Introduction

Weather is the main environmental factor limiting crop productivity. These conditions are often modulated by synoptic scale weather events and global teleconnections as the phenomenon of El Niño Southern Oscillation (ENSO). For ^{Salinger et al. (2000)} , ENSO is an important force of short-term of climate variability and responsible for changes in the ocean-atmosphere system, which reveal economic, social, environmental, political and academic implications as detailed by ^{Zebiak et al. (2014)} . This phenomenon is characterized by the temperature anomaly in the equatorial Pacific Ocean surface in the region of3.4 El Niño, the warm phase is known as El Niño which consists of a positive anomaly while La Niña is the cold phase or negative anomaly, the Neutral phase is when the anomaly ranges on mean values 26.8 ± 0.5 °C (^{Latif and Keenlyside, 2009}; ^{Maturana et al., 2004}).

There are several models called indexes that explain with a degree ofaccuracy the behaviour ofthe ENSO; however, as a rule among American countries, we use ONI (Oceanic Niño Index) index developed by the Climate Prediction Center of the National Oceanic and Atmospheric Administration (NOAA). The ONI is the moving average ofthe temperature anomaly of the sea surface in three consecutive months (^{Guevara-Díaz, 2008}).

The use of indices that describe the phases of ENSO and its interaction with rainfall has been well documented, for example in South America, ^{Pinilla and Pinzón (2012)} , ^{Ramírez-Builes and Jaramillo-Robledo (2009)}, and ^{Ruiz-Cabarcas and Pabón-Caicedo (2013)} , reported decreases in precipitation and increases in Niño phase during Niña phase. Similarly, the ENSO is related to the impact on crops such as ^{Monasterio et al. (2011)} , pointed low yields in Niña phase mainly by an increase in precipitation in the flowering stage and grain filling in maize.

Similarly, ^{De La Casa and Ovando (2006)} found statistically positive anomalies in maize yield in El Niño and negative phase La Niña. In Mexico the impacts of ENSO have been sized by different authors, highlighting the work of ^{Adams et al. (2003)} who found in years El Niño decreased maize established surface and an increased price in contrast to years La Niña where price decreases because of an increase in production. This is closely related to the distribution of rainfall, this being confirmed by ^{Méndez-González et al. (2007)} and ^{Pavia et al. (2006)} which showed significant teleconnection between the phases of ENSO and monthly rainfall.

Some crops are severely affected due to increased extreme droughts, erratic rainfall distribution or extremely humid rainfed because of phenomena ENSO (^{FAO, 2010}). Precipitation in particularly its rainfed distribution best explains the variability of maize yield (^{Llano and Vargas, 2011}). In Mexico, maize is not only important in food or industrial terms but also socially and culturally (cereals produced in Mexico is 65% maize). Similarly, the 8.5 million hectares of cultivated maize in Mexico, 88% are in the spring or summer seasonal cycle in this cycle are obtained on average 2.3 t ha^-1 (^{SAGARPA, 2011}).

In Nayarit, just over 40 thousand ha of rainfed maize representing 1.9% of the national area cultivated. While Santa María del Oro in Nayarit is important for assessments of maize by the National Research Institute for Forestry, Agriculture and Livestock (INIFAP), as well as being of municipalities with higher yield (5 t ha^-1) and have an established area of over 2000 ha (^{SIAP, 2013}). Despite the relative importance of the crop for the region, the impact of ENSO on meteorology in different phenological stages is not known yet. Analyse the impact of the phases of ENSO in some meteorological variables in the period from planting to flowering by type of vegetative cycle was the objective of the present work. The results provide support for decision-making by producers and technicians for selection of appropriate types of varieties rainfed depending on the stage of ENSO.

Materials and methods

Study area. The experimental site represents the rainfed maize belt in Nayarit, Mexico. It was located in the vicinity ofthe town San José de Mojarras, municipality Santa María del Oro at an elevation of 940 meters, where the moisture regime with strict rainfed between June and OctoberPhenological data base. Originating from the Research Program of Maize by INIFAP. A database with 15 years of records was used from 98-99 to 12-13 cycle. This consisted of 344 records of commercial and regional varieties of maize, of which planting date (FS), flowering date (FF) were obtained. The following criteria were used to categorize the varieties according to their growth habit: < 57 days early flowering, medium > 57 and <62 days to flowering and late flowering > 62 days.

