Sun–earth relationship inferred by tree growth rings in conifers from Severiano De Almeida, Southern Brazil

Journal of Atmospheric and Solar-Terrestrial Physics 73 (2011) 1587–1593

Contents lists available at ScienceDirect

Journal of Atmospheric and Solar-Terrestrial Physics journal homepage: www.elsevier.com/locate/jastp

Sun–earth relationship inferred by tree growth rings in conifers from Severiano De Almeida, Southern Brazil A. Prestes a,n, N.R. Rigozo b, D.J.R. Nordemann c, C.M. Wrasse a, M.P. Souza Echer c, E. Echer c, M.B. da Rosa b, P.H. Rampelotto b a

~ Jose´ dos Campos, SP, Brazil Instituto de Pesquisa e Desenvolvimento, Universidade do Vale do Paraı´ba, Av. Shishima Hifumi, 2911, Urbanova, 12244-000 Sao Centro Regional Sul de Pesquisas Espaciais (CRS/INPE), P.O. Box: 5021, 97105-900, Santa Maria, RS, Brazil c Instituto Nacional de Pesquisas Espaciais, Caixa Postal 515, 12245-970 Sa~ o Jose´ dos Campos, SP, Brazil b

a r t i c l e i n f o

abstract

Article history: Received 31 March 2010 Received in revised form 20 November 2010 Accepted 28 December 2010 Available online 11 January 2011

This study of Sun–Earth relationships is based on tree growth rings analysis of araucarias (Araucaria angustifolia) collected at Severiano de Almeida (RS) Brazil. A chronology of 359 years was obtained, and the classical method of spectral analysis by iterative regression and wavelet method was applied to find periodicities and trends contained in the tree growth. The analysis of the dendrochronological series indicates representative periods of solar activity of 11 (Schwabe cycle), 22 (Hale cycle), and 80 (Gleissberg cycle) years. The result shows the possible influence of the solar activity on tree growth in the last 350 years. Periods of 2–7 years were also found and could represent a response of the trees to local climatic conditions. Good agreement between the time series of tree growth rings and the 11 year solar cycle was found during the maximum solar activity periods. & 2011 Elsevier Ltd. All rights reserved.

Keywords: Tree growth rings Sun–earth-climate relationship Spectral analysis Wavelet analysis

1. Introduction The environment at the surface of the Earth exists only because of the energy flux received from the Sun. Solar radiation influences atmospheric and oceanic circulations, which also influence the biosphere (National Research Council, 1994). Without solar radiation, photosynthesis stops. Solar radiation and high energy particles continuously collide into gases and plasmas in the atmosphere and magnetosphere, which protect life on Earth (Raisbeck and Yiou, 1984). Changes in the quantity of total solar energy input into the planet system are caused by three mechanisms: 1. Changes in Earth’s orbital parameters, i.e., the obliquity of Earth’s axis and the eccentricity of its orbit alter the incidence and distribution of the radiation incident on the planet (Eddy, 1980). 2. Processes inside the planet system, which regulate the quantity of energy received by the Earth (Lean et al., 1992). 3. Solar activity variations that modulate the energy emitted by the Sun (Wilson and Hudson, 1988). The study of solar variations related to the energy flux is completely observational and very recent (about the past 40 n

Corresponding author. E-mail address: [email protected] (A. Prestes).

1364-6826/$ - see front matter & 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jastp.2010.12.014

years), which limits the understanding of their effects on climate and the possibility of long-term climatic predictions for the future. For these reasons, it is necessary to indirectly monitor solar variations and other geophysical phenomena at a more extensive scale into the past. Tree growth rings, which represent records of chronological series, are witnesses of the environment and climate that affected their growth (Fritts, 1976). Several simultaneous environmental factors influence tree growth including: solar radiation, temperature, water precipitation and soil content, humidity, nutrients, neighborhood, pests, disease, etc. Depending on environmental conditions in which the tree and its species exist, some of these factors may prevail. Temperature, light and precipitation play an important role in regions with changing seasons and induce different growth rates caused by different cell size allowing direct visual recognition of the well-known tree growth rings. The thickness variation of yearly rings reflects the sensitivity of the tree to environmental factors at the location where it grows (Nordemann et al., 2005). Thus, ˜oprecipitation and temperature fluctuations caused by El Nin Southern Oscillation (ENSO) and other natural forcing mechanisms could have been recorded in tree growth rings. Because ENSO is known to have a very strong influence on the climate of South America (Neelin and Latif, 1998). In South America, research on tree ring chronologies have been mostly conducted for climate record studies based on sampling from Chile and Argentina (Hughes et al., 1982); however, there have been a few studies in Brazil. The Schwabe 11-yr solar cycle

