -23.32 -21.55 -21.72 -21.16 T19* - - - - -23.42 -22.36 -23.84 -24.02 T20 -24.1 -24.42 -25.05 -24.83 -24.62 -24.5 -23.67 -23.53
18 Table 2: Raw δ OCELL measurements for tree-ring quartiles
Tree Q12015 Q22015 Q32015 Q42015 Q12016 Q22016 Q32016 Q42016 T1 31.37 31.93 32.11 33.01 31.68 32.55 30.37 28.47 T2 33.24 33.35 33.34 33.18 31.95 32.65 31.95 31.24 T3 - - 32.24 31.9 - 33.15 33.84 33.06 T4 31.29 33.14 31.39 32.23 32.33 32.67 32.5 30.44 T5 30.79 - 31.46 - - - 32.7 29.36 T6 - 33.44 33.6 32.93 31.22 33.89 - 31.42 T7 32.15 32.91 32.54 32.4 31.73 31.48 30 29.99 T8 31.63 31.17 31.63 32.45 32.81 31.86 33.84 32.12 T17 35.34 35.85 - - 33.29 32.99 34.11 32.25 T18 32.92 31.52 31.48 32.14 32.67 33.15 32.855 32.97 T19* - - - - 32.83 31.44 32.19 31.89 T20 34.56 34.43 35.13 35.07 34.27 33.59 33.32 32.09
112
Appendix E
18 Table 1: Significant differences in subseasonal δ OCELL, iWUE, and Pex for large and small trees
Subseason Comparison Variables W-value P-value 18 Large MS2015 - Large MS2016 δ OCELL 17 < 0.01 18 Large PM2015 - Small PM2015 δ OCELL 38 0.04 18 Large EWS2016 - Small EWS2016 δ OCELL 39.5 0.01 18 Large PM2016 - Small PM2016 δ OCELL 33 0.01
Small MS2015 – Small PM2015 iWUE 49 <0.01
Large MS2015 - Large PM2015 Pex 103 < 0.01
Large MS2015 - Large EWS2016 Pex 83 < 0.01
Large MS2015 - Large MS2016 Pex 87 < 0.01
Large PM2016 – Large PM2015 Pex 58 < 0.01
Large PM2016 – Large EWS2016 Pex 48 < 0.01
Large PM2016 – Large MS2016 Pex 55 0.02
Large EWS2016 – Small EWS2016 Pex 4 0.02
Large PM2016 – Small PM2016 Pex 2 0.01
Large MS2016 – Small MS2016 Pex 2 0.01
113
Appendix F
18 Correlation analysis for subseasonal ecophysiological parameters and δ Ocell in the
2015 and 2016 tree-ring quartiles
18 18 A correlation analysis δ Ocell and ecophysiological parameters that affect δ Ocell was conducted for each subseason in large and small trees. Subhourly atmospheric measurements were averaged over the tree-ring quartile phenodates for the following parameters; total precipitation (Total PPT), relative humidity (RH), and vapor-pressure
18 18 deficit (VPD). The δ O VSMOW from the precipitation samples (δ Oppt) were sum-
18 weighted for each tree-ring quartile phenodate, and the δ Oxw were averaged over the phenodate. The wood anatomy traits of LD and CWT were similarly investigated. Finally, intrinsic water use efficiency (iWUE) was calculated for each phenodate (Supp. Info – iWUE calculations).
18 Significant correlations (p < 0.05) between subseasonal δ Ocell and ecophysiological parameters varied between large and small trees. RH, VPD, and iWUE
18 were correlated with subseasonal δ Ocell in large trees during MS2015, while Total PPT
18 was correlated with subseasonal δ Ocell in small trees. The only parameter during the
18 18 PT2015 correlated with subseasonal δ Ocell was CWT in large tees. Subseasonal δ Ocell
18 during EWS2016 was correlated with iWUE in large trees and δ Oxw in small trees. Total
18 PPT was correlated with subseasonal δ Ocell in large trees during MS2016. Subseasonal
18 δ Ocell during PT2016 was correlated with CWT in large trees and iWUE in small trees.
114
Figure F1: Pearson’s correlation coefficient (r) represented by ellipses between
18 subseasonal δ OCELL and ecophysiological parameters for large (A) and small (B) trees.
