• Berumen Barbosa, M., 2003: Geografía económica de Oaxaca. 91 pp. [Available online at http://www.eumed.net/cursecon/libreria/mebb/index.htm.]

  • Clavijero, F., 1852: Historia de la antigua o Baja California. Juan R. Navarro, 258 pp.

  • Englehart, P. J., , and A. V. Douglas, 2001: The role of eastern North Pacific tropical storms in the rainfall climatology of western Mexico. Int. J. Climatol., 21, 13571370.

    • Search Google Scholar
    • Export Citation
  • Englehart, P. J., , M. Lewis, , and A. Douglas, 2008: Defining the frequency of near-shore tropical cyclone activity in the eastern North Pacific from historical surface observations (1921–2005). Geophys. Res. Lett., 35, L03706, doi:10.1029/2007GL032546.

    • Search Google Scholar
    • Export Citation
  • Escobar, A., 2004: Siglo XIX (1822–1900). Vol. 2, Desastres Agrícolas en México. CIESAS, 280 pp.

  • Fuentes, O., , and M. Vázquez, 1997: Probabilidad de presentación de ciclones tropicales en México. Cuaderno de Investigación, Vol. 42, CENAPRED, 138 pp.

  • Gallego, D., , R. García-Herrera, , N. Calvo, , and P. Ribera, 2007: A new meteorological record for Cádiz (Spain) 1806–1852: Implications for climatic reconstructions. J. Geophys. Res., 112, D12108, doi:10.1029/2007JD008517.

    • Search Google Scholar
    • Export Citation
  • García-Acosta, V., , J. M. Pérez-Zevallos, , and A. M. del Villar, 2003: Catálogo Histórico, Vol. 1, Desastres Agrícolas en México, CIESAS, 506 pp.

  • García-Herrera, R., , L. Gimeno, , P. Ribera, , and E. Hernández, 2005: New records of Atlantic hurricanes from Spanish documentary sources. J. Geophys. Res., 110, D03109, doi:10.1029/2004JD005272.

    • Search Google Scholar
    • Export Citation
  • García-Herrera, R., , L. Gimeno, , P. Ribera, , E. Hernández, , E. González, , and G. Fernández, 2007: Identification of Caribbean basin hurricanes from Spanish documentary sources. Climatic Change, 83, 5585, doi:10.1007/s10584-006-9124-4.

    • Search Google Scholar
    • Export Citation
  • García-Oliva, F., , E. Excurra, , and L. Galicia, 1991: Pattern of rainfall distribution in the central Pacific coast of Mexico. Geogr. Ann., 73, 179186.

    • Search Google Scholar
    • Export Citation
  • Gilman, D., , E. Fuglister, , and J. Mitchell, 1963: On the power spectrum of “red noise.” J. Atmos. Sci., 20, 182184.

  • Jauregui, E., 2003: Climatology of landfalling hurricanes and tropical storms in Mexico. Atmósfera, 4, 193204.

  • Latorre, C., , and L. Penilla, 1988: Influencia de los ciclones en la precipitation de Baja California Sur. Atmósfera, 1, 99112.

  • Ludlum, D., 1963: Early American Hurricanes 1492–1870. American Meteorological Society, 198 pp.

  • Mendoza, B., , and M. Pazos, 2009: A 22 yr hurricane cycle and its relation with geomagnetic activity. J. Atmos. Solar-Terr. Phys., 71, 20472054.

    • Search Google Scholar
    • Export Citation
  • Salinas-Zavala, C., , A. Leyva-Contreras, , D. Lluch-Belda, , and E. Díaz-Rivera, 1990: Distribución geográfica y variabilidad climática de los regimenes pluviométricos en Baja California Sur, México. Atmósfera, 3, 217237.

    • Search Google Scholar
    • Export Citation
  • Serra, S., 1971: Hurricanes and tropical storms of the west coast of Mexico. Mon. Wea. Rev., 99, 208302.

  • Sumner, H., 1944: North Atlantic hurricanes and tropical disturbances of 1943. Mon. Wea. Rev., 71, 179183.

  • Torrence, C., , and G. P. Compo, 1998: A practical guide to wavelet analysis. Bull. Amer. Meteor. Soc., 79, 6178.

