Response of tree radial growth to evaporation, as indicated by earlywood and latewood

 

Respuesta del crecimiento radial a la evaporación, a partir de maderas temprana y tardía

 

José E. Chacón-de la Cruz; Marín Pompa-García*

 

Facultad de Ciencias Forestales, Universidad Juárez del Estado de Durango. Río Papaloapan y bulevar Durango s/n, col. Valle del Sur. C. P. 34120. Durango, Durango, México. Correo-e: mpgarcia@ujed.mx Tel.: 01 618 130 1096 (*Autor para correspondencia).

 

Received: October 28, 2014.
Accepted: Junuary 12, 2015.

 

ABSTRACT

Most dendrochronological studies are based on the relationship between total radial growth and precipitation; however, very few studies have considered the responses of early (EW) and late (LW) wood growth or the role of evaporation. In this study, residual indices for forests in the state of Durango (Mexico) during the period 1964-2010 were used to evaluate the response of EW and LW growth in Pinus cooperi Blanco to evaporation. DENDROCLIM software was used to correlate dendrochronological data to total monthly evaporation. Evaporation from January of the previous year to December of the current year was considered in a simple interval analysis. The coefficients of correlation indicated a consistent negative relationship between both EW and LW growth and evaporation in the fall/winter prior to the growth season. Nevertheless, LW growth was more consistent in terms of climatic sensitivity, which may indicate that measurement of this parameter may be more useful in dendroclimatic studies.

Keywords: Dendrochronology, correlation, sensitivity, hydric stress.

 

RESUMEN

La mayoría de los estudios dendrocronológicos se basan en la relación del crecimiento radial y la precipitación; sin embargo, pocos estudios han considerado las respuestas de la madera temprana (EW, por sus siglas en inglés) y tardía (LW, por sus siglas en inglés) al efecto de la evaporación. En este estudio se usaron índices residuales para evaluar la respuesta de crecimiento de EW y LW de Pinus cooperi Blanco al efecto de la evaporación en bosques del estado de Durango, México, durante el periodo 1964-2010. El software DENDROCLIM fue utilizado para correlacionar datos dendrocronológicos con la evaporación total mensual. En un análisis de intervalo simple se utilizaron datos de evaporación a partir del mes de enero del año previo a la estación de crecimiento a diciembre del año de la formación del anillo. Los coeficientes de correlación indicaron una relación negativa consistente de EW y LW con la evaporación del otoño-invierno previo a la estación de crecimiento. No obstante, el crecimiento de la madera tardía fue más consistente en términos de sensibilidad climática, lo cual puede indicar que la medición de dicha variable puede ser de utilidad en estudios dendroclimáticos.

Palabras clave: Dendrocronología, correlación, sensibilidad, estrés hídrico.

 

INTRODUCTION

The climatic parameters temperature and precipitation are often considered in dendrochronological studies because they are directly related to soil moisture content and evaporative demand (rising vapour pressure deficit). Measurement of these parameters has become standard for evaluation of radial growth in response to climatic variations. It has proven valuable for reconstructing chronologies (Cerano-Paredes et al., 2013; Stahle et al., 2011), analyzing the effects of local (Gochis, Brito-Castillo, & Shuttleworth, 2006) and global (Yocom et al., 2010) climate change, and constructing future scenarios in which climate, vegetation and water availability play a principal role (Meko et al., 2013). However, evaluation of radial growth as a function of hydric stress associated exclusively with precipitation and temperature may overlook information provided by other variables. For example, the effect of evaporation on growth has yet to be fully established. This could yield new data for the study of climate-plant relationships, as evaporation is governed by factors such as solar incidence, wind speed and relative humidity. Pompa-García, Rodríguez-Flores, Aguirre-Salado, and Miranda-Aragón (2013) modeled evaporation dynamics in relation to radial growth in Pinus cooperi Blanco during the period 1964-2010 and found that evaporation plays an important role in increasing moisture stress.