Variables include relative humidity (RH), vapor pressure deficit (VPD), total precipitation
(Total PPT), lumen diameter (LD), cell wall thickness (CWT), the δ18O VSMOW of
18 18 18 precipitation (δ OPPT), and the δ O VSMOW of xylem water (δ Oxw). Significant correlations are surrounded by either a black border (p < 0.05), or red border (p < 0.01).
115
Appendix G
13 13 Table 1: Raw measurements of the δ CCELL for earlywood (δ CEW) and latewood 13 (δ CLW) isotope chronologies
13 13 13 13 13 13 13 13 Year δ CEW δ CLW Year δ CEW δ CLW Year δ CEW δ CLW Year δ CEW δ CLW
1640 -20.53 -19.81 1676 -19.85 -19.49 1712 -20.47 -19.61 1748 -19.03 -19.53 1641 -20.62 -20.44 1677 -20.06 -19.03 1713 -18.86 -19.83 1749 -19.83 -19.84 1642 -20.82 -19.73 1678 -18.29 -18.49 1714 -19.73 -19.28 1750 -19.72 -19.10 1643 -19.85 -20.17 1679 -19.32 -19.68 1715 -19.27 -18.97 1751 NA -18.61 1644 -19.84 -19.64 1680 -20.09 -20.02 1716 -19.29 -19.66 1752 NA NA
1645 -20.45 -20.70 1681 -19.53 -19.56 1717 -19.34 -19.76 1753 -18.20 -18.36 1646 -20.96 -20.81 1682 -19.96 -19.42 1718 -20.32 -19.36 1754 -18.28 -19.07 1647 -20.62 -19.46 1683 -20.12 -19.38 1719 -19.69 -19.13 1755 -16.83 NA 1648 -19.50 -20.45 1684 -18.87 -18.97 1720 -20.64 -20.42 1756 -18.42 -18.45 1649 -20.22 -19.14 1685 -19.33 -20.01 1721 -19.06 -18.78 1757 NA NA 1650 -18.89 -20.49 1686 -19.48 -19.76 1722 -17.95 -18.51 1758 -19.36 -19.17 1651 -20.49 -19.35 1687 -20.16 -20.15 1723 -19.80 -19.42 1759 -18.56 -19.56 1652 -19.30 -19.26 1688 -19.70 -19.41 1724 -19.71 -19.56 1760 -19.94 -19.62 1653 -18.33 -18.71 1689 -19.46 -19.47 1725 -20.13 -21.20 1761 -19.19 -19.37 1654 -18.57 -19.80 1690 -19.64 -19.08 1726 -20.56 -19.52 1762 -19.01 -18.89
1655 -19.54 -19.43 1691 -19.06 -20.24 1727 -20.15 -19.01 1763 -18.12 -18.65 1656 -20.10 -20.14 1692 -20.24 -19.25 1728 -19.12 -18.82 1764 -19.22 -19.65 1657 -19.84 -19.35 1693 -19.76 -19.61 1729 -18.63 -18.53 1765 -20.88 -21.02 1658 -19.83 -19.13 1694 -19.99 -19.89 1730 -19.63 -20.02 1766 -20.03 -19.31 1659 -19.90 -19.50 1695 -19.76 -19.28 1731 -19.59 -18.83 1767 -18.72 -19.38 1660 -20.80 -19.71 1696 -19.68 -19.23 1732 -19.87 -19.62 1768 -19.11 -19.49 1661 -20.84 -19.65 1697 -19.47 -18.72 1733 -18.25 -18.83 1769 -19.00 -19.21 1662 -20.09 -19.43 1698 -17.63 -18.62 1734 -19.56 -18.53 1770 -19.32 -19.89 1663 -19.56 -19.84 1699 -19.92 -19.41 1735 -18.64 NA 1771 -19.22 -20.13 1664 -20.00 -18.57 1700 -19.11 -19.35 1736 NA NA 1772 -19.66 -18.73 1665 -18.04 -18.52 1701 -20.14 -19.77 1737 -18.60 -18.76 1773 -18.92 -19.46
