Introduction

The state of Sinaloa occupies the first place in surface established with 30 095 ha and is the third mango producer at national level with 235 772 t of fruit. In the southern region of Sinaloa (Mazatlan, Concordia, Rosario and Escuinapa), 84% of the state area is concentrated (^{SIAP, 2013}). The main cultivars are: Ataulfo, Haden, Keitt, Kent, Manila and Tommy Atkins. In Sinaloa, as in other states, the mango is attacked by pests of importance and quarantine, such as fruit flies of the genus Anastrepha spp. Or of economic importance such as the white mango scale (EBM) Aulacaspis tubercularis Newstead (Hemiptera: Diaspididae) (^{Urias 2006}; ^{Urias-López et al., 2010}).

The white scale is a pest, which caused damage initially in Compostela, Nayarit. It is of commercial importance because it causes direct damages in fruit quality (^{Urías-López et al., 2010}) and causes problems in the commercialization, mainly when it comes to fruits for export (^{Urías-López et al., 2010}, ^{Abo-Shanab, 2012}). The scale can also be found on tender stems, but it is more frequently located in the leaves, which decreases the photosynthetic area of the trees (Urías 2006). The control of this pest is based on the use of insecticides and unconventional products such as natural oils or petroleum products, soaps and detergents (^{Le Lagadec and Louw 2009}; ^{García- Álvarez et al., 2009}; ^{Urías-López et al., 2013}).

Although white scale is widely distributed in mango producing areas, few studies document the behavior of populations for management purposes, this is probably due to the fact that in a few countries it manifests itself as a pest. Of the cases that white scale damages the mango, South Africa (^{Le Lagadec et al., 2006}), Ecuador (Arias et al ., 2004) and Egypt (^{Abo-Shanab, 2012}) are mentioned. In a study of population fluctuation of scales in Nayarit, Mexico, it was determined that this pest recorded three population stages during the year: one of low population from the end of the rainy season from september to january, another from population increase from march to july and another with the lowest presence during the year of July-August (^{Urías-López et al., 2010}). The highest populations of the scale during the year are associated with warm periods and no rainfall, whereas those with the lowest populations of the plague during the summer rainy season (^{Urías-López and Flores-Canales 2005}; ^{González-Carrillo et al., 2008}, ^{Urias-López et al., 2010}).

The distribution of white scale in Mexico is found in all Pacific coastal states such as: Sinaloa, Colima, Jalisco, Guerrero and Oaxaca, as well as in the Gulf of Mexico in the state of Veracruz (^{Noriega et al., 2012}). In Nayarit, it was determined that the pest is distributed throughout most of the state, but with higher densities in the south and center of the state than in the central zone (^{García-Álvarez et al., 2014}).

Because A. tubercularis limits the commercialization of the fruit due to low quality of the product, mainly in the export market, since in Sinaloa there are no studies on the behavior of the populations in mango cultivation for management purposes, the objective of this work was to know the population fluctuation during the year and the geographical distribution of the white scale in Sinaloa.

Materials and methods

Two trials were carried out in the years 2012-2013 and 2013-2014 in southern Sinaloa, where the state’s largest mango surface is established (76%), mainly in the municipalities of El Rosario and Escuinapa. The first test was carried out to determine the annual population fluctuation of the scale and the second to determine the geographical distribution and degree of incidence of the pest. The general sampling methodology used for both trials is the one suggested by ^{Urías-López et al. (2010)}, this consisted in selecting five mango trees with similar characteristics, to which four branches were marked oriented to each cardinal point. In each terminal branch, two leaves of the penultimate vegetative flow were sampled, one located inside and the other outside the outbreak, totaling eight leaves/tree or 40 leaves per orchard. In each sample the number of colonies (males) and females detected per leaf were recorded.

Population fluctuation of white scale. This study was carried out in three mango commercial orchards located in the El Pozole community of El Rosario municipality, Sinaloa.

The white scale populations were examined biweekly for two years (from april 2012 to june 2014), the cultivars were Ataulfo, Kent and Keitt. In order to reinforce information on the population fluctuation of A. tubercularis with climate factors, temperature data were obtained in each orchard, for which data logger devices (HOBO® Brand) were installed. Rainfall data were obtained from the database of the Center for Research in Food and Development, Culiacán, Sinaloa (CIAD).

Geographic distribution of the white scale To know the geographical distribution and degree of incidence of A. tubercularis in southern Sinaloa, during the first year 80 commercial orchards were sampled, grouped in nine localities. In the second year, 141 commercial orchards were sampled from 10 localities, located in the main mango producing municipalities of the state (Rosario and Escuinapa). Sampling was carried out mainly during the fruiting season, between March and June, which is reported as the period of maximum incidence of the pest according to ^{Urías-López et al. (2010)} and ^{García et al. (2014)}. The distribution and incidence in the cultivars Ataulfo, Haden, Keitt, Kent, Manila and Tommy Atkins were determined using the methodology described in the first section.

