Introduction

The tequila agave is the raw material to elaborate tequila, a product of high national and international demand with a production over 242.2 million liters of tequila with 788 200 t of agave in 2014 (^{CRT, 2015}). Agave tequila is produced in the appellation of origin of tequila (DOT), comprising the state of Jalisco and some municipalities in the states of Nayarit, Guanajuato, Michoacan and Tamaulipas. The dynamics of the area planted with agave tequila has shown a tendency to increase due to the high demand of tequila (^{SIAP-SAGARPA, 2015}). However, the production of agave in DOT has shown irregularities in recent years by factors such as uncontrolled growth of the planted area with agave, overproduction and shortage of agave, presence of various plant health problems, lack of appropriate technology to the socioeconomic conditions for agave producers, which in turn affects productivity between contrasting areas and deterioration of natural resources, especially soil loss.

Wilt is one of phytosanitary problems facing tequila agave, but accentuated when there is overproduction so that the price of agave decreases and the crop is abandoned, and when there is shortage of agave this problem is reduced, because the demand for the tequila industry absorbs all the agave grown; still, the phytosanitary problem of agave wilt persists.

Agave wilt is attributed to Fusarium oxysporum (^{Rubio, 2007}), phytopathogenic fungus widely distributed in cultivation soils and reported in crops like tomatoes (^{Larkin and Fravel, 2002}), lawn (^{Goldberg, 2006}), oil palm (^{Flood, 2006}), pea (^{Landa et al., 2001}; ^{Landa et al., 2006}), cotton (^{DeVay et al, 1997}), banana (^{Ploetz, 2006}), white pine seedlings (^{Ocamb and Juzwik, 1995}), among others.

^{Nelson (1981)} mentioned that wilt symptoms by fusarium in herbaceous plants, consists of an incipient wilting of lower leaves or buds. Subsequently come chlorosis in lower leaves followed by permanent wilting of these leaves. Symptoms appear gradually in the youngest leaves of the plant, often occurring on one side of the same. Finally the affected part of the plant withers and dies. This process differs in perennial crops such as banana, date palm, mimosa, among others. In herbaceous plants, the disease occurs in days, while in woody plants such as bananas develops in 2-5 months (^{Nelson, 1981}).

These wilt characteristics of fusarium are observed in tequila agave and is reported as the most prevalent phytosanitary problem in Jalisco (^{CRT, 2010}). F. oxysporum attacks agave plants regardless of their growth; begins with a pale leaf, mainly in the core of pineapple and pushing the whorl, leaves that subsequently wilt, affecting the core of pineapple causing plant death (Flores et al, 2010). Wilting in agave is the result of a series of cycles of infection by fusarium, which considers infection processes, colonization and survival (^{Nelson, 1981}).

The cycles of infection develop in time and space, mild or severe, explosive or subtle, host interaction and triggered by the environment (soil and climate) or human interference, which operate at different levels (^{Kranz and Hau, 1980}) or sub process (Zadoks and Schein, 1979).

Within the climate component from the epidemiological system by fusarium, temperature and soil moisture are the main primary factors governing the cycles of infection (^{Nelson, 1981}) and Fusarium species with aerial action, water activity on the foliage is important (^{Marin et al, 1995}; Ramirez et al, 2005). The temperature required for fusarium species to develop cycles of infection varies widely. ^{Katan (1989)} summarizes the relative growth of F. oxysporum in function of the temperature on the model shown in Figure 1. The temperature for optimum growth of F. oxysporum in the soil is defined between 25 and 28 °C, with growth from 15 to 38 °C (^{Smith et al., 1988}; ^{Katan, 1989}). Depending on the crop, other fusarium species have an optimum temperature between 22 to 28 °C and extreme ranges between 2 and 39 °C (^{Lacey, 1989}). ^{Timmer (1982)} found that the F. oxysporum on citrus showed the greatest severity between 21 to 26 °C, with significant decrease with temperature below 15 °C and higher than 32 °C. In coca plant, fusarium showed optimal development at 25 °C, with extremes between 10 and 35 °C (^{Fravel et al, 1996}).