  • Velasco, V., , and B. Mendoza, 2008: Assessing the relationship between solar activity and some large scale climatic phenomena. Adv. Space Res., 42, 866878.

    • Search Google Scholar
    • Export Citation
  • Villanueva González, E., 2002: Presencia de huracanes en Baja California Sur: El caso del ciclón Liza. Master’s thesis, Universidad Autonoma de Baja California Sur Social Sciences and Humanities Department, 227 pp.

  • View in gallery

    Percentage of coincidence between TC presence and positive SPI values for the following states: Baja California Sur (BCS), Colima (COL), Chiapas (CHI), Guerrero (GRO), Jalisco (JAL), Nayarit (Nay), Oaxaca (OAX), Sinaloa (SIN), and Sonora (SON).

  • View in gallery

    Location of the records of historical data by century. Circles represent data from the sixteenth and seventeenth centuries; asterisks represent data from the eighteenth century and squares from the nineteenth century. The lack of information is noticeable between sixteenth and eighteenth centuries. The labels correspond to the same description as in Fig. 1.

  • View in gallery

    Time series of LFTCs between 1536 and 1948.

  • View in gallery

    Spectral analysis for the 1949–2010 time series. (bottom left) the wavelet power spectrum in grayscale with the COI (the border between the black and gray backgrounds). (top) The time series and (right) the global spectrum with the red noise level (dotted curve).

  • View in gallery

    As in Fig. 4, but for the (a) 1701–2010 and (b) 1850–2010 time series.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 71 71 4
PDF Downloads 34 34 2

Landfalling Tropical Cyclones along the Eastern Pacific Coast between the Sixteenth and Twentieth Centuries

View More View Less
  • 1 Graduate Program in Earth Sciences, Universidad Nacional Autónoma de México, Mexico City, Mexico
  • | 2 Instituto de Geofísica, Universidad Nacional Autónoma de México, Mexico City, Mexico
© Get Permissions
Full access

Abstract

Numerous studies have been conducted to document long-term trends in tropical cyclone (TC) activity. However, the eastern Pacific has not received as much attention as other basins. Here the authors attempt the identification of TC formation in the Mexican eastern Pacific Ocean before 1950. Using bibliographical and historical file consultation, they constructed a catalog of events related to intense storms and possible TCs that made landfall on the Mexican Pacific coasts. Between 1536 and 1948 they found a total of 119 events related to TCs. Then, using the Saffir–Simpson scale and the climatology of the region as the criteria to evaluate each event, they found 85 TCs. Furthermore, they constructed a historical time series of TCs between 1701 and 2010. The spectral analysis showed periodicities of ~2.6, 4, 5, 12, 16, 39, and 105 years that coincide with some large-scale climatic phenomena and also with solar activity. In particular, the ~12-yr cycle is the most persistent periodicity in this study.

Corresponding author address: Marni Pazos, Departamento de Ciencias Espaciales, Instituto de Geofísica, Universidad Nacional Autónoma de México, Mexico City 04510, Mexico. E-mail: marni@geofisica.unam.mx

Abstract

Numerous studies have been conducted to document long-term trends in tropical cyclone (TC) activity. However, the eastern Pacific has not received as much attention as other basins. Here the authors attempt the identification of TC formation in the Mexican eastern Pacific Ocean before 1950. Using bibliographical and historical file consultation, they constructed a catalog of events related to intense storms and possible TCs that made landfall on the Mexican Pacific coasts. Between 1536 and 1948 they found a total of 119 events related to TCs. Then, using the Saffir–Simpson scale and the climatology of the region as the criteria to evaluate each event, they found 85 TCs. Furthermore, they constructed a historical time series of TCs between 1701 and 2010. The spectral analysis showed periodicities of ~2.6, 4, 5, 12, 16, 39, and 105 years that coincide with some large-scale climatic phenomena and also with solar activity. In particular, the ~12-yr cycle is the most persistent periodicity in this study.