1666 -18.48 -18.52 1702 -19.54 -18.99 1738 -18.96 -18.48 1774 -19.06 -19.57 1667 -18.42 -18.48 1703 -19.35 -19.77 1739 -19.15 -19.28 1775 -19.77 -19.60 1668 -19.10 -19.15 1704 -19.04 -19.44 1740 -19.32 -19.36 1776 -20.02 -19.36 1669 -18.25 -19.13 1705 -20.16 -20.10 1741 -19.31 -19.24 1777 -19.02 -18.98 1670 NA NA 1706 -20.22 -19.56 1742 -20.23 -19.50 1778 -18.95 -18.79 1671 -19.04 -19.19 1707 -19.17 -19.57 1743 -19.76 -19.36 1779 -19.29 -18.93 1672 -19.42 -18.99 1708 -19.58 -19.26 1744 -19.41 -19.34 1780 -18.92 -19.10 1673 -19.76 -19.40 1709 -19.47 -19.71 1745 -19.12 -19.78 1781 -18.85 -19.61 1674 -20.16 -19.45 1710 -20.17 -19.48 1746 -20.28 -20.63 1782 NA NA 1675 -19.39 -20.57 1711 -20.07 -19.66 1747 -20.27 -19.28 1783 -19.28 -19.88
116
13 13 13 13 13 13 13 13 Year δ CEW δ CLW Year δ CEW δ CLW Year δ CEW δ CLW Year δ CEW δ CLW
1785 NA NA 1821 -19.45 -19.15 1857 NA NA 1893 -20.41 -21.26
1786 -19.68 -19.62 1822 NA NA 1858 -19.39 -19.49 1894 -19.77 -19.85
1787 -19.04 -19.05 1823 NA NA 1859 -19.54 -19.73 1895 -20.70 -19.80
1788 NA NA 1824 -19.30 -20.13 1860 -20.07 -20.02 1896 -19.64 -20.49
1789 -19.41 -20.03 1825 -19.61 -19.68 1861 -20.39 -20.00 1897 -21.02 -20.19
1790 NA NA 1826 -19.21 -19.75 1862 -20.42 -20.02 1898 -20.06 -19.78
1791 -19.59 -20.35 1827 -19.38 -20.11 1863 -20.38 -20.46 1899 -19.52 -19.48
1792 -20.07 -20.68 1828 -19.81 -19.62 1864 -19.58 -20.06 1900 -20.23 -19.58
1793 -20.56 -20.28 1829 -18.86 -20.34 1865 -19.52 -19.04 1901 -20.42 -19.95
1794 -19.10 -18.80 1830 -20.32 -20.69 1866 -19.21 -19.32 1902 -19.99 -19.80
1795 -19.31 -19.49 1831 -20.30 -20.60 1867 -20.48 -19.57 1903 -20.54 -20.34
1796 -19.58 -19.82 1832 -20.62 -20.12 1868 -21.40 -21.02 1904 -21.03 -21.78
1797 -19.52 -20.07 1833 -20.81 -20.89 1869 -21.25 -20.93 1905 -21.36 -20.22
1798 -19.40 -20.52 1834 -20.52 -19.97 1870 -20.87 -20.48 1906 -21.78 -21.66
1799 -20.26 -19.92 1835 -20.01 -19.70 1871 -19.90 -20.56 1907 -21.95 -20.56
1800 -18.89 -19.90 1836 -19.33 -19.49 1872 NA -19.78 1908 -21.49 -21.44
1801 -19.53 -19.46 1837 -19.82 -20.23 1873 NA NA 1909 -21.87 -20.60
1802 -20.25 -20.50 1838 -20.71 -20.51 1874 -20.26 -20.16 1910 -21.31 -20.10
1803 -19.66 -21.87 1839 -21.19 -21.81 1875 -20.65 -20.29 1911 -21.53 -20.95
1804 -20.09 -19.37 1840 -20.90 -20.05 1876 -20.70 -20.99 1912 -21.56 -21.03
1805 -18.62 -20.29 1841 -19.47 -19.39 1877 -20.08 -19.86 1913 -21.07 -20.56
1806 -19.67 -19.13 1842 -19.16 -19.74 1878 -20.86 -20.55 1914 -21.94 -20.56
1807 NA NA 1843 -20.04 -20.32 1879 -21.58 NA 1915 -21.40 -20.42
1808 -19.18 -19.14 1844 -20.15 -19.83 1880 -20.02 -19.59 1916 -21.14 -21.10
1809 NA NA 1845 -19.59 -19.21 1881 -20.86 -20.72 1917 -23.07 -22.45
1810 -18.97 -19.10 1846 -19.98 -20.19 1882 -19.94 -21.94 1918 -21.44 -20.71