1588

A. Prestes et al. / Journal of Atmospheric and Solar-Terrestrial Physics 73 (2011) 1587–1593

has been found in tree rings from Southern Brazil (Rigozo et al., 2002, 2004; Nordemann et al., 2005) and Chile (Nordemann et al., 2005; Rigozo et al., 2006), and the fourth harmonic of Suess cycle (52 yr) has been found in tree ring data from Conco´rdia, Brazil (Rigozo et al., 2004) and Osorno, Chile (Nordemann et al., 2005). The 80–100 yr Gleissberg solar activity cycles have been found in samples from Southern Brazil (Rigozo et al., 2004) and Chile (Rigozo et al., 2006). Nordemann et al. (2005) observed that there was an increase of Osorno tree ring thickness in Chile during the intervals with low sunspot numbers (1800–1835 and 1875–1935) and the contrary during the intervals with high sunspot numbers (1725–1800, 1835–1850, and 1950–1991). This shows the susceptibility of the tree growth rings to long periods of variations in solar activity. The mathematical analysis of ring thickness time series aims to identify the spatial and geophysical phenomena that caused the variations recorded each year during the tree’s life. The method used includes spectral analysis by iterative regression and wavelet analysis. The study described in this work is based on a mathematical analysis of the time series of growth rings in trees sampled in Southern Brazil. Characteristic features in the variations of their thickness, such as periodicities, trends, and events, are examined in order to obtain a greater understanding of effects of solar activity, climatic, and geophysical phenomena on the South America continent.

2. Time series and analysis methods

Sunspot Number

Tree Ring Width (mm)

For the time series used in this study: (1) The tree ring samples of Araucaria angustifolia species were collected in Severiano de Almeida (RS) from Brazil, lat.: 271 250 S–long.: 521 060 W—Altitude: 476 m. (Fig. 1A). (2)- The sunspot time series (Fig. 1B) were obtained from National Geophysical Data Center, Boulder, Colorado¸ http://www. ngdc.noaa.gov/. (3) The Southern Oscillation Index (SOI) (Fig. 1C) was obtained from the website http://www.cru.uea.ac.uk/cru/data/soi. htm. A spectral analysis by an iterative regression method was used to look for the periodicities embedded in tree growth rings. This method is an iterative least square fit and uses a simple sine function with three unknown parameters, a0 ¼amplitude, a1 ¼angular frequency, and a2 ¼ phase (Wolberg, 1967; Rigozo and Nordemann,

1998). The starting point of the method is the so-called conditional function F ¼ Y-a0 sinða1 t þ a2 Þ

ð1Þ

where Y is the observed signal, t is the time and a0, a1, and a2 are the three unknown parameters described above. Every periodicity embedded in the time series corresponds to a set of values of the three parameters, which are determined one at a time by applying the iterative process to the original time series with the limiting condition of maintaining the angular frequency a1 inside a restricted domain of the allowed interval of angular frequencies. The advantage of the method is that it provides the standard deviation for each parameter, as determined from the statistical fluctuations and, if desired, from the errors on every value of the time series. This allows a selection of the periodicities with amplitudes above 95% confidence. The wavelet transform is a powerful tool for non-stationary signal analysis and permits identification of the main periodicities in a time series and their evolution (Kumar and FoufoulaGeorgiou, 1997; Torrence and Compto, 1998; Percival and Walden, 2000). The wavelet transform of a series of discrete data is defined as the convolution between the series and a scaled and translated version of the wavelet function chosen. By varying the wavelet time scale and translating the scaled versions of the wavelet, it is possible to build a graph showing the amplitudes versus frequency (or period) and how they vary with time. In this work, a complex Morlet wavelet was used because it is the most adequate to continuously detect variations of periodicities in geophysical signals. The Morlet wavelet is a plane wave modulated by a Gaussian function (Torrence and Compto, 1998; Percival and Walden, 2000)

