Geography Department, University of Northern Iowa ...

400+ Years of ENSO-like Climate Cyclicity from Tree Ring Width-data, Wind River Range, Wyoming, USA
 Dennis Dahms - Derek Richards - Patrick Pease

ABSTRACT Spectral analysis of detrended ring width-data from a series of +400-year-old Douglas Fir trees on the SE flank of the Wind River Range indicates that tree growth from 1589 to 2013 shows a 2.5 to 4.5-year cyclicity (99%). This falls within the limits of the generally-accepted 2-7 year ENSO cyclicity for the western Pacific. Our results also show a 16-year frequency (95%) suggesting possible additional influence from the Pacific Decadal Oscillation (PDO). Ring-widths here are most closely correlated to soil moisture conditions through the interaction(s) of abundant winter snowpack, summer rainfall, and average May-thru-August temperatures during the 424 years from 1589 to 2013. Nearby climate records from the 1948 to 2013 period show that more favorable growth conditions occur here (higher snowpack + summer precipitation) during the El Niño cycle of ENSO. Our results fill a gap in knowledge of ENSO-like teleconnections during the Late Holocene that exists for the southern region of the Greater Yellowstone Ecosystem.

Geography Department, University of Northern Iowa

Cedar Falls, Iowa

(C) (A) (B) (D)

Figure 1: (A) Location of the Wind River Range (Wyoming) in the western U.S. (B) Location of the lower Middle Popo Agie basin. (C) Location of Sinks Canyon in the lower Popo Agie basin SW of Lander. (D) Location on the North rim of Sinks Canyon where Douglas Firs were sampled for this study. (Image source: www.flashearth.com.

METHODS SAMPLING Cores from 14 Douglas Firs with minimum diameters of ca. 1 meter were obtained in June, 2013. Trees were sampled within a ~5 km2 area on North-to-SE slopes above the North rim of Sinks Canyon (Fig 1-D). Trees were sampled mostly from the perimeters of mixed stands of Douglas Fir-Lodgepole Pine. Each tree was cored twice at offset of ~90o. Ring-thickness was measured to 0.001 mm precision in February-April 2009 using a VELMEX‘TA measurement system with an Acu-Rite SENC150 glass scale rotational linear encoder (Acu-Rite Companies, Inc.) and Quadra-Chek-10 digital encoder (Metronics, Inc.) with MEASURE J2X software (VoorTech Consulting). Ring thicknessmeasurements were transferred from the J2X software to Excel format (.xls). Cross-dating was performed using both the skeleton plot and graphical methods (Cropper, 1979; Gartner, 2007; Schweingruber, 1988; Stokes & Smiley, 1996). COFECHA was then used to check the cross-dating (Holmes, 1983). ARSTAN (AutoRegressive Standardization) was then used to remove the natural growth function and to produce a stand-level detrended chronology (Cook, 1985; Salzer & Kipfmueller, 2005; Speer, 2010). A negative exponential curve best-fit our data.

CLIMATE DATA

Climate data was obtained from the Historical Climatology Network and National Climatic Data Center for the climate stations of Townsend Creek SNOTEL (snow telemetry; 1981-2013 for summer precipitation and winter snow water equivalent SWE; 1990-2013 for temperatures) and the Lander airport (1948-2013 for summer temperature and precipitation). The May-through-August monthly averages were used as this is considered to be the general growing season and climate conditions during these months should have the most impact on growth

CLIMATE RECONSTRUCTION

COFECHA analysis shows a series inter-correlation of 0.022 with an average mean sensitivity of 0.273. Thus, the chronology has weak series inter-correlation but the rings are sensitive enough to show variability. Decade averages decreased the variability of the single-yr reconstruction. The coolest decade was the 1840’s with an average of 57°F (14°C). The warmest decade was the 2000’s with an average of 67°F (19°C). Figure 2a: Reconstructed temperatures vs the instrumental temperature record from Lander Airport and Townsend Ck SNOTEL from 1947-2013.

(Kipfmueller, 2008). Principal Components Analysis (PCA) and Stepwise Multiple Regression were used to explore the relationships between the detrended chronologies and climate data (Briffa et al., 1992; Gray et al., 2004; Kipfmueller, 2008; Salzer and Kipfmueller, 2005; and Woodhouse and Brown (2001). A ‘Transfer Function’ was created to build a model of past climate conditions from the ring-width data. The model was verified using the Predicted Residual Sum-ofSquares (PRESS) method (Kipfmueller, 2008; Salzer & Kipfmueller, 2005).

Temperature reconstruction was made for the 425 yrs of ring-width data (1589-2013) by using the transfer function with the detrended residual chronology (Kipfmueller, 2008). This extended the reconstructed temperatures back to 1589 despite very low sample depth in the earlier years of the reconstruction. The sample depth is the number of cores for each year in the chronology and, generally, a sample depth of greater than 10 is needed in order to accurately compare the ring-widths to climate (Speer, 2010). The sample depth here was 10 trees at 1775 and less than 5 trees before 1649.

SPECTRAL ANALYSIS We performed spectral analyses on both the detrended chronology and the temperature reconstruction using three separate analyses procedures: the Blackman-Tukey correlogram and CrossSpectra (BT) and the Multi-Taper-Method (MTM) using the Mac ‘kSpectra’ program (SpectraWorks, Inc). To verify the absence of aliasing, we performed stepwise re-interpolations of the data every 0.2 from 1.2-to-3.0 years. We repeated MTM analysis on each of the stepwise interpolations. All interpolations showed results similar to our original signals at 95-99% confidence.