1811 -19.03 -18.58 1847 -20.14 -20.11 1883 -20.52 -20.62 1919 -20.86 -21.12
1812 -18.98 -20.16 1848 -20.09 -20.15 1884 -20.60 -19.88 1920 -21.42 -20.65
1813 NA NA 1849 -20.64 -20.11 1885 -21.11 -20.64 1921 -21.98 -21.53
1814 -18.96 -19.93 1850 -20.51 -20.18 1886 -21.29 -20.64 1922 -22.72 -23.32
1815 -19.38 -19.70 1851 -19.87 -20.08 1887 -21.82 -20.83 1923 -20.70 -20.40
1816 -19.76 -20.21 1852 -21.20 -21.35 1888 -21.86 -21.19 1924 -19.88 -19.65
1817 -19.64 -19.94 1853 -20.70 -19.95 1889 -21.43 -20.72 1925 -20.04 -20.16
1818 -20.00 -20.14 1854 -20.43 -20.08 1890 -21.36 -20.94 1926 -21.37 -20.51
1819 -20.20 -20.43 1855 -19.87 -19.41 1891 -22.11 -20.99 1927 -20.29 -19.95
117
13 13 13 13 13 13 Year δ CEW δ CLW Year δ CEW δ CLW Year δ CEW δ CLW
1929 -20.55 -20.84 1965 -22.10 -21.82 2001 -20.07 -19.27
1930 -20.74 -20.04 1966 -20.77 -18.75 2002 NA NA
1931 -21.15 -20.39 1967 -19.74 -20.77 2003 -19.45 -19.36
1932 -22.25 -21.60 1968 -20.92 -20.28 2004 -20.10 -19.45
1933 -20.28 -20.96 1969 -21.35 -20.65 2005 -21.06 -20.95
1934 -19.32 -19.65 1970 -20.58 -19.93 2006 -20.26 -19.83
1935 -20.65 -20.43 1971 -20.67 -19.87 2007 -18.64 -19.16
1936 -20.96 -21.07 1972 -19.70 -19.15 2008 -19.85 -19.85
1937 -21.18 -20.37 1973 -21.00 -19.73 2009 -20.46 -19.75
1938 -21.19 -21.16 1974 -19.98 -19.42 2010 -20.51 -19.96
1939 -21.66 -20.17 1975 -20.18 -19.94 2011 -20.93 -20.41
1940 -20.55 -20.45 1976 -22.28 -21.42 2012 -19.07 -19.61
1941 -22.57 -22.63 1977 -20.51 -19.73 2013 -19.33 -19.60
1942 -20.41 -20.19 1978 -21.14 -19.38 2014 -19.53 -19.63
1943 -21.32 -19.95 1979 -21.06 -20.12 2015 -19.60 -19.83
1944 -20.91 -19.84 1980 -21.46 -19.78 2016 -21.01 -19.36
1945 -20.26 -20.67 1981 -20.16 -19.63 2017 -20.58 -20.27
1946 -20.24 -20.59 1982 -20.72 -20.44
1947 -20.66 -19.89 1983 -21.54 -21.43 1948 -20.73 -20.73 1984 -20.35 -21.90 1949 -21.56 -20.57 1985 -22.71 -21.99 1950 NA NA 1986 -20.67 -19.96 1951 -20.26 NA 1987 -21.44 -20.05 1952 -20.58 -20.12 1988 -21.21 -20.50 1953 -20.02 -20.15 1989 -19.75 -19.01 1954 -20.26 -19.98 1990 -19.57 -19.98 1955 -20.64 -21.44 1991 -20.85 -19.69 1956 -21.14 -20.34 1992 -21.22 -19.91 1957 -20.79 -20.15 1993 -22.68 -21.64 1958 -20.83 -19.90 1994 -20.52 -19.94 1959 -19.44 -19.91 1995 -21.82 -20.12 1960 -20.54 -19.86 1996 -18.92 -18.43 1961 -19.97 -20.01 1997 -18.70 -19.10 1962 -20.27 -19.89 1998 -21.50 -20.96 1963 -19.79 -19.71 1999 -20.83 -20.71
118
Appendix H
13 Table 1: Linear models of the relationship between paleoclimate proxies using δ CCELL and historical PRISM precipitation data during the instrumental period (1948-2017).