Corresponding author address: Marni Pazos, Departamento de Ciencias Espaciales, Instituto de Geofísica, Universidad Nacional Autónoma de México, Mexico City 04510, Mexico. E-mail: marni@geofisica.unam.mx

1. Introduction

Many studies on the number of historical tropical cyclones (TCs) have been generated in the Atlantic Ocean, for which data are obtainable since the fifteenth century (e.g., García-Herrera et al. 2005, 2007; Ludlum 1963). However, for the case of the eastern Pacific Ocean, the earliest and most reliable records start in 1949 from the Atlantic basin hurricane database (HURDAT) of the National Oceanic and Atmospheric Administration (NOAA).

Jauregui (2003) showed that, between 1949 and 2003, 65 hurricane-strength TCs made landfall on the Mexican Pacific coast (MPc), while only 27 TCs made landfall on Mexico’s Gulf of Mexico coast. This study also indicates the states that are more often hit by TCs along the MPc: Baja California Sur, Sinaloa, and Michoacán. This agrees with the data published in 1997 by the Mexican National Disaster Prevention Center [Centro Nacional de Prevención de Desastres (CENAPRED); see http://www.cenapred.gob.mx/es/Investigacion/RHidrometeorologicos/FenomenosMeteorologicos/CiclonesTropicales/].

We use the Unisys Weather database of TCs in the northeast Pacific Ocean between the years 1949 and 2010 concerning landfalling tropical cyclones (LFTCs) in the basin. From the storm tracks displayed in this database we found 447 TCs with an origin in the northeast Pacific Ocean. Of these, 95 reached the MPc, which means that the influence of the heavy rains and strong winds affected the area. Many of them affected more than one state in their path. In Table 1 we show the number and percentage of LFTCs in each state ordered from its geographical location from north to south of the MPc. It is clear that the states of Baja California Sur, Sinaloa, and Michoacán have the highest percentage of LFTCs, in agreement with the study of Jauregui (2003). The difference between the results in Jauregui (2003) and Table 1 is that Jauregui made the study only with hurricanes and this study includes all TCs.

Table 1.

LFTCs in the MPc years 1949–2010.

Table 1.

Even though the eastern Pacific is a very active region of tropical cyclogenesis, it has not received as much attention as other basins. We see the need to extend the LFTC time series in the MPc to provide a better understanding of the TCs that formed in the eastern Pacific, as well as the variability this data might show through spectral analysis.

We examined historical files from several states of the MPc, which are frequently hit by these phenomena, which allowed us to construct a time series that was analyzed to find the periodicities through wavelet analysis.

2. Methodology

Some reconstructions of past meteorological phenomena are carried out from historical instrumental records of weather stations, as well as testimonials of newspapers and government or missionary travel reports, as done before in several works (García-Herrera et al. 2005, 2007; Gallego et al. 2007; Villanueva-González 2002).

Since historical research often lacks criteria by which to establish the wind velocity of TCs, in this study we are considering TCs as low pressure systems including tropical depressions, tropical storms, and hurricanes.

To decide whether the damages described in the historical archives and catalogs were probably caused by a TC, we identified them through the description of the event using the following criteria. 1) As a reference of the damages caused by a TC, we considered those according to the Saffir–Simpson scale (SSS). In this scale, a category 1 hurricane is able to crack trees or pull them from their roots, destroy fragile buildings and blow off roofs that are not well attached, and cause flooding and damage to ships. 2) We took into account the climate of the location where the event was reported. 3) We compared the precipitation and the TC occurrence in such locations.

The yearly storm track charts of the Mexican Meteorological Service [Servicio Meteorológico Nacional (SMN)] provide another form of TC identification in the Atlantic and Pacific Ocean since 1921. These charts were very useful in order to identify the tracks and places where these events made landfall. This dataset was also used by Serra (1971), Englehart and Douglas (2001), and Englehart et al. (2008) to study the importance of TCs in the MPc.

Precipitation–tropical cyclone relationship

Englehart and Douglas (2001) showed that tropical storms normally constitute 20%–60% of rainfall along the MPc and, in more extreme cases, can contribute as much as 25%–30% to seasonal rainfall totals experienced in western interior locations.

The Standardized Precipitation Index (SPI) is an index based on the probability of recording a given amount of precipitation. The probabilities are standardized; therefore, an index of zero indicates the median precipitation amount. The index is negative for drought and positive for wet conditions. As the dry or wet conditions become more severe, the index becomes either more negative or positive.