CðtÞ ¼ p1=4 expðiotÞexpð0:5 t2 Þ

ð2Þ

X, Y being time series and WXn(s), WYn(s) their wavelet transform, the wavelet cross spectrum is: x y W xy n ðsÞ ¼ W n ðsÞW n ðsÞ

ð3Þ

WYnn(s)

WYn(s).

where is the complex conjugate of The power is 9WXY n (s)9. The wavelet cross power indicates the scale of high covariance between two time series (X, Y).

5 4 3 2 1 0 -1 -2 200

Tree R ing Series

Sunspot

150 100 50 0 2

SO I

SOI

1 0 -1 -2 1640

1680

1720

1760

1800 1840 Time (year)

1880

1920

1960

2000

Fig. 1. (A) The mean width of tree ring time series from Severiano de Almeida. (B) Yearly averages of sunspot number. (C) Southern Oscillation Index—SOI.

A. Prestes et al. / Journal of Atmospheric and Solar-Terrestrial Physics 73 (2011) 1587–1593

3. Results and discussion

Tree Ring Width (mm)

These time series were studied by harmonic spectral (iterative regression) and by wavelet analysis. The tree ring thickness chronology was obtained averaging 12 different tree ring samples, in order to eliminate the noise caused by each tree’s individual variations. Before determining the mean chronology of the location, a mathematical function that represents its individual growth trend was subtracted from every sample tree series. Then a mean time series was calculated for each sampled location, as shown in Fig. 2 for the tree rings collected at Severiano de Almeida. Fig. 3 presents the amplitude spectrum obtained by iterative regression spectral analysis. Periodicities at 72.5, 48.9, 23.9, and 11 yr can be seen and may represent a possible influence of solar activity and the Gleissberg cycle (80 yr), fourth harmonic of the 200 yr Suess cycle (  50 yr), Hale cycle (22 yr), and Schwabe cycle (11 yr), respectively. The low period cycles from 2 to 7 yr possibly ˜ o events (Gray et al., represent local climatic factors and/or El Nin 1992). Other two interesting periodicities found are at the 18.5 yr period, which several authors have associated to lunar tide influenced Saros cycle (O’Brien and Currie, 1993), and at 33.3 yr ¨ ¨ period that may be due to Bruckner cycle (Bruckner, 1890; Sazonov, 1979). Roig et al. (2001), studying chronological series of fossil trees from  50,000 years ago as well as living trees from Chile, found similar spectral properties indicating that similar factors have affected the tree’s radial growth since the end of the Pleistocene. The similar spectral properties in the two series were the periods between: 136–153, 81–94, 47–53, 35, 24, 17.8, 11.8, 6.6, 5.1, 4.58, 4.3, 3.7, 3.2, and 2.77 years. They attributed the periods of 81–94, 24, 11.8 years to the solar modulation. There are cycles that can appear due to the beating of two different periods. Some cycles of 7 and 8 years can appear due to the beating of others two cycles (Hoyt and Schatten, 1997). Kurths et al. (1993) showed that the combinations of two periods