Oceanic Niño index. From the website of the Climate Prediction Center ofthe National Oceanic andAtmospheric Administration United States, monthly data from oceanic Niño index concerning the region 3.4 were obtained (5 °N-5° S, 120° - 170° W), corresponding to the anomaly of the quarterly moving average. The values were classified into three phases: warm or El Niño (> 0.5 °C anomaly), Cold or La Niña (anomaly < -0.5 °C) and neutral with anomalies between -0.5 °C and 0.5 °C (^{NOAA, 2014}).

Meteorological data. We used the available database of the National Weather Service station Cerro Blanco 18005 code that is located at coordinates north latitude 21° 22' 36" west longitude 104° 37' 06" at an elevation 965 m, which consisted of daily maximum temperature records (Tmax), minimum temperature (Tmin), average temperature (Tmed= (Tmax + Tmin)/2) and precipitation.

Study variables. For the quantification of the weather in the vegetative stage of each variety and established cycle period "FS" to "FF" was considered. The variables were calculated as follows: precipitation (P)= precipitation amount accumulated in the period. Degree Days development (GDD)= Tmed-Tb; where: Tb= 10 °C (Shaw, 1975; Ruiz et al, 2011). Potential evapotranspiration (ET₀)= 0.0023 * (Tmed+17.78) R_a (Tmax-Tmin)⁰⁵; where: R_a= extraterrestrial solar radiation in mm/day. Humid days (DH)= represents the number of days with precipitation higher than 1 mm. Simple daily intensity index of precipitation (ISIDP)= average precipitation in wet days (P/ DH) amount. Humidity Index (IH)=ratio ofprecipitation to potential evapotranspiration (P/ET0).

Data management. A database of growth habit, production cycle, yield and meteorological variables organized by phase ENSO was integrated.

Statistical analysis. In order to identify possible differences by phase ENSO an analysis of variance was used. When significant differences occurred, a comparison test was performed (Tukey with an alpha of 0.05% confidence level) using the statistical program Minitab version 16.

Table 1 Phases of ENSO per production cycle. 

Results and discussion

The effect of the phenomenon of El Niño Southern Oscillation on meteorology in Santa María del Oro for the period from planting to flowering in rainfed maize in the 15 cycles analysed showed significant differences in the different phases. The La Niña was characterized by wetter conditions than normal contrast what happens in the El Niño phase, associated with drier conditions. This coincides with the findings of ^{Pavia et al. (2006)} ; ^{Moeletsi et al. (2011)} ; ^{Ruiz-Cabarcas and Pabón-Caicedo (2013)} , where the phases of ENSO contrast mainly remained positive precipitation anomalies for La Niña and negative for El Niño.

The analysis of meteorological variables from FS to FF, distinguishing the phase of ENSO can be seen as rainfall (P) showed statistical differences between the three phases, La Niña episode has the highest cumulative amount in the period (43.6%), equivalent to 200 mm of difference to El Niño years had the lowest accumulation (467.36 mm). In regard to damp days (DH), years with El Niño phase had the lowest amount with 30.77 days, while the neutral and La Niña years were not statistically different from each other (higher than 38 days).

Simple daily intensity index of precipitation (ISIDP) showed differences between the three phases being the fastest years La Niña wet day precipitation (17.19 mm) in contrast Niño years (15.24 mm). Potential evapotranspiration (ET₀) showed no difference between Niño and neutral years (> 320.9 mm); however, La Niña years were 9 mm lower with respect to El Niño. Moisture content (IH) showed statistical differences between the various stages of ENSO, but in terms of classification rate only in phase transits El Niño "wet-wet" environment whereas in years La Niña phase with neutral and the atmosphere is very wet. Finally, no differences were found for phase ENSO in development accumulated degree days (GDD) being the range found between 818.75 and 807.51 °C. ^{Ruiz et al. (1999)} quoting various authors indicated that maize thrives in multiple environmental conditions but, this plasticity is influenced by crop growth habits, which is found with those found in this study (Table 2).

Table 4 Influence of ENSO on meteorological variables from sowing to flowering on improved maize varieties under rainfed conditions. 

Means with same literal in line do not present significant differences according to the test of Tukey a p≤ 0.05

Precipitation

The results of analysis of variance showed statistically significant differences in P. For early varieties in year La Niña, we recorded over 750 mm of rainfall at flowering (Figure 1), while in Neutral years for middle and late varieties P accumulates over 650 mm. For the three growing seasons the years with El Niño phase accumulate less P from planting to flowering (420-520 mm). Works, such as those made by ^{Granados-Ramírez and Sarabia-Rodríguez (2013)} state that, the water requirements for rainfed maize should be above 400 mm at flowering, while ^{Ruiz et al. (1999)} and ^{Ruiz et al. (2013)} pointed out that maize thrives accumulated more than 600 mm in the complete cycle P. The results of this study show how these requirements are provided in the various stages of ENSO and the three vegetative only flowering cycles, so that in terms of P can be considered as a non-limiting factor for crop development in this region, on this phenological stage.