We used the SPI to find a relationship between TCs whose path is near the shore and the precipitation amount during the hurricane season (May–December) from 1949 to 2010, using data from the North American Drought Monitor Standard, the Precipitation Index Data Files, NOAA, and the National Climatic Data Center, of stations with similar location as the historical records. The hypothesis is that for every year in the MPc, a positive SPI corresponds to the presence of at least one TC, and its absence with a negative SPI. Figure 1 shows the percentage of coincidence between TCs and SPI per state. The lowest coincidences were in Chiapas (~27%) and Oaxaca (~45%), where the rainy season, mostly due to the monsoon, is less influenced by the TC activity than for the Mexican northwestern states. Also, the number of TCs in this area is sparse (see Table 1). Thus, these results agree with studies related to precipitation and the geographical zone covered in this work (Latorre and Penilla 1988; Salinas-Zavala et al. 1990; García-Oliva et al. 1991).

Fig. 1.
Fig. 1.

Percentage of coincidence between TC presence and positive SPI values for the following states: Baja California Sur (BCS), Colima (COL), Chiapas (CHI), Guerrero (GRO), Jalisco (JAL), Nayarit (Nay), Oaxaca (OAX), Sinaloa (SIN), and Sonora (SON).

Citation: Journal of Climate 26, 12; 10.1175/JCLI-D-11-00411.1

3. Data

Figure 2 shows the locations of the events reported in the historical files that we examine. The results are summarized in the following sections. All the event descriptions have been translated from Spanish.

Fig. 2.
Fig. 2.

Location of the records of historical data by century. Circles represent data from the sixteenth and seventeenth centuries; asterisks represent data from the eighteenth century and squares from the nineteenth century. The lack of information is noticeable between sixteenth and eighteenth centuries. The labels correspond to the same description as in Fig. 1.

Citation: Journal of Climate 26, 12; 10.1175/JCLI-D-11-00411.1

a. Catalog of Agricultural Disasters in Mexico

The historical Catalog of Agricultural Disasters in Mexico [CADM; volume I by García-Acosta et al. (2003) and volume II by Escobar (2004)] describes damages caused not only by meteorological events, but also by earthquakes and other natural phenomena throughout the whole country between the years 958 and 1900.

Here, only data from the western coast were taken into account. In some cases, the date was missing, but from the description of damages it was possible to infer if the event could be a TC by comparing with the characteristics of damages made by a category 1 hurricane as given by the SSS. We found information concerning 38 intense storms or potential TCs that affected the MPc from 1537 to 1900. Only 29 could be considered a TC and these appear in Table 2. For three of the other events (1537, 1732, and 1787) there were earthquakes at the same time. Also in this catalog, for some events that are described as TCs, the damages were caused by very strong winds but no rain presence and they occurred in central zones of Mexico; this indicates that in those times “hurricane” was used not only for events related to strong rains and winds, giving an ambiguous sense to the word. For that reason, we considered only those events that have descriptive damages rather than all those with which the term hurricane is associated.

Table 2.

Meteorological events possibly related to LFTCs in the MPc between the years 1609 and 1900 [from García-Acosta et al. (2003) and Escobar (2004)].

Table 2.

b. Baja California Sur

Baja California Sur is a state located at the northwest of Mexico in the Baja California peninsula, between 28° and 22° 52′N. It has 2705.39 km of coastlines according to the National Institute of Statistics, Geography and Informatics [Instituto Nacional de Estadística y Geografía (INEGI)]. According to Table 1, it has the highest percentage of LFTCs in the MPc (Fuentes and Vázquez 1997; Jauregui 2003).

In this state, because of the influence of the high pressure center from the North Pacific, the prevailing climate is dry and warm and with low levels of precipitation (179 mm yr−1). However, in summer and autumn the southeast part of the peninsula is influenced by the cyclonic activity of the tropical Pacific Ocean, with up to 69% of average contribution to the annual precipitation (Latorre and Penilla 1988; Salinas-Zavala et al. 1990).