1589

could generate another two ((1/12.6)+(1/17.1)¼1/7.3 and (1/12.6) –(1/17.1)¼1/47.9). Raspopov et al. (2000 and 2001) interpreted ¨ the Bruckner cycle as a resultant of the non-linear effect of the solar activity in the terrestrial environment, which is a result of the combinatorial frequency of the Gleissberg cycle (T1 ¼1/n1 ¼  90 years) and of the Hale cycle (T2 ¼1/n2 ¼22 years) (T_¼1/n ¼ 1/n1–1/n2 ¼ 30; T+ ¼1/n ¼1/n1 +1/n2 ¼ 17). Thus, considering that the length of the solar cycle varies, the periods in the  17–18 and  30–35 intervals may be caused by the non-linear effect of the solar activity in atmospheric processes. The climatic variability of  18 years may be a combination of the lunar tides periodicities in the climate, cycle of Saros, along with the solar activity. Another consideration in the ¨ Bruckner cycle is the presence of two frequency bands with periods centered in 30 and 45 years discovered by Sazonov (1979), who emphasized that at times of high solar activity shorter periods occurred (T¼  30 years), and at times of low solar activity longer periods occurred (T¼  45 years). Raspopov et al. (2000 and 2001) considered the period of  45 years to be the second harmonic of the 90 year cycle. Prestes (2009), studying chronological series from Passo Fundo (RS), found some periods to be combinations of two other periods found in the series: ((1/11)+(1/16.2)ffi1/6.6 and (1/11)–(1/16.2)ffi 1/35) or ((1/23)+(1/73.1)ffi1/17.5 and (1/23)–(1/73.1)ffi 1/33.5), moreover, the beating between periods of 11 and 16.2 years, revealed the periods of 6.6 and 35 years, in addition, the periods of 23 and 73.1 revealed the periods of 17.5 and 33.5 years, respectively. Fig. 4 shows the amplitude spectrum for tree ring thickness time series, sunspot number, and Southern Oscillation Index—SOI, at a confidence level of 95%. Tree ring spectra periodicities were observed around the 11, 22, 50, and 80 year solar cycles; as well ˜o events. as periodicities around 5 yr associated to El Nin Rigozo et al. (2008), studying tree rings from Passo Fundo (RS) through spectral analysis, wavelets, and cross-wavelets, found a strong tree growth ring response to the increase of solar activity, mainly at the solar activity maximum. The relation between solar

T1

–4 0 –4 8 0 3 0 –3 3 0 –3 3 0 –3 2 0 –2 3 0 6 3 3 0

T2

T3

T4

T5

N1

N2

N3

3 0 4 2 0 –2 3 0

N4

N5

Mean

1650

1680

1710

1740

1770

1800

1830

1860

1890

1920

1950

1980

2010

Time (year) Fig. 2. Mean time series of growth ring thickness of trees from Severiano de Almeida – Rio Grande do Sul, after removing their long trend curves.

1590

A. Prestes et al. / Journal of Atmospheric and Solar-Terrestrial Physics 73 (2011) 1587–1593

Suess 4 harmonic Hale Cycle Gleissberg Bruckner Saros Cycle Schwabe Cycle Suess Cycle 0.4

El-Niño Events

217.4

T1

0.2 0.0 0.6 184.2

T2

0.3 0.0 0.6

T3

0.3 0.0 0.8

T4

0.4 0.0 0.8

T5

Amplitude

0.4 0.0 0.4

N1

0.2 0.0 0.4

N2

0.2 0.0 0.4

N3

0.2 0.0 0.4

129

N4

0.2 0.0 0.4

N5

0.2 0.0 0.4

72.5 48.9

0.2

12.8 14.5 13.8 11 23.9 15.3 9.3 9 12.2

5.7

7.7 7.5

5.1

SA Mean

5

0.0 0.00

0.05

0.10

0.15

0.20

0.25

0.30

-1

Frequency (year ) Fig. 3. Tree ring thickness amplitude spectrum above 95% confidence.

0.4

72.5 14.5

(A)Tree Ring

12.8

48.9

0.3

13.8 12.2 23.9 15.3

0.2

9 5

11

0.1

7.5

7.7

9.3

5.7 5.1

0.0 35

(B) Sunspot Number 11

225.4

Amplitude

28 21

8.8

10

104.2

8.5 12

14 7

54.7

28.5

8.1 9.6

43.5

0 3.0

(C) SOI

2.0

3.5

6.4

2.5

13.6

5.7

9 7.4

45.9

3.4

4.8 5.2

3.6

4.2

2.9

1.5 1.0 0.5 0.0 0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Frequency (years-1) Fig. 4. (A) Amplitude spectra for the mean width of tree rings, (B) sunspot number, and (C) Southern Oscillation Index—SOI.