Figure 1 Cumulative precipitation on flowering per vegetative stage regarding ENSO 

Wet days

In regard to the vegetative cycle analysis, the phase El Niño years accumulated less "DH" (Figure 2). The varieties of intermediate and late cycles showed no significant difference between Neutral and La Niña years, while early varieties are statistically different between the three phases of ENSO.

Figure 2 Wet days to flowering per vegetative stage regarding ENSO. 

Simple daily intensity index of precipitation

Varieties with intermediate growth cycle showed no statistical difference for phase ENSO, the average being 16 mm rainfall event (Figure 3). The early varieties in years La Niña showed higher mean differences with 20 mm of precipitation in contrast with El Niño and neutral years, averaging between 15 and 17 mm. In late varieties Neutral excelled year with precipitation averages 17 mm per day unlike years El Niño with 14 mm stage. La Niña phase showed no statistical difference with the other phases.

Figure 3 Simple daily intensity index of growing season precipitation stage regarding ENSO. 

Potential evapotranspiration

Late varieties showed no statistical difference for phase ENSO (320-350 mm). While the early and intermediate varieties, years with stage La Niña had the lowest averages in ET0 as shown in the Figure 4. ^{Texeira et al. (2013)} found for a region of Buenos Aires, Argentina that events ENSO is associated with positive anomalies in evapotranspiration under La Niña and negative for El Niño, contrary to the findings in Nayarit, though the ENSO is manifested in the southern region ofthe continent inversely to the north by what can be considered as a similarity between the regions.

Figure 4 Potential evapotranspiration to flowering from vegetative cycle regarding ENSO. 

Moisture content

In the vegetative cycle analysis in La Niña, the highest numbers in this index in all habits were obtained (Figure 5). In the middle and late varieties in the Neutral and La Niña phases, no significant differences in contrast to the Niño phase were observed, where the lowest averages were gathered. ^{Ruiz et al. (2013a)} identified 678 accessions representing 24 maize landraces, which they were adapted to arid environments, unlike the above, this study identified that although phase ENSO in the "IH "ever higher values were recorded at 1, i.e. moist environment, so it can be inferred that better moisture conditions are presented in Nayarit in the regions studied by ^{Ruiz et al (2013b)}.

Figure 5 Moisture index per vegetative cycle regarding ENSO. Grados día de desarrollo 

Degree days development

Late varieties accumulated more than 850 "GDD" at any stage ENSO. The early accumulated between 760-790 GDD", being in Neutral years where the accumulation was significantly lower (760 GDD) and finally the intermediate varieties between 810-830 "GDD". These were in the years with El Niño phase which accumulated more "GDD" in the three growing seasons (Figure 6). These results agree with rain early group races such as Conejo, Zapalote Chico, Ratón and Nal-Tel, described by ^{Ruiz et al. (2013a)}. This allows us to speculate on the phenotypic plasticity present in maize (wide adaptation) in landraces and the vulnerability of some commercial maize on the specifics of their habit and development environment (close match). ^{Mercer and Perales (2010)} concluded that landraces are better adapted to changing environmental conditions due to their larger genetic variation that confers them phenotypic plasticity to adapt to unfavourable environments.

Figure 6 Degree days of development to flowering per vegetative cycle regarding ENSO. 

Conclusions

The influence of the phenomenon of El Niño Southern Oscillation on meteorology in the production region of rainfed maize in Nayarit, Mexico in the period from planting to flowering was identified. The phases of this phenomenon affect deferred according to the type of maize vegetative cycle, established under rainfed conditions. The La Niña events are characterized by predominantly wet conditions while the El Niño phase was significantly lower in the five moisture related indicators. There were no statistical differences between phases with respect to the accumulation of GDD. Precipitation being dominated by the phase ENSO, showed five meteorology associated with it, statistical differences with respect to the vegetative cycle indicators, with the early varieties which showed marked differences between the three phases.

While the intermediate and late phase differed only in El Niño (under less humid conditions). The results open the possibility of continuing to work to consider the rest of the phenological stages of the crop in order to characterize the conditions for rainfed maize. With all this we can conclude that, the phases of ENSO influence clearly in some meteorological variables which are important in the development of the crop, which can be used as a nascent early warning system combined with forecasts models of ENSO, published on a monthly basis by the International Research Institute for Climate and Society.