The Pablo L. Martínez Historical Archive in the city of La Paz has historical files dating back to the second half of the nineteenth century. Several sources were consulted, and data from 1882 to 1918 were found. Official bulletins, newspapers, and government reports mentioned events related to TCs and damage caused by heavy rains and strong winds. From rainfall charts, we obtained information about the strongest and longest rainfalls within the TC season period and those that affected several locations, which could indicate the track of the storm. Another source used was the master’s degree thesis of Villanueva-González (2002), containing TC data from 1827 to 1948. Several cases do not have a description of damages; these were confirmed by the yearly storm track charts of the SMN. The book History of Ancient or Baja California by Clavijero (1852) provides information about Spanish travels during the conquest period and the missionary journeys of the Jesuit Order in Baja California. These chronicles tell about sea storms and the damages incurred, mainly by vessels, that probably were caused by a TC. This book contained information between the years 1702 and 1767. For the events that occurred in 1717 and 1723, the described damages coincide with those caused by a hurricane according to the SSS.

Table 3 shows the historical event information that affected Baja California Sur. The total number of registered events found in the consulted sources is 32. Some of them coincide with data found in the CADM, such as the events in the years 1882, 1895, and 1897. Of these 32 events, 28 could be considered to be TCs. Moreover, the data were compared with the tracks provided by the SMN. Events from 1927 to 1948 are listed with no description, except for 1939 and 1941, but for each of those years there is at least one storm track affecting the zone.

Table 3.

Meteorological events possibly related to LFTCs in Baja California Sur during the years 1536–1941 (Clavijero 1852; Villanueva-González 2002; Pablo L. Martinez Historical Archive).

Table 3.

c. Jalisco

Jalisco is located at the central western side of the country, from 22°45′ to 18°55′N, and it has 341.93 km of coastline in the Pacific Ocean (INEGI). The climate in the Jalisco coast is subhumid and warm and is present in 25% of the state’s surface. At the east side of the coast, the semiwarm climate prevails in 42% of the surface. Both types of climate are characterized by rains and thunderstorms in July and August. The rainfall pattern in this region is affected mainly by the influence of trade winds and TCs (García-Oliva et al. 1991).

At the historical files in the city of Guadalajara, we consulted a division named “Foment-Natural Disasters” covering the period 1860–1978, with missing files between 1914 and 1924. We found events related to TC presence such as heavy rains and floods and their consequent damages. We chose seven events that might be related to a TC; these are listed in Table 4. The event registered in August 1932 was compared to the SMN chart for that year and there is a storm track that affected Jalisco in the same time period. Because of the semiarid climate of the zone, rain is not abundant (850 mm annual average). Therefore, if a rainy event is reported we assume that a TC could be present. We have also considered those events that occurred on consecutive dates as the same event.

Table 4.

Meteorological events possibly related to LFTCs in Jalisco between the years 1865–1932 (Historical Archive of Jalisco).

Table 4.

d. Oaxaca

Oaxaca is located in the south of Mexico, between 18°42′ and 15°39′N and its coastlines are 598 km in length along the eastern Pacific Ocean (Berumen Barbosa 2003).

The climate in the coastal zone is subhumid and warm with summer rains, and the annual precipitation ranges from 800 to 2000 mm. At the northwest of the state, the climate is humid and warm with abundant rain in summer and annual precipitation between 1500 and 3000 mm. The eastern side is characterized also by abundant rains in summer, and there is a band at the central north with rains the whole year.

In the General Archive of the state of Oaxaca, we consulted documents from 1900 to 1949 searching for reports of TCs on the coast of Oaxaca in government reports, the official newspaper of the state, and documents from the Foment and Government division. Of eight events found, four could be considered a TC (see Table 5). Damages reported in the other four events have scarce information that prevents us from relating them with a TC. Events for 1926 and 1935 were compared with the SMN charts of those years, and in both cases there is a storm track affecting Oaxaca according to the registered dates.

Table 5.

Meteorological events possibly related to LFTCs in Oaxaca during years 1902–35 (Historical Archive of Oaxaca).

Table 5.

e. Chiapas

Chiapas is at the southeast of Mexico, with a coastline of 255.69 km (INEGI). It is located between 17°59′ and 14°32′N. From the coastal zone of Chiapas to the northeast side of the state, a subhumid climate with summer rains prevails, with annual precipitation ~2000 mm. Northward of this zone rains are more abundant (2000–3000 mm yr−1). The northern part of the state is the state’s rainiest area (3000–4500 mm yr−1). Rains in this area are present the whole year, due to the moist winds from the Gulf of Mexico.