SPECTRAL ANALYSIS

Analysis of the Detrended Ring Width-Data reveal cycles in ring thickness of 2.5-to-4.5 years, consistent with the generally-accepted 2.5-8-yr pacing of ENSO. Both the Multi-Taper (Fig. 3a) and Blackman-Tukey (not shown) methods show statistically significant confidence levels above 99%.

!

— Figure 3a

! ! ! ! ! ! ! !

Figure 2b: Reconstructed temperatures for the 425 yrs of the treering width data (1589-2013).

!

Figure 3b —

!

Analysis of the Temperature Reconstruction (Fig. 3b) also reveals ENSO-like cycles of 2.0-to-4.5 yrs (99%) along with a weaker frequency signal at ~16 years (90-95%). The ~16 yr frequency is consistent with the Pacific Decadal Oscillation (PDO).

INTERPRETATIONS CLIMATE RECONSTRUCTION — Ring widths failed to significantly correlate to the precipitation data, considering that previous studies in the Bighorn Mountains found precipitation to be significant (Gray et al. 2003, 2004; Brown & Wu 2005; Brown 2006). However, summer temperatures vs precipitation from the Lander Airport & Townsend Creek SNOTEL datasets yielded a correlation of -0.672. This suggests that ring widths here correlate most directly with soil moisture conditions. Presumably, cool summer temperatures allow for higher soil moisture conditions, while warm summer temperatures lead to lower soil moisture conditions (Madden & Williams, 1978) — thus, summer soil moisture is more a function of summer temperature than summer precipitatiion. ENSO — Our work corroborates Gray et al’s (2004) results that an apparent teleconnection that links ocean-atmosphere cycles of the Equatorial (ENSO-annual) and North (PDO-decadal) Pacific to annual and decadal temperature-precipitation patterns of the Wyoming Rocky Mountains. When we compared yearly SNOW MOISTURE + PRECIPITATION conditions from the Lander Airport and the Townsend Creek SNOTEL, we find that the known cycles of El Niño since 1947 coincide with slightly higher moisture conditions at these stations, both of which are within 15 km of our samples. Thus, some coincident pattern of El Niño and PDO apparently is the forcing mechanism driving the slightly more-favorable conditions that affect the growth patterns seen in the Douglas Firs within our study area. ACKNOWLEDGEMENTS

Thanks to the University of Iowa’s Center for Global and Regional environmental Research (CGRER) for travel support for the study (Richards), and to Paolo Cherubini (Zurich-ETH), Malcolm Hughes (Univ of Arizona) and Michael Mann (Penn State University) for their kind suggestions. References

Blackman, R. B., & Tukey, J. W., 1958. The measurement of power spectra from the point of view of communications engineering. New York, NY: Dover.

Briffa, K. R., Jones, P. D., & Schweingruber, F. H., 1992. Tree-ring density

reconstructions of summer temperature patterns across Western North America since 1600. Journal of Climate, 735-754.

Brown, P. M., 2006. Climate effects on fire regimes and tree recruitment in Black Hills ponderosa pine forests. Ecology 87, 10, 2500-10.

Brown, P. M., & Wu, R., 2005. Climate and disturbance forcing of episodic tree recruitment in a Southwestern ponderosa pine landscape. Ecology 86, 11, 3030-3038.

Cook, E., 1985. A time-series approach to tree-ring standardization. Ph.D. dissertation, The University of Arizona. Tucson, AZ.

Cropper, J. P., 1979. Tree-ring skeleton plotting by computer. Tree Ring Bulletin, 39, 48-59.

Gärtner, H., 2007. Glacial Landforms, Tree-rings and Dendrogeomorphology. Edited by Scott Elias. In Encyclopedia of Quaternary Science,Volume 2, p. 978-988.

Gray, S. T., Betancourt, J. L., Fastie, C. L., & Jackson, S. T., 2003. Patterns and

sources of multidecadal oscillations in drought-sensitive tree-ring records from the central and southern Rocky Mountains. Geophysical Research Letters, 30,

49.

Gray, S. T., Fastie, C. L., Jackson, S. T., & Betancourt, J. L., 2004. Tree-Ring-Based Reconstruction of Precipitation in the Bighorn Basin, Wyoming, since 1260 a.d. Journal of Climate, 17 (19), 3855-3865.

Grissino-Mayer, H., 2008. User guide to COFECHA output files.

http://www.ncdc.noaa.gov/paleo/treering/cofecha/userguide.html

Holmes, R., 1983. Computer-assisted quality control in tree-ring dating and measurement. Tree-Ring Bulletin, 43, 69-78.

Kipfmueller, K. F., 2008. Reconstructed summer temperature in the northern Rocky

Mountains wilderness, USA. Quaternary Research, 70 (2), 173-187.

Mann, M.E., & Lees. J., 1996. Robust Estimation of Background Noise and Signal Detection in Climatic Time Series. Climatic Change, 33, 409-445.

Paillard, D., Labeyrie, L., & Yiou, P., 1996. Macintosh Program performs time-series analysis. Eos, Transactions American Geophysical Union, 77 (39), 379.

Salzer, M., & Kipfmueller, K., 2005. Reconstructed Temperature and Precipitation on a

Millennial Timescale From Tree-Rings In The Southern Colorado Plateau,

U.S.A. Climatic Change, 70 (3), 465-487.

Schweingruber, F. H., 1988. Tree-rings: Basics and applications of dendrochronology. Dordrecht, Netherlands: D. Reidel Pub. Co.

Speer, J. H., 2010. Fundamentals of tree-ring research. Tucson: University of Arizona Press.

Stokes, M. A., & Smiley, T. L., 1996. An introduction to tree-ring dating. Tucson: University of Arizona Press.

Woodhouse, C.A., & Brown, P.M., 2001. Tree-ring evidence for great plains drought. Tree-Ring Research, 57 (1), 89-103.