Figure 1 Relative growth of Fusarium oxysporum in function of the temperature (adapted from ^{Katan, 1989}). 

Soil moisture complements suitable temperature conditions to generate infection cycles by pathogens. F. oxysporum is an aerobic fungus, with requirements of soil moisture for optimum growth and survival near field capacity, but under saturation or flood conditions, its population is reduced (^{Nelson, 1981}). Depending on the physical characteristics of the soil, optimum matrix potential of soil moisture have been reported for fusarium from -10 kPa to -100 kPa, with less proliferation -500 kPa (^{Fravel et al., 1996}). Water activity is an important element in cycles of infection of fusarium, reporting optimal values in 0.98 and minimum of 0.87 (^{Lacey, 1989}; ^{Carrillo, 2003}). Temperature-soil moisture interaction is important in the cycle of infection, ^{Katan (1989)} mentioned that in melon the incidence of wilt by fusarium, the temperature is associated with soil moisture and inoculum concentration; so that the disease level on wet soil was higher at low temperatures and with dry soil was high in low and high temperatures (^{Fravel et al, 1996}).

Epidemiological system components for fusarium act under a systemic structure and are made up of basic biological elements, plant health control system, farming systems and agro-ecosystem management (Hau and ^{Kranz, 1980}). The resulting interactions between these elements generate a complexity of such magnitude that hinders their understanding, operation and adequate control in the crop. If a broader spatial perspective is involved, as the DOT, technological and environmental changes occurred in recent years modify the frontier where phytosanitary problems arise, as reported wilt dynamics mentioned by the ^{CRT (2010)}.

The main purpose of the studies of climate change on agriculture is to estimate the expected changes in weather and its influence on living organisms, changes in the concentration of CO2 in the atmosphere, among others (^{Manici et al., 2014}). The effect of climate change on crops and soil pathogens such as F. oxysporum requires understanding to anticipate pertinent changes in agricultural management and ecosystems, spatial and temporal modification in climate generate responses in pathogens and crops that require to be identified (^{Shaw and Osborne, 2011}). This complicates by the lack of knowledge related to the complexity of host-pathogen-climate-pest interaction and agronomic damage, the specificity of host-pathogen attacks, diversity of pathogens (fungi, viruses or batteries) and sequence of induced stresses by climate that increase host susceptibility to the pathogen increasing its effects (^{Grulke, 2011}).

One way to address the shortage of information on climate-host-pathogen relationship under conditions of climate change is to use historical information of these relationships that could provide understanding on these interactions. In the tequila agave, the National Institute of Forestry, Agriculture and Livestock Research (INIFAP) has conducted studies in Jalisco (^{Flores, 2000}; ^{Flores et al., 2009}), with useful results to elucidate part of the pathogen climate-host-interaction under the influence of climate change and the partnership with the phytosanitary problem of wilting in DOT. These studies can be used to identify environmental changes that are occurring associated to this crop and plant pathology risk by fusarium in agave growing areas, to take appropriate measures to reduce or avoid risk in the raw material supply chain agave-tequila. The aim of this study was to evaluate the effect of climate change in Jalisco on wilt of agave tequila, based on the use of indices obtained from field monitoring of tequila agave during 1997-1998 and 2008-2009.

Materials and methods

This analysis was made for the state of Jalisco, located in the pacific center region of Mexico. Two studies from INIFAP on tequila agave wilt in the period 1997-1998 and 2008 to 2009 were used. In this study the Producer-Experimenter model (^{Villarreal, 2000}) was used as a diagnostic method to identify the level of wilting present. In the period 1997-1998, 97 fields were monitored and in 2008-2009 were 105. The age of the agave plantations was 1-7 years for both monitoring periods, with different levels of production technology. The distribution of field in the state of Jalisco for the periods of analysis is shown in Figure 2.