The Archive of Chiapas is kept by the University of Arts and Sciences of Chiapas. Documents between 1900 and 1949 were consulted for references to possible TCs in government reports, bulletins, newspapers, the state’s official newspaper, and documents from the Foment and Government area.

Six events were found and three of them were identified as TCs (see Table 6). There were no SMN storm track charts related to the events recorded in this state.

Table 6.

Meteorological events possibly related to LFTCs in Chiapas between the years 1906 and 1923 (Historical Archive of Chiapas).

Table 6.

f. Sinaloa

Sinaloa is located in the northwest region of Mexico and adjoins the Pacific Ocean with 640.17 km of coastline. Its geographical position is between 27°07′ and 22°20′N (INEGI). According to Table 1, it has the second highest percentage of LFTCs in the MPc.

The climate in Sinaloa is mainly warm and most of the coastal zone is also dry with annual precipitation between 200 and 800 mm. The eastern side is mostly temperate with summer rains and annual precipitation between 700 and 1500 mm.

To determine the number of historical TCs that made landfall in the Sinaloa coast we consulted Sinaloa’s Chronic Research Institute, which provided information from books and other publications that reported disasters caused by TCs; government information at the Historical Files of Sinaloa (AHS); and the Confederation of Agricultural Associations in Sinaloa (CAADES) through its Agro-Meteorological Office.

These data were compared with the SMN charts for events between 1922 and 1943 and were related to a storm track. A total of 14 events were identified as TCs in Sinaloa between 1883 and 1943 (see Table 7).

Table 7.

Meteorological events possibly related to LFTCs in Sinaloa between the years 1883 and 1943 (CAADES; The Sinaloa’s Chronic Research Institute; AHS).

Table 7.

In Table 8 there is a summary of the number of events found in each state.

Table 8.

Meteorological events possibly related to LFTCs in the MPc years 1536–1948.

Table 8.

4. Analysis and results

With the historical data obtained, we constructed a time series between 1536 and 1948 shown in Fig. 3 and we applied a spectral analysis using the wavelet method to study the variability of the LFTCs founded.

Fig. 3.
Fig. 3.

Time series of LFTCs between 1536 and 1948.

Citation: Journal of Climate 26, 12; 10.1175/JCLI-D-11-00411.1

To analyze local variations of power within a single nonstationary time series at multiple periodicities we apply the wavelet method using the Morlet function (Torrence and Compo 1998). This function was specifically developed for geophysical phenomena signals and provides a good time resolution. Meaningful periodicities (confidence level greater than 95%) must be inside the cone of influence (COI), which is the region of the wavelet spectrum outside where the edge effects become important (Torrence and Compo 1998). We also include the global spectra in the wavelet plots, which is the average power at each period over the whole time span considered. This is a mean to show the contributions of the power from each periodicity inside the COI (e.g., Velasco and Mendoza 2008). We establish our significance levels in the global wavelet spectra with a simple red noise model (Gilman et al. 1963). We only took into account those periodicities above the red noise level.

First, we applied a wavelet analysis to the TC Weather Unysis database in order to find the periodicities in the LFTC time series between 1949 and 2010. The global spectrum shows peaks of 2.7 and ~5 yr, weakly present in ~1975 and 1985, and also shows peaks of ~12 and ~16 yr. Both periodicities are present along the entire time span inside the COI.

To compare spectrally the instrumental data along the time span (1949–2010) with the historical results, the wavelet analysis was performed in time spans of 62 yr of historical data. The time spans are 1701–62, 1763–1824, 1825–86, and 1887–1948. For data between 1536 and 1700, the analysis was not performed due to the low data density. The analysis was made also for low and high frequencies. The global spectrum obtained for each time span was correlated with the global spectrum of the 1949–2010 period. The correlation values are in Table 9.

Table 9.

Correlation values between the instrumental data of LFTC number (1949–2010) and 62-yr time spans of historical data of LFTC number.

Table 9.