A. Prestes et al. / Journal of Atmospheric and Solar-Terrestrial Physics 73 (2011) 1587–1593

activity and tree growth is more evident in Fig. 5 for the 8–13 yr band . The solar signal is in the opposite phase as the tree rings after 1935. Before 1935, the solar signal is in phase with the tree rings, with only a small variation. This could indicate the intensification of climatic and/or anthropogenic factors. Rao and Hada (1990) showed that during the negative phase in ˜ o) an excess of precipitation generally occurs in SOI (El Nin Southern Brazil, Northern Argentina, and Uruguay, and in the ˜ a) there is a deficiency of precipitapositive phase in SOI (La Nin tion. Fig. 6(A) shows the mean width of the tree ring time series and Southern Oscilation Index (SOI) filtered between 2 and 7 yr,

while Fig. 6B shows the same signals filtered from 13 to 14 yr. ˜ o the trees grew more than during La Nin ˜ a, showing During El Nin the strong response of the trees to precipitation. The period of 13.6 yr is clearly seen in the tree ring time series, indicating the ˜ o in this trees’ response to the strong influence of the El Nin region. The evidence of the solar activity in the last millennium, for example of 14C and 10B, indicates that the sun had periods of lower and higher activity than has been observed in last the two centuries, such as the Maunder minimum from 1645 to 1745 and great maximum from 1100 to 1300. These upheavals in solar

100

Tree Ring Series Pass Band Filter 8-13 yr ------ Sunspot Number Pass Band Filter 8-13 yr

0.8

80

0.6

60

0.4

40

0.2

20

0.0

0

-0.2

-20

-0.4

-40

Sunspot

Amplitude (Tree Ring Data)

1591

-60

-0.6

-80

-0.8 1700

1750

1800

1850

1900

1950

-100 2050

2000

Time (year) Fig. 5. Signals filtered of the mean width of tree ring time series and sunspot number between 8 and 13 yr.

Positive phase, occurrence La Niña Niña, less precipitation.

4

Negative phase, occurrence of the El Niño, more precipitation.

1.5

SOI Tree Ring Series Pass Band Filter 2-7 yr

3

A

1.0 0.5

SOI

1 0

0.0

-1

-0.5

Tree Ring

2

-2 -1.0

-3

-1.5

-4 1860

1880

1900

1920

1940

1960

1980

2000

0.05 0.04 0.03 0.02 0.01 0.00 -0.01 -0.02 -0.03 -0.04 -0.05

0.15

Tree Ring Series Pass Band Filter 13-14 yr ------- SOI Band Pass Filter 13 - 14 yr

B

0.10 0.05 0.00

SOI

Amplitude (Tree Ring Data)

Time (year)

-0.05 -0.10 1860

1880

1900

1920

1940

1960

1980

2000

-0.15 2020

Time (year) Fig. 6. (A) Signals filtered of the mean width of tree ring time series and Southern Oscillation Index—SOI between 2 and 7 yr; (B) Signals filtered between 13 and 14 yr.

1592

A. Prestes et al. / Journal of Atmospheric and Solar-Terrestrial Physics 73 (2011) 1587–1593

behavior may have been accompanied by significant long-term changes in radiative output and flow of atomic particles from the sun, with a possible terrestrial effect (Eddy, 1976; Rigozo et al., 2001).

The wavelet cross spectrum of sunspot number and tree growth ring (Fig. 7) clearly illustrates a relationship between both series. Good agreement between the time series of tree growth rings and the 11 yr solar cycle was found during the

Fig. 7. Power cross spectrum between sunspot numbers and tree growth rings, with confidence cone (white curve) and confidence level of 95% (white contour).

Fig. 8. Power cross spectrum between SOI and tree growth rings, with confidence cone (white curve) and confidence level of 95% (white contour).