Statistic analysis. In order to know the white scale populations between sampling dates, between cultivars or between localities, in the last two cases according to each test, variance analysis was performed with a randomized block design using the statistical package ^{SAS Institute (2000)}. For the comparison of means, the Tukey test (p≤ 0.05) was used. The variables studied were: colonies of scales (male colonies) and females, with these two variables a third, called total scales, was generated.

Results and discussion

Climatic information: weather data from El Rosario, Sinaloa, showed in the two years of study that in August was the warmest with 28.2 and 29.3 °C in 2012-13 and 2013-14, respectively. The coldest month in the first year was February (19 °C) and in the second year January with 20.5 °C. On the other hand, rainfall in the first year reached its highest values during the period of june-october, while in the second year it was june-august (Table 1).

Population fluctuation of the white scale in Sinaloa. During the first year, there were statistical differences (p= 0.0001) in the density of females and the total (more female colonies) of scales between sampling dates and between cultivars. The white scale was presented from february to july in the three cultivars (Keitt, Kent and Ataulfo). In Ataulfo, in May the highest average female was recorded (1.3 females/leaf); from april to july, there were low populations with averages of less than 0.05 females/leaf (Figure 1). Regarding the total of scales (colonies more females/leaf), the cultivar Ataulfo presented the highest average of the total of scales/leaf (2.4 in the month of may). The cultivars Keitt and Kent presented low average of the total of scales (colonies more females/leaf), with values smaller than 0.08 scales/leaf during the period of april to july (Figure 2).

Figure 1 Female populations of A. tubercularis on mango. El Rosario, Sin. 2012-2013 

Figure 2 Total population of A. tubercularis in mango. El Rosario, Sin. 2012-2013 

Second year of study. In the second year, significant differences were also observed in the density of the flake populations between sampling dates and between cultivars (p= 0.0001). The highest average female/leaf (3.08) was recorded in april 2014 in the cultivar Ataulfo. In this cultivar the lowest values were presented from july to october with averages at low of 0.5 females/leaf (Figure 3). The white scale populations in the cultivars Kent and Keitt behaved in a similar way, however, the Kent cultivar presented low values; in july it had a peak with an average of 1.25 females/leaf and there was no population from august to october 2014 (Figure 3). Regarding the total scales variable (colonies plus females/leaf), the cultivar Ataulfo recorded the highest averages during almost all the sampling dates (up to 3.53 scales/leaf). In the Kent and Keitt in cultivars for most of the year stocks of all scales did not exceed averages above a scale per leaf (Figure 4).

Figure 3 Female populations of A. tubercularis in mango. El Rosario, Sin. 2012-2013. 

Figure 4 Total population of A. tubercularis in mango. El Rosario, Sinaloa, 2013-14. 

The previous results coincide in part with those obtained by ^{Urías-López et al. (2010)} and ^{Salazar-Santiago (2012)}, who relate the increases in the populations with periods without rain and high temperatures, which coincide with the phenological stages of fruit ripening and maturation. ^{García et al. (2014)} found that white flies prefer to cultivate Ataulfo instead of Tommy Atkins, which coincides with this study because, in all cases, ‘Ataulfo’ presented higher averages of infestation. During the august-october rainy season the lowest values ​were presented in all three cultivars (Figure 4). In this regard, ^{Urías López et al. (2010)} mention that the absence of white scale during the summer is due to the mechanical effect of rainfall. Information that coincides with the result of the present study in the two years, the decrease of the population is attributed to the presence of rains in summer.

In general, the populations remained low during the rest of the year due to the low temperatures during the winter months and the rainfall during november and december (Table 1). According to the above, the population of white scale in Sinaloa goes through three periods of development; in the first the population has almost null levels (<0.01 scales/leaf) and occurs during the rainy season from August to October, this is in line with what was reported by ^{Urías-López et al. (2013)}, who points out that a decline in scale populations is due to the effect of rainfall. There was also a second period from november to february, where the white scale population was kept at low densities, below 1 scale/leaf (period of low population).

There was an increase in the population of white scale from february and ends at the beginning of the rains with three development phases, one of low population, one of increase and one more, where the population falls to very low levels in time of rain. In this case, ^{Bautista et al. (2013)} mentioned that white scale populations increase from March until the beginning of the rains, which coincides with this study.

Geographic distribution of the white scale by localities. In the first year a significant statistical difference (p= 0.0001) was detected in the average of colonies/leaf between the different localities. In the orchards sampled in the town of Teacapan, Sin. The highest average was recorded with 0.05 colonies/leaf, this being the average of the total of all the samples made during the year (Table 2).

The localities of El Rosario, Apoderado and Agua Verde did not present infestation of white scale. On the other hand, no significant differences were detected between localities in the female populations nor for the total scales (colonies/leaf plus females/leaf) (Table 2).