These authors also suggested additional effects of drought and extreme temperatures. The tree ring index (TRI) is commonly used as an indicator of response to changes in climate (Woodhouse & Lukas, 2006); however, recent studies have explored the potential use of late and early wood growth measurements, particularly in markedly seasonal environments (Meko et al., 2013). For example, Griffin, Meko, Touchan, Leavitt, and Woodhouse (2011) considered how both early wood (EW) and late wood (LW) growth were correlated with seasonal precipitation in Pseudotsuga menziesii (Mirb.) Franco. Although total and early growth were reliable indicators of precipitation in the dry season, these parameters did not reflect the changes in this variable during the wet season. However, late growth did not respond to variations in precipitation between October and April, but was highly sensitive to rain during the wet season.

The present study considers the association between evaporation and both LW and EW growth in P. cooperi. This coniferous tree, which has high ecological and economic value, is representative of northern Mexico and its growth has previously been shown to relate to climatic drivers (Cruz-Cobos, De los Santos-Posadas, & Valdez-Lazalde, 2008; Pompa & Jurado, 2013), but which has not previously been the subject of such analysis. In this context, the aim of the study was to compare the responses of EW and LW growth to instrumental evaporation in the Sierra Madre Occidental mountain range, a zone that is influenced by climatic patterns with ecological, economic and hydrological implications for the whole region. We hypothesized that EW and LW are statically different in climatic sensitivity.

 

MATERIALS AND METHODS

The study site was located in the Sierra Madre Occidental in the state of Durango, Mexico, between coordinates 24° 8' 27'' N - 105° 3' 18'' W and 24° 4' 43'' N - 105° 1' 15'' W and at a mean elevation of 2362 m (Figure 1). The vegetation in the area is predominated by Pinus and Quercus species (González-Elizondo, González-Elizondo, & Márquez, 2007).

At least three wood cores were removed from each of 18 P. cooperi trees, at a height of 1.3 m, with a Pressler borer (Haglof, Sweden). A low quality site with the following characteristics was chosen for study: elevation 2362 m, slope no greater than 5°; and semi-cold temperate climate with rainfall in summer. Only undamaged trees with no deformities or signs of competition for light or nutrients were included in the study, to highlight the environmental effects on the growth of the trees and to maximize the climatic signal. The expressed population response (EPS) was calculated as an indicator of sample size sufficiency, as often done in paleoclimatic studies (Mérian, Pierrat, & Lebourgeois, 2013). The cores were dried, mounted and sanded. Crossed dating was then conducted according to Stokes and Smiley (1996). Tree growth was measured under a stereoscopic microscope, and EW and LW growth were differentiated, to an accuracy of 0.001 mm, by using a Velmex 117 system (USA). The quality of the dating and measurements were evaluated using COFECHA software (Holmes, 1983).

Biological and geometric effects not associated with climate were minimized using ARSTAN software (Cook & Holmes, 1984) to obtain standardized residuals of annual EW and LW with a mean of 1 and 0 variance. In this process, the series were fitted to a negative exponential function and were then processed using the spline interpolation technique for cubic softening to reduce the non-climatic variance while preserving the annual or sub-decadal information. Early residual (ER) and late residual (LR) indices of radial growth were calculated by dividing the observed radial growth by the adjusted values. The results were analyzed by using autoregression models to eliminate temporal autocorrelation.

Monthly evaporation (mm) for the period 19642010 was considered as a climatic variable. This data was obtained from the Santa Bárbara (Durango) climatological station (Comisión Nacional del Agua [CONAGUA], 2012), located less than 50 km from the sampling sites. The recordings included the period from January of the year prior to the growth season until December of the current year (Figure 1).

The effect of evaporation on EW and LW growth was determined by using the DENDROCLIM2002 program (Biondi & Waikul, 2004) to calculate Pearson's correlation coefficients between evaporation and both EW and LW and random resampling to estimate the significance of the correlations. This process identified the months in which association between these variables was the strongest. Furthermore, we also evaluated the association of LW and EW to winter evaporation (WEJ variability within this region, using scatter plots.