A. Prestes et al. / Journal of Atmospheric and Solar-Terrestrial Physics 73 (2011) 1587–1593

maximum solar activity periods, for the intervals from 1704 to 1790 and 1940 to 1990. In addition, the Gleissberg solar cycle was observed between 1830 and 1930. Fig. 8 shows the relationship between the Southern Oscillation Index (SOI) and tree growth ring series obtained by wavelet cross spectrum. During extreme events in the SOI, the meteorological conditions are significantly modified in diverse regions of the ˜ o events, as occurred in 1982/83, caused Earth. Intense El-Nin great social, economic, and ecological impacts around of the planet. This event caused intense rainfalls in southern Brazil. The wavelet cross spectrum shows the trees responded strongly to excess of precipitation in 1982/83.

4. Conclusion Sun–Earth relationship studies based on tree growth rings from Brazil showed interesting results. The main result is the observation in tree samples of periodicities associated to solar activity: Gleissberg cycle, fourth harmonic of the Suess cycle, Hale cycle, and Schwabe cycle. The 11 yr cycle showed that Severiano de Almeida tree ring growth was facilitated by an increase of solar activity. These results are consistent with Rigozo et al. (2004); however, in this work two new periodicities were observed for the Southern region of Brazil, 18.5 and 33 yr, representing the ¨ lunar cycle and the Bruckner cycle, respectively. Short periods observed in tree growth ring series are due to ˜ o and La-Nin ˜ a events. local environmental conditions of El-Nin These exert strong influence on the climate of southern Brazil, ˜ o events in mainly on the precipitation. The influence of El Nin ˜ o phenomenon brings higher tree rings showed that the El-Nin precipitation in this region causing more tree growth, whereas ˜ a phenomenon occurs lower precipitation causes when the La-Nin less tree growth, as can be seen in Fig. 6.

Acknowledgements The authors wish to thanks CNPq, CAPES, and FAPESP for supporting this research: A. Prestes FAPESP—(2009/02907-8); N. R. Rigozo—CNPq (APQ 470252/2009-0, APQ 470455/2010-1 and research productivity, 301033/2009-9), and FAPERGS (1013273); C.M. Wrasse thanks to CNPq for the grant (304277/2008-8); M.P Souza Echer—CNPq (151609/2009-8); CNPQ/PQ (300211/ 2008-2) and FAPESP (2007/52533-1); P. H. Rampelotto from CNPq. References ¨ Bruckner, E., 1890. Klimaschwankungen seit 1700. Geographische Abhandlungen 14, 325. Eddy, J.A., 1976. The Maunder minimum. Science 192, 1189–1192. Eddy, J., 1980. The historical record of solar activity. The Ancient Sun, 119–134. Fritts, H.C., 1976. Tree Ring and Climate. Academic Press Inc., London. Gray, W.M., Sheaffer, J.D., Knaff, J.A., 1992. Hypothesized mechanism for stratospheric QBO influence on ENSO variability. Geophysical Research Letters 19, 107–110.

