Dendrochronology and Isotope Chronology of Juglans neotropica and Its Response to El Niño-Related Rainfall Events in Tropical Highlands of Piura, Northern Peru

Abstract

Tropical trees represent an important potential archive of climate and ecological information, but their dendrochronology based on conventional techniques has been challenging. We conducted a pilot study of the wood anatomy and dendroclimatological potential of Juglans neotropica Diels (Juglandaceae), an IUCN Red List species, using 225 radii sampled from 57 trees in Piura (4°55' S, 79° 56' W), northern Peru. A total of 112 radii from 40 trees passed quality control and are included in the tree-ring width chronology for this species. J. neotropica has demonstrably annual rings, and results are consistent with reports that the species has a dormant period during the dry season, which locally is approximately June-November. Local precipitation is correlated (p = 0.10, 1-tailed test) with tree-ring growth, lagged by one year, consistent with other studies of tropical tree species. The age distribution of the sample collection of J. neotropica is young and invariant, probably because of selective cutting by local villagers. To supplement ring-width analysis, we conducted the first oxygen isotopic (δ¹⁸O) and radiocarbon (∆¹⁴C) analysis for this species on radii from two individuals; results are preliminary given sample size limitations, but consistent with dendrochronological dating, within uncertainties, in all three chronometric analyses. A two-sample composite annually-averaged δ¹⁸O anomaly data series is correlated significantly with gridded regional growing season (December-May) precipitation (1973/74-2005/06). Qualitatively consistent with simulation of ring width and δ18O, responses to El Niño events are manifested in positive ring-growth anomalies and negative isotopic anomalies following known event years. The combination of tree-ring, radiocarbon, stable isotopic analyses, and the application of sensor and chronological modeling provides a degree of confidence in the results that would not have been possible by relying on any single approach and indicates the potential for further investigation of this and other tropical tree species with uncertain ring boundaries.

Keywords: El Niño Southern Oscillation (ENSO); climate variability; isotopic analysis; oxygen 18; radiocarbon dating; tropical forests; wood anatomy.

Figure 1

Study area and locations of sampling and climate stations.

Figure 2

Monthly maximum, mean and minimum precipitation (mm) in Frias, Peru for 1963–2003. (Source: [113]). Standard box plot convention is followed (box encloses 25th–75th percentile with median line; outer bars are 10th and 90th percentiles; outliers are shown as dots). Horizontal bars indicate the mean periods of flowering and fruiting of J. neotropica [93].

Figure 3

Microscope images of J. neotropica microtone slices showing ring boundaries (arrows) near the core (A), in the transition zone from heartwood to sapwood (B) and near the bark (C). A piece of the bark is seen in the lowermost part of (C), indicating the direction of growth from top to bottom. Scale bars in mm. (D,E) Wood structure images. All photos by T.M. Ektvedt.

Figure 4

(Top) J. neotropica standard chronology from 1962/63–2005/06 compared to sample size (number of trees, grey line). The bars on the standard chronology are ±1 standard error of the standard chronology among the samples. Chronology values from 1962–1971, which may not be reliable due to small sample size, are indicated as a dashed line. (Middle) Age-modeled oxygen isotope records of JN37-1W (blue) and JN41-2N (red), with monthly-interpolated mean values in black, and radiocarbon age estimates (see Table 2) plotted as nablas with 2s error bars (top). Bottom lines with circles are error bars by decade centered on the decade midpoint: 2000s ± 2 years, 1990s ± 2 years, 1980s ± 3 years, 1970s ± 4 years, 1960s ± 5 years. (Bottom) Precipitation at Frias (black bars) and Santo Domingo (grey bars). El Niño years (yellow shading on years) are based on the Oceanic Niño Index [142], with event intensity grading (weak (w), moderate (m), strong (s) and no info (-) (Table 3). The two strong El Niño-events that resulted in precipitation of more than 3000 mm are shaded for comparison across panels.

Figure 5

Cumulative growth analyses for study samples of J. neotropica. (a) Cumulative incremental growth (mm) vs. time for 83 increment series in the J. neotropica chronology reported in this study. Darker (lighter) greyscale indicates longer (shorter) time interval spanned per increment core. (b) As in (a), except for time normalized to the length of each increment series, respectively. (c) As in (a), except for cumulative growth normalized by total growth of each series, respectively. The vertical dashed line indicates the 1982/83 El Niño event. (d) As in (c), except for time normalized by length of each increment series, respectively. Blue curve is the median of the median normalized growth-time values, with all normalized cumulative growth series interpolated to a common 12-point linear normalized timescale. Dashed lines trace the values of the median normalized cumulative growth (0.5), reached at 0.39 of the median normalized interpolated age.