 

RESULTS AND DISCUSSION

A total of 28 core samples (mean diameter at breast height, 48.2 cm) from P. cooperi were analyzed. The EPS value was 82 %. The chronology had a duration of 177 years, between 1834 and 2010, while the study period comprised 47 years during the period from 19642010. The EPS value provided valuable information to confirm the quality of the assessment of climate-growth relationships (Mérian et al., 2013). The tree-ring index for EW and LW showed similar trends during the period they overlapped, indicating that the P. cooperi shared an interannual growth variation (Figure 2). In general, evaporation showed an inverse relationship with tree-ring growth.

Figure 2 shows a comparison between EW and LW growth and evaporation, expressed as a ring index (RI). For EW growth, the smallest increase occurred in 1983 and the maximum increase in 2010, while LW growth was minimal in 1991 and maximal in 2010. A negative association between both growth types and evaporation was profiled graphically throughout the series. Short periods with positive responses were evident in 19671968, 1970-1971, 1975-1977, 1993-1994 and 2009-2011; however, LW growth was more consistent and closely reflecting than the magnitude and direction of the variations in evaporation, which supports the hypothesis that climatic signal may be better represented by analyzing EW and LW growth separately.

Negative correlation coefficients were obtained, with values of -0.38 to 0.29 (P < 0.05) for LW and ER; these are considered low, but significant. Evaporation was negatively and constantly correlated with LW growth in the period between September of the previous year and January of the current year, as well as in June and September of the same year. The association between evaporation and EW growth was variable. Evaporation was positively correlated with EW growth in the months of February, April and May prior to the growth stage and negatively with EW growth between September and January. In the period September to February, the correlation values and trends for both types of wood growth were similar (Figure 3).

In general, the correlation coefficients were low; however, the trends were consistent during the study period, especially in the months of September, October and November of the previous year, for both types of wood growth, and in the period February to April, for EW growth.

Graphs showing the influence of WE on EW and LW are displayed on Figure 4. WE exhibits negative strong association on radial tree growth (r = -0.4). In a reconstruction of winter-spring precipitation by using P. menziesii, Díaz, Therrell, Stahle, and Cleaveland (2002) found a positive association between EW growth and precipitation for the period November to May. This pattern reflects the effect of the moisture produced by rain, but does not distinguish the effect of evaporation, especially between March and May, when hydric stress may be greater because of the lack of rain. This could explain the contrasting growth responses to drought before or during the growing season (Figure 3 and Figure 4). In the analysis prior to the reconstruction of precipitation based on TRI, Pompa and Jurado (2013) report a positive response to the availability of water and a graphic correspondence between precipitation and TRI for the same period analyzed in the present study, in an area neighboring the study site. However, the correlation coefficients found in the present study for LW growth with respect to evaporation are higher than those reported for the relationship between TRI and precipitation.

Pompa et al. (2013) also reported a negative effect of evaporation on growth. The results of the aforementioned study are consistent with those of the present study in terms of the direction of the correlation, but differ in intensity, indicating a greater response of total growth than of LW growth. However, although the response of LW growth was less intense, it was more seasonally specific and it was concentrated in the dry period between September and January, while total growth responded more variably over a wider period including both wet and dry months. In an exploration of the relationship between LW growth and seasonal precipitation, Griffin et al. (2011) reported high positive correlation coefficients for the wet season, but a weak response to the hydric stress caused by the low precipitation in the dry season. This emphasizes the need to eliminate the effect of EW on LW growth for a more accurate explanation of the variations in the rains caused by the North American monsoon. A climate reconstruction based only on LW growth revealed that the historical occurrence of recurrent and random extreme precipitation events is an effect of the monsoon region, which contrasts with the stability recorded in the last century for this area (Griffin et al., 2013). In this sense, the results of the present study indicate that LW growth could be used in the study region to directly evaluate the response of P. cooperi to hydric deficit during stages prior to growth, without the application of any adjustments with respect to the effect of EW growth.