1593

Hoyt, D.V.E., Schatten, K.H., 1997. The Role of Sun in Climate Change. Oxford: University Press, p. 279. Hughes, M.K., Kelly, P.M., Pilcher, J.R., LaMarche, V.C., 1982. Climate from Tree Ring. Cambridge University Press, Cambridge. Kumar, P., Foufoula-Georgiou, E., 1997. Wavelet analysis for geophysical applications. Reviews of Geophysics 35, 385–412. ¨ Kurths, J., Spiering, C., Muller-Stoll, W., Striegler, U., 1993. Search for solar periodicities in Miocene tree ring widths. Terra Nova 5, 359–363. Lean, J., Skumanich, A., White, O., 1992. Estimating the Sun’s radiative output during the Maunder Minimum. Geophysical Research Letters 19, 1595. National Research Council, 1994. Solar Influences on Global Change. National Academy Press, Washington, D. C. ˜ o dynamics. Physics Today 51, 32–36. Neelin, J.D., Latif, M., 1998. El Nin ˜ o signals Nordemann, D.J.R., Rigozo, N.R., Faria, H.H., 2005. Solar activity and El-Nin observed in Brazil and Chile tree ring records. Advances in Space Research 35, 891–896. O’Brien, D.P., Currie, R., 1993. Observation of the 18.6-year cycle of air pressure and a theoretical model to explain certain aspects of this signal. Climate Dynamics 6, 287–298. Percival, D.B., Walden, A.T., 2000. Wavelet Methods for Time Series Analysis. Cambridge University Press. Prestes, A., 2009. Relac- a~ o sol-terra estudada atrave´s de ane´is de crescimento de conı´feras do holoceno recente e do tria´ssico/Alan Prestes. Sa~ o Jose´ dos Campos: INPE, p.148: (INPE-15756-TDI/1499). Raisbeck, G.M., Yiou, F., 1984. Production of long-lived cosmogenic nuclei and their applications. Nuclear Instruments and Methods in Physics Research B5 (2), 91–99. Rao, V.B., Hada, K., 1990. Characteristics of rainfall over Brazil: annual variations and connections with the southern oscillations. Theoretical and Applied Climatology 42, 81–91. Raspopov, O.M., Shumilov, O.I., Kasatkina, E.A., Turunen, E., Lindholm, M., 2000. 35-Year climatic Bruckner cycle—solar control of climate variability? In: Proceedings of the Solar & Space Weather Euroconference, the Solar Cycle and Terrestrial Climate, 2000, Santa Cruz de Tenerife. ¨ Raspopov, O.M., Dergachev,V.A., Shumilov, O.I., Kolstrom, T., Lindholm, M., ´ ., Vasiliev, S.S., Kuzmin, A.V., Yu Kirtsidely, I., ¨ Merilainen, J., Eggertsson, O Kasatkina, E.A., 2001. Dendrochonological evidence of long-term variations in ¨ solar activity and climate. In: Kaennel Dobbertin M., Braker O.U. (Eds.), Proceedings of the International Conference Tree Rings and People, Davos. Rigozo, N.R., Nordemann, D.J.R., 1998. Ana´lise por regressa~ o iterativa de periodicidades em se´ries temporais de registros geofı´sicos. Revista Brasileira de Geofı´sica 16 (2–3), 149–157. Rigozo, N.R., Echer, E., Vieira, L.E.A., Nordemann, D.J.R., 2001. Reconstruction of wolf sunspot numbers on the basis of spectral characteristics and estimates of associated radio flux and solar wind parameters for the last millennium. Solar Physics, Netherlands 203 (1), 179–191. Rigozo, N.R., Nordemann, D.J.R., Echer, E., Vieira, L.E.A., 2004. Search for solar periodicities in tree-ring widths from Conco´rdia (S.C., Brazil). Pure Applied Geophysics 161, 221–233. Rigozo, N.R., Nordemann, D.J.R., Echer, E., Zanandrea, A., Gonzalez, W.D., 2002. Solar variability effects studied by tree-ring data wavelet analysis. Advances in Space Research 29, 1985–1988. Rigozo, N.R., Nordemann, D.J.R., Souza Echer, M.P., Echer, E., Silva, H.E., Prestes, A., 2006. Solar and climate imprints in tree ring width time series from Chile (1587–1994 A.D.). Trend Applied Science Research 1, 73–78. Rigozo, N.R., Prestes, A., Nordemann, D.J.R., da Silva, H.E., Souza Echer, M.P., Echer, E., 2008. Solar maximum epoch imprints in tree-ring width from Passo Fundo, Brazil (1741–2004). Journal of Atmospheric and Solar-Terrestrial Physics 70, 1025–1033. Roig, F.A., Le-Quesne, C., Boninsegna, J.A., Briffa, K.R., Lara, A., Grudd, H., Jones, P.D., Villagra´n, C., 2001. Climate variability 50,000 years ago in mid-latitude Chile as reconstructed from tree rings. Nature 410, 567–570. ¨ Sazonov, B.I., 1979. Bruckner cycle of droughts, Criterions and characteristics of droughts phenomena in USSR. Hydrometeoizdat, Leningrad, pp. 82–92. Torrence, C., Compto, G.P., 1998. A practical guide to wavelet analysis. Bulletin of the American Meteorological Society 79, 61–78. Wilson, R.C., Hudson, H.S., 1988. Sun luminosity variations in solar cycle 21. Nature 332, 810–812. Wolberg, J.R., 1967. Prediction Analysis. D. Van Nostrand.