Figure 2 Distribution of fields with agave tequila monitored in 1997-1998 and 2008-2009 and distribution of weather stations in Jalisco. 

Fields selected with agave tequila had an area of one hectare with plants of 1-7 years. On these surface 50 plants were marked with distribution in five golds with groups of 10 plants each point. Bioclimatic indices associated with the presence of wilting were used. During the monitoring period 1997-1998 the number of rolled leaves, registered from July 1997 to January 1998 were used. In the period 2008-2009 the number of leaves emerged from the month of July 2008 to January 2009 were recorded.

In the period 1997-1998 rolled leaves rate (THE) index was used as wilt indicator in agave, expressed by the following relationship: ; where HE1 is the number of rolled leaves at the beginning of the study, HE2 is number of leaves to January 1998, (t2-t1) is the number of passing days in the recording period rolled leaves.

In the period 2008-2009 the leaf emergence rate (TEH) index was used as an indicator of agave growth, expressed by the following relationship: where EH1 is the number of emerged leaves of the whorl from the agave plant at the beginning of the study, EH2 is the number of emerged leaves to January 2009, (t2-t1) are the passing days in the monitoring period of agave leaves.

Results and discussion

Wilt risk areas in 1961-1995

Figure 3a shows the risk areas by wilting (ZRM) for ZR95 and Table 1 shows the area occupied by each risk level in Jalisco. The low risk is prevalent and is associated with very hot, dry areas, such as coastal and northern and high northern of Jalisco. The levels of high and medium risk are found in Cienega, Altos Sur, Valles and Sur regions from Jalisco.

Figure 3 Spatial distribution of wilt risk areas in tequila agave in Jalisco, estimated with climate from a) 1961 to 1995; and b) 1961-2009. 

Table 1 Estimated to risk levels by wilting of agave tequila for climates 1961-1998 and 2008-2009 area. 

The results of THE index from the monitoring period 1997-1998 were grouped by risk areas and are summarized in Table 2 and the relationship between THE by ZRM are shown in Figure 4. This figure presents the rate change by risk level is 0.0498 leaves / day with highly significant differences between levels of risk.

Table 2 Statistics of rolled leaves rate (THE) by wilt of agave tequila in monitored fields during 1997-1998, sorted into three levels of risk for Jalisco. 

Figure 4 Relationship between wilting areas of agave tequila with rolled leaves rate from monitored fields for the period 1997-1998 in Jalisco. 

THE was greater in the high-risk level with 0.0998 leaves / day, in the medium risk area with 0.0098 leaves / day and the lowest in the low risk area with 0.0022 leaves / day. These values are indicators of the dehydration rate of the agave leaves as result of wilt, so that in the high ZRM a leaf rolls every 10 days, while in medium and low ZRM requires over 100 days to roll a leaf.

Another important effect is the relationship between THE with agave age and ZRM, as shown in Figure 5. This figure shows that THE for the low-risk area, is maintained at a minimum value during the seven-year of agave cycle. For medium risk area, after the fourth year THE begins to increase until the seventh year where it reaches the highest value. In the high risk area it has the strongest effect of THE with agave age, in virtually the whole agave cycle, starting from planting to the seventh year.

Figure 5 Relationship between risk areas by wilting with rolled leaves rate and leaf emergence rate with agave age, with climate from a) 1961 to 1995; and b) 1961-2009, respectively. 

Wilt risk areas in 1961-2008

Figure 3b shows risk areas by wilting for climate 1961-2008 (ZR08) and Table 5 indicates the area occupied by each level of risk in Jalisco. The surface by risk zone has the same trend as ZR95, but in this period the areas of low and medium risk increase 46 275 and 3 299 ha, respectively. In the high risk zone the surface decreases 9 569 ha. This can be considered the first sign of the effect of climate change on ZRM of tequila agave in Jalisco.