Climate Variable Proxy Linear Model Adjusted-R2 p-value 13 13 JJAS δ CDIFF JJAS = -60.1* δ CDIFF + 147.8 0.29 < 0.0001 13 13 δ CEW JJAS = 13.9* δ CEW + 405.3 0.01 0.18 13 13 δ CLW JJAS = -30.9* δ CLW - 500.4 0.08 0.01
13 13 JJA δ CDIFF JJA = -50.7* δ CDIFF + 113.7 0.3 < 0.0001 13 13 δ CEW JJA = 4.2* δ CEW + 176.3 0 0.63 13 13 δ CLW JJA = -36.5* δ CLW - 642.6 0.18 0.0002
13 13 JAS δ CDIFF JAS = -60.3* δ CDIFF + 140.1 0.3 < 0.0001 13 13 δ CEW JAS = 15.5* δ CEW + 430.7 0.02 0.14 13 13 δ CLW JAS = -28.8* δ CLW - 465.7 0.07 0.02
13 13 Cool-Season δ CDIFF Cool-Season = 83.7* δ CDIFF + 246.5 0.16 0.0005 13 13 δ CEW Cool-Season = -92.1* δ CEW -1607.6 0.33 <0.0001 13 13 δ CLW Cool-Season = -54.1* δ CLW -800.9 0.07 0.02
13 13 fSUMMER δ CDIFF fSUMMER = -0.4* δ CDIFF + 0.71 0.42 < 0.0001 13 13 δ CEW fSUMMER = 0.21* δ CEW + 4.83 0.18 0.0002 13 13 δ CLW fSUMMER = -0.05* δ CLW - 0.43 0 0.5
119
Appendix I The following table and figures include a preliminary analysis of anomalies, droughts, and false rings from the tree-ring isotope chronologies. A possible relationship between 13 δ CDIFF and atmospheric rivers is represented by the last figure.
13 Table I1. List of anomalies (-2.0 ≥ z-scores ≥ 2.0) in the δ CDIFF isotope chronology and presence of false rings
13 Year δ CDIFF False Ring 1642 2.0 Absent 1647 2.0 Absent 1649 2.0 Absent 1650 -2.5 Present 1651 2.0 Absent 1660 2.0 Absent 1661 2.0 Absent 1664 2.5 Absent 1677 2.0 Absent 1726 2.0 Absent 1727 2.2 Absent 1734 2.0 Absent 1803 -3.5 Absent 1805 -2.7 Absent 1829 -2.2 Absent 1882 -4.0 Absent 1893 -2.2 Absent 1896 -2.2 Absent 1904 -2.0 Absent 1955 -2.0 Present 1966 2.7 Absent 1967 -2.3 Absent 1978 2.3 Absent 1980 2.2 Absent 1984 -3.3 Present 1995 2.2 Absent 2016 2.2 Absent
120
13 Figure I1. The δ CCELL in the earlywood and latewood of tree rings for individual trees
13 during drought periods for a) 1948-1955 and b) 2010-2017. Positive δ CDIFF (z-scores)
13 were measured in the 1949 (1.0) and 2016 (2.0) tree rings. A negative δ CDIFF was measured in in the 1955 (-2.0) tree ring. 1955 and 2016 were classified as anomalies in
13 the δ CDIFF tree-ring isotope chronology.
* = significant difference between earlywood and latewood (p < 0.1)
121
Figure I2. The difference between δ13C in the α-cellulose of latewood and earlywood
13 (δ CDIFF) when a false ring was absent or present in an annual tree ring pre- and post-
Industrial Revolution (~1850 AD). Notice the y-axis units are parts per mil (‰) instead of z-scores.
122
Figure I3. Sum of winter integrated water vapor transport (IVT) associated with
13 atmospheric rivers (ARs) that made landfall at 32.5 ˚N to values of δ CDIFF. AR chronology from Gershunov et al. (2017).