Our findings have notable implications to interpret the role played by evaporation on tree-ring growth. We show that LW is also very sensitive to water deficit and high evaporation rates which suggests that evaporation by itself may be a likely direct driver of growth decline and forest dieback rather than warmer temperatures or low rainfall (Liang, Eckstein, & Liu, 2008). Radial growth shows considerable dependence on the climatic conditions of the winter months preceding the growing season and hence on evaporation conditions.

 

CONCLUSIONS

EW and LW growth respond negatively to evaporation in P. cooperi. In particular tree radial growth was constrained by drought stress related to high evaporation rates during the previous winter and the current spring. However, LW growth proved a particularly sensitive indicator of hydric deficit by evaporation during the winter prior to the growth season. In contrast, EW growth responded to evaporation positively in spring and negatively during the summer prior to the growth season, albeit with a lower degree of association. These characteristics highlight the potential usefulness of LW growth as a study parameter for climatic reconstruction, prognosis of climatic trends and the analysis of markedly seasonal forest ecosystems such as those in the Sierra Madre Occidental. Dry and hot winters in northern Mexico negatively affect forest productivity. This negative impact indicates that rising evaporation rates constitute an important driver of forest growth decline.

 

ACKNOWLEDGEMENTS

We would like to thank the support given by Julián Cerano and Omar Durán at the INIFAP CENID-RASPA Lab in Mexico, who helped in the gathering and processing of dendrochronological data. We thank Carlos A. Aguirre-Salado, three anonymous reviewers and the editorial team for their useful comments that improved the manuscript.

 

REFERENCES

Biondi, F., & Waikul, K. (2004). DENDROCLIM2002: A C program for statistical calibration of climate signals in tree-ring chronologies. Computers & Geosciences, 30, 303-311. doi: 10.1016/j.cageo.2003.11.004.         [ Links ]

Cerano-Paredes, J., Méndez-González, J., Amaro-Sánchez, A., Villanueva-Díaz, J., Cervantes-Martínez, R., & Rubio-Camacho, E. A. (2013). Reconstrucción de precipitación invierno-primavera con anillos anuales de Pinus douglasiana en la Reserva de la Biosfera Sierra de Manantlán, Jalisco. Revista Chapingo Serie Ciencias Forestales y del Ambiente, 19(3), 413-423. doi: 10.5154/r.rchscfa.2013.02.007.         [ Links ]

Comisión Nacional del Agua (CONAGUA). (2012). Consultado 15-04-2013 en http://smn.cna.gob.mx/emas/.

Cook, E. R., & Holmes, R. H. (1984). Program ARSTAN user manual. Tucson, AZ, USA: University of Arizona Press.         [ Links ]

Cruz-Cobos, F., De los Santos-Posadas, H. M., & Valdez-Lazalde, J. R. (2008). Sistema compatible de ahusamiento-volumen para Pinus cooperi Blanco en Durango, México. Agrociencia, 42, 473-485. Obtenido de http://www.scielo.org.mx/pdf/agro/v42n4/v42n4a10.pdf.         [ Links ]

Díaz, S. C., Therrell, M. D., Stahle D. W., & Cleaveland, M. K. (2002). Chihuahua (Mexico) winter-spring precipitation reconstructed from tree-rings 1647-1992. Climate Research, 22, 237-244. Obtenido de http://www.intres.com/articles/cr2002/22/c022p237.pdf.         [ Links ]

Gochis, D. J., Brito-Castillo, L., & Shuttleworth, W. J. (2006). Hydroclimatology of the North American Monsoon region in northwest Mexico. Journal of Hydrology, 316, 53-70. Obtenido de http://www.cpc.ncep.noaa.gov/products/outreach/proceedings/cdw29_proceedings/Gochis.pdf.         [ Links ]