Regarding the distribution of the pest among municipalities, the analysis showed a statistically significant difference (p= 0.0001). In Escuinapa, a greater number of insects were registered with a total of 0.04 females/leaf plus colonies/ leaf, whereas the municipality of El Rosario did not present population. ^{Urías-López et al. (2010)}; ^{García-Álvarez et al. (2014)} indicated that the white scale appeared for the first time in the south of Nayarit and advanced gradually until reaching the north of Sinaloa. In this production area of Nayarit, low populations of scales are detected (^{García-Álvarez et al., 2014}), densities that were similar to those recorded in this study for the southern part of Sinaloa. However, the populations detected in Sinaloa are very low compared to the zones of greatest scale infestation in Nayarit (^{Urías López et al., 2010}; ^{García-Álvarez et al., 2014}). The density and distribution of the white scale in southern Sinaloa was so low that in 74.7%, of the orchards sampled were free of the pest, only 25.3% registered low incidence and no orchards with average incidence were detected (between one and two scales per leaf) or with high incidence (more than two scales/leaf) (Figure 5).

Figure 5 Spatial distribution of the total incidence (females plus colonies/leaf) of A. tubercularis in the mango producing region of southern Sinaloa. 

In the second year of sampling, significant differences were also detected (p= 0.0001) of the populations of scale between localities; the gardens of the locality Capomal presented the highest average with 1.85 females/leaf and a total (females/ leaf plus colonies/leaf) of 1.98 scales/leaf. The other localities did not show statistical differences between them (Table 3). However, it can be observed that white scale is present in most mango producing localities in southern Sinaloa.

The municipality of Escuinapa returned to present the highest incidence of scale with 0.23 females/leaf, on average per year. Regarding the number of colonies/leaf, in the municipality of El Rosario the largest amount was recorded with 0.21 colonies/leaf. Regarding the total of scales (females/leaf plus colonies/leaf) no significant statistical difference was detected between municipalities, which suggests that the white scale is evenly distributed in low populations in the south of the state. One of the possible reasons for the existence of low populations, with respect to those of the southern and central part of Nayarit (^{Urías et al., 2006} and ^{García-Álvarez et al., 2014}), could be the management of orchards by part of mango producers. Usually in this zone of Sinaloa, the producers carry out periodic activities of fertilization, pruning, as well as application of fungicides and insecticides in order to control pests and diseases such as the fruit fly, which could also indirectly affect the scale.

Distribution by cultivar. During the first year of sampling statistical differences (p= 0.0001) were detected between mango cultivars. The Manila variety showed the highest averages with a total (females/leaf plus colonies/leaf) of 0.25 scales/leaf. The Keitt cultivar recorded a total of 0.06 females leaf plus colonies/leaf. No statistical difference was detected between the cultivars Ataulfo, Kent, Tommy, Keitt and Haden regarding the variables evaluated (Table 4).

In the second year of sampling (2014) the cultivar Manila again showed a significantly higher average of scales than the other cultivars. The values recorded with this cultivar were 0.54 females plus colonies/leaf. The cultivars Ataulfo, Tommy Atkins and Haden recorded averages of 0.22, 0.15 and 0.13 females plus colonies/leaf, respectively. The cultivars Kent and Keitt did not exceed 0.04 females plus colonies/leaf (Table 5).

In some studies, higher populations of ‘Ataulfo’ mango foliage are recorded in relation to other cultivars, but are cases in which ‘Manila’ was not present (Urias et al., 2010 and ^{García-Álvarez et al., 2014}. In this study finding the largest population of scales in the Manila cultivar, coincides with ^{Montiel et al. (2014)}, who found the highest populations of white scale (up to 10 females/leaf) in ‘Manila’ compared to other mango cultivars in Actopan, Veracruz. In this sense, ^{Urías-López et al. (2010)}, indicate that the difference of the scale populations between cultivars, in general terms may be due to differences in the age of the orchards, size of the trees and density of leaves, orchards with large and relatively young trees, with abundant foliage, favor a favorable environment for the development of the white scale, which coincides with the fact that, the cultivars of the cultivar Manila were adults with foliage abundance, in the case of differences between cultivars of fruit flake populations, ^{Osuna (2006)} points out that it is probable that the difference between cultivars, particularly Ataulfo is due to the fact that during the stage before maturation this cultivar presents a higher degree of acidity and low content of sugars with respect to other varieties.

Conclusions

The population of A. tubercularis recorded three stages of development: during and after the rainy season (july-october) the population was the lowest. Low prevalence during low temperatures (november-february) and with the highest populations during the period of high temperatures and/or without rains (march-july).

The presence of mango white scale was recorded in 90% of the mango producing area of southern Sinaloa, in the locality El Capomal municipality of Escuinapa, Sinaloa. ‘Manila’ was the most infested cultivar.