The identification of ZRM in agave fields from the period 2008-2009 are summarized in Table 3 and the relationship of TEH by ZRM are shown in Figure 6. The slope on TEH by risk level is -0.0257 leaves / day. This value is highly significant between levels with tendency to decrease with change of ZRM. The average values of TEH in low, medium and high ZRM is 0.144 leaves / day, 0.108 leaves / day and 0.097 leaves / day, respectively.

Table 3 Statistics of leaf emergence rate (TEH) in agave tequila from monitored fields during 2008-2009, sorted into three levels of risk for Jalisco. 

Figure 6 Relationship between risk areas by wilting in agave tequila with leaf emergence rate on fields monitored during the period 2008-2009 in Jalisco. 

The relationship of TEH with agave age and ZRM are shown in Figure 5b. This figure shows that with the advance in age of agave, the TEH tends to decrease in the three ZRM, although in high ZRM the TEH is low since the early years of the plantation, while in low ZRM the TEH has a higher value (Table 3). These values indicate that a leaf is formed in low, medium and high ZRM, at 6.95 days, 9.25 days and 10.3 days, respectively.

Change of risk areas by wilting

Figure 7 shows the change in distribution on ZRM from Jalisco. This map is the result of the algebraic sum between ZRM of ZR08 minus ZR95. When the result is positive, the risk by wilting is reduced one or two levels, with zero value indicates no change of risk and negative value represents an increase in the risk level of one or two levels.

Figure 7 Change in level of risk by wilt in agave tequila, estimated with the comparison of risk areas by wilting for 1997 and 2008. 

Table 4 shows the affected surface by changes in ZRM and levels of change occurring. The surfaces in changing ZR08 minus ZR95 show that agro-climatic conditions for agave wilt is changing in the period of analysis and geographic factors like mountains, water bodies, arid areas, among others, largely define the spatial distribution of wilt (^{Heesterbeek and Zadoks, 1987}). It is noteworthy that the increase in risk by wilting is happening in the high parts of Jalisco while in coastal areas and medium altitude of less than 1 500 m, the risk by wilt is decreasing, result attributable to the increase of environmental temperature.

Table 4 Change in risk areas by wilting (ZRM), and the surface subject to change in ZRM, for the period 1997-2008. 

The change in climate conditions of ZR08 regarding ZR95 for F. oxysporum, indicates that has been affected 49.5% of the area from Jalisco, but at least 23.3% of the area changed their level of risk, favoring agro-climatic conditions for wilting in agave, observed as negative values in the change of ZRM. These trends are similar to those shown in studies where projections of climate change and its effect on crops are made (^{Masutomi et al, 2009}) or pathogens such as F. oxysporum (^{Tantaoui et al, 1996}; Ferrocino et al, 2014; ^{Shabani et al, 2014}).

In places where the severity of this phytosanitary problem for agave is clear, the risk by wilting should be considered as a factor of reduction of areas for the production of raw material for the tequila industry. It is also necessary to use future scenarios of climate change to project as it could be the expected trend in risk areas by fusarium of DOT to produce agave tequila.

Conclusions

The relationship between risk areas and monitoring of agave in 1997-1998 with the bioclimatic index of rolled leaves rate turn with a correlation of 0.37 with highly significant difference (p> 0.01). The age of agave has a close link with THE and risk areas by wilting.

Risk areas by wilt was also correlated with the bioclimatic index leaf emergence rate (TEH) with a correlation of 0.38 and highly significant (p> 0.01), using agave monitoring during the period 2008-2009.

Evidence of the effect of climate change on risk areas by wilt in Jalisco is shown. Two tendencies were found, one aimed to reduction of risk by Fusarium for agave tequila in 26.2% of the area, but another that increases the risk by fusarium in over 23.3% of the surface from Jalisco. This change was based on the empirical relationship of the rolled leaves rate and leaf emergence rate for 1997 and 2008, which were used as indicators of the association between agave wilt in three risk zones and age of agave tequila.

It is necessary to use climate change scenarios to project as it could be the future the trend in risk areas by fusarium in the DOT to produce agave tequila.