González-Elizondo, M. S., González-Elizondo, M., & Márquez, M. A. (2007). Vegetación y ecorregiones de Durango. México: Plaza y Valdés.         [ Links ]

Griffin, D., Meko, D. M., Touchan, R., Leavitt, S. W., & Woodhouse, C. A. (2011). Latewood chronology development for summer-moisture reconstruction in the U. S. Southwest. Tree-Ring Research, 67(2), 87-101. doi: 10.3959/2011-4.1.         [ Links ]

Griffin, D., Woodhouse, C. A., Meko, D. M., Stahle, D. W., Faulstich, H. L., Carrillo, C., ...Leavitt, S. W. (2013). North American monsoon precipitation reconstructed from tree-ring latewood. Geophysical Research Letters, 40(5), 1-5. doi: 10.1002/grl.50184.         [ Links ]

Holmes, R. L. (1983). Computer-assisted quality control in tree-ring dating and measurement. Tree-Ring Bulletin, 43, 69-78. Obtenido de http://arizona.openrepository.com/arizona/bitstream/10150/261223/1/trb-43-069-078.pdf.         [ Links ]

Liang, E. Y., Eckstein, D., Liu, H. Y. (2008). Climate-growth relationships of relict Pinus tabulaeformis at the northern limit of its natural distribution in northern China. Journal of Vegetation Science, 19, 393-406. doi: 10.3170/20088-18379.         [ Links ]

Meko, D. M., Touchan, R., Villanueva-Díaz, J., Griffin, D., Woodhouse, C. A., Castro, C. L., ...Leavitt, S. W. (2013). Sierra San Pedro Mártir, Baja California, cool-season precipitation reconstructed from earlywood width of Abies concolor tree rings. Journal of Geophysical Research: Biogeosciences, 118,1660-1673. doi: 10.1002/2013JG002408.         [ Links ]

Mérian, P., Pierrat, J. C., & Lebourgeois, F. (2013). Effect ofsampling effort on the regional chronology statistics and climate-growth relationships estimation. Dendrochronologia, 31, 58-67. doi:10.1016/j.dendro.2012.07.001.         [ Links ]

Pompa-García, M., & Jurado, E. (2013). Seasonal precipitation reconstruction and teleconnections with ENSO based on tree ring analysis of Pinus cooperi. Theoretical and Applied Climatology. doi: 10.1007/s00704-013-1018-6.

Pompa-García, M., Rodríguez-Flores, F. J., Aguirre-Salado, C. A., & Miranda-Aragón, L. (2013). Effect of evaporation on forest growth. Agrociencia, 47, 829-836. Obtenido de http://scielo.unam.mx/pdf/agro/v47n8/v47n8a7.pdf.         [ Links ]

Stahle, D. W., Villanueva-Diaz, J., Burnette, D. J., Cerano-Paredes, J., Heim, J. R. R., Fye, F. K., ...Stahle, D. K. (2011). Major Mesoamerican droughts of the past millennium. Geophysical Research Letters. doi: 10.1029/2010GL046472.

Stokes, M. A., & Smiley, T. L. (1996). An introduction to tree-ring dating. Tucson, AZ, USA: University of Arizona Press.         [ Links ]

Woodhouse, C. A., & Lukas, J. J. (2006). Multi-century tree-ring reconstructions of Colorado streamflow for water resource planning. Climate Change, 78, 293-315. doi: 10.1007/s10584-006-9055-0.         [ Links ]

Yocom, L. L., Fulé, P. Z., Brown, P. M., Cerano, J. N., Villanueva-Díaz, J., Falk, D. A., & Cornejo-Oviedo, E. (2010). El Niño-Southern Oscillation effect on a fire regime in northeastern Mexico has changed over time. Ecology, 91, 1660-1671. Obtenido de http://library.eri.nau.edu/gsdl/collect/erilibra/index/assoc/HASH0e72.dir/doc.pdf.         [ Links ]