As historical periods show spectral consistency with the instrumental period, we made a dataset of LFTCs from all the historical data and instrumental data. This dataset covers the time span between 1701 and 2010 to see the lower frequencies, and between 1850 and 2010 where we apply the same wavelet analysis as Fig. 4. Figure 5a shows in the global spectrum the low-frequency periodicities of ~105 yr. Figure 5b presents the wavelet analysis of the time series from 1850 to 2010. The global spectrum shows periodicities of ~2, 4, 5, 12, and 39 yr.

Fig. 4.
Fig. 4.

Spectral analysis for the 1949–2010 time series. (bottom left) the wavelet power spectrum in grayscale with the COI (the border between the black and gray backgrounds). (top) The time series and (right) the global spectrum with the red noise level (dotted curve).

Citation: Journal of Climate 26, 12; 10.1175/JCLI-D-11-00411.1

Fig. 5.
Fig. 5.

As in Fig. 4, but for the (a) 1701–2010 and (b) 1850–2010 time series.

Citation: Journal of Climate 26, 12; 10.1175/JCLI-D-11-00411.1

In Table 10 a summary of periodicities is presented.

Table 10.

Periodicities of LFTCs in the MPc.

Table 10.

The wavelet analysis yielded results indicating periodicities of LFTCs in the MPc of ~2.6, 4, 5, 12, 16, 39, and 105 yr for the time span of 1701–2010. The most conspicuous periodicities during the instrumental period (1949–2010) are ~12 and 2.6 yr, while during the whole time span, ~105 and 12 yr were most conspicuous. All the periodicities are weak between 1750 and 1850, which can be attributed to the lower number of data compared with posterior time spans (see upper panels of Fig. 5). Some of these periodicities, taking into account the uncertainties, coincide with some natural large-scale climatic phenomena such as the Atlantic multidecadal oscillation (AMO), the Pacific decadal oscillation (PDO), and the Southern Oscillation index (SOI) (e.g., Velasco and Mendoza 2008) or external phenomena such as the 11-yr solar activity cycle.

The ~4-yr periodicity coincides with the SOI, and the 100-yr periodicity with the AMO (Velasco and Mendoza 2008). The ~12-yr period is the most persistent periodicity of the analyzed time series, coinciding with the PDO and the ~11-yr solar cycle. Other studies have also found this periodicity considering TCs in the MPc (e.g., Englehart et al. 2008; Mendoza and Pazos 2009).

5. Discussion and conclusions

The event selection presented here is a first version of a catalog of possible TCs that affected the MPc between the sixteenth and twentieth centuries. The access to instrumental records of some MPc states containing data of precipitation, pressure, temperature, and wind velocity previous to 1949 would be very useful to identify more TCs, and also to improve the results of this work, since the historical data are limited to testimonies and populated places.

In Baja California Sur, in the colonization period (1521–1810) the Spanish fleet was affected several times by intense storms that could be reported in files kept by the General Archive of Indies. This archive can be used as another source for identification of TCs as has been done for the Caribbean basin (García-Herrera et al. 2007).

From all the events found in historical files, we selected 119 related to hydrometeorological phenomena that occurred near the shore on dates within the TCs season and for the type of damages reported according to SSS at least for category 1 hurricanes. We found a total of 85 possible TCs and constructed a time series between 1536 and 1948.

According to this study, Baja California Sur is the state with the largest number of LFTCs, which agrees with previous results using data for the second half of the twentieth century (e.g., Jauregui 2003).

Finally, with these results, we hope this paper will help to pave the way for future studies in an effort to better understand the TC climatology of the Mexican Pacific coast.

Acknowledgments

This work was supported by CONACyT Grant F282795. We thank Adrian Garcia, Mr. Rogers and Rosario Heras in Sinaloa, Noe Gutierrez and Martin Sanchez in Chiapas, Luis Rochin, and all the staff of the Pablo L. Martinez Historical Archive and Alberto Hernandez from the SMN for their kind assistance during the consult of the archives and for the information that they provided.

REFERENCES

  • Berumen Barbosa, M., 2003: Geografía económica de Oaxaca. 91 pp. [Available online at http://www.eumed.net/cursecon/libreria/mebb/index.htm.]

  • Clavijero, F., 1852: Historia de la antigua o Baja California. Juan R. Navarro, 258 pp.

  • Englehart, P. J., , and A. V. Douglas, 2001: The role of eastern North Pacific tropical storms in the rainfall climatology of western Mexico. Int. J. Climatol., 21, 13571370.

    • Search Google Scholar
    • Export Citation
  • Englehart, P. J., , M. Lewis, , and A. Douglas, 2008: Defining the frequency of near-shore tropical cyclone activity in the eastern North Pacific from historical surface observations (1921–2005). Geophys. Res. Lett., 35, L03706, doi:10.1029/2007GL032546.

    • Search Google Scholar
    • Export Citation
  • Escobar, A., 2004: Siglo XIX (1822–1900). Vol. 2, Desastres Agrícolas en México. CIESAS, 280 pp.

  • Fuentes, O., , and M. Vázquez, 1997: Probabilidad de presentación de ciclones tropicales en México. Cuaderno de Investigación, Vol. 42, CENAPRED, 138 pp.

  • Gallego, D., , R. García-Herrera, , N. Calvo, , and P. Ribera, 2007: A new meteorological record for Cádiz (Spain) 1806–1852: Implications for climatic reconstructions. J. Geophys. Res., 112, D12108, doi:10.1029/2007JD008517.

    • Search Google Scholar
    • Export Citation
  • García-Acosta, V., , J. M. Pérez-Zevallos, , and A. M. del Villar, 2003: Catálogo Histórico, Vol. 1, Desastres Agrícolas en México, CIESAS, 506 pp.

  • García-Herrera, R., , L. Gimeno, , P. Ribera, , and E. Hernández, 2005: New records of Atlantic hurricanes from Spanish documentary sources. J. Geophys. Res., 110, D03109, doi:10.1029/2004JD005272.

    • Search Google Scholar
    • Export Citation
  • García-Herrera, R., , L. Gimeno, , P. Ribera, , E. Hernández, , E. González, , and G. Fernández, 2007: Identification of Caribbean basin hurricanes from Spanish documentary sources. Climatic Change, 83, 5585, doi:10.1007/s10584-006-9124-4.

    • Search Google Scholar
    • Export Citation
  • García-Oliva, F., , E. Excurra, , and L. Galicia, 1991: Pattern of rainfall distribution in the central Pacific coast of Mexico. Geogr. Ann., 73, 179186.

    • Search Google Scholar
    • Export Citation
  • Gilman, D., , E. Fuglister, , and J. Mitchell, 1963: On the power spectrum of “red noise.” J. Atmos. Sci., 20, 182184.

  • Jauregui, E., 2003: Climatology of landfalling hurricanes and tropical storms in Mexico. Atmósfera, 4, 193204.

  • Latorre, C., , and L. Penilla, 1988: Influencia de los ciclones en la precipitation de Baja California Sur. Atmósfera, 1, 99112.

  • Ludlum, D., 1963: Early American Hurricanes 1492–1870. American Meteorological Society, 198 pp.

  • Mendoza, B., , and M. Pazos, 2009: A 22 yr hurricane cycle and its relation with geomagnetic activity. J. Atmos. Solar-Terr. Phys., 71, 20472054.

    • Search Google Scholar
    • Export Citation
  • Salinas-Zavala, C., , A. Leyva-Contreras, , D. Lluch-Belda, , and E. Díaz-Rivera, 1990: Distribución geográfica y variabilidad climática de los regimenes pluviométricos en Baja California Sur, México. Atmósfera, 3, 217237.

    • Search Google Scholar
    • Export Citation
  • Serra, S., 1971: Hurricanes and tropical storms of the west coast of Mexico. Mon. Wea. Rev., 99, 208302.

  • Sumner, H., 1944: North Atlantic hurricanes and tropical disturbances of 1943. Mon. Wea. Rev., 71, 179183.

  • Torrence, C., , and G. P. Compo, 1998: A practical guide to wavelet analysis. Bull. Amer. Meteor. Soc., 79, 6178.

  • Velasco, V., , and B. Mendoza, 2008: Assessing the relationship between solar activity and some large scale climatic phenomena. Adv. Space Res., 42, 866878.

    • Search Google Scholar
    • Export Citation
  • Villanueva González, E., 2002: Presencia de huracanes en Baja California Sur: El caso del ciclón Liza. Master’s thesis, Universidad Autonoma de Baja California Sur Social Sciences and Humanities Department, 227 pp.

Save