HOURLY AND DAILY CHANGES ON AIRBORNE UREDINIOSPORES OF Phakopsora pachyrhizi

The Asian rust (Phakopsora pachyrhizi) is one of the most destructive diseases of soybean in Brazil. Despite its importance, little is known about the airborne inoculum production dynamics of pathogen throughout the soybean cycle in Brazil. The objective of this study was to assess the temporal variation of air collected P. pachyrhizi urediniospores using a Burkard 7-day-spore-trap during 2006-2007 and 2007-2008, from November to March of each soybean (MSoy-8001) growing season. The disease severity was quantified on soybean planted in October 18, 2006 and in October 30, 2007. From November/06 through March/07 were collected over 175,000 urediniospores, and, over 131,000 from November/07 to March/08. In 06-07, most of the spores were collected from January 9, 2007 through February 1, 2007 (±3,000 urediniospores day), and, in 07-08 most urediniospores was collected from February 8, 2008 throughout March 2, 2008 (±2,000 urediniospores day). The maximum amount of urediniospores collected in a single day in 06-07 (±15,000; 1/25/2007) was higher than in 07-08 (±12,000; 2/13/2008). In both soybean growing seasons most of the hourly urediniospores was collected from 10:00 am through 6:00 pm and the highest number from December to March, was at 3:00 pm (±16,000 in 06-07 and ±14,000 in 07-08). The disease started earlier in 2006-07 [38 days after planting (DAP)] than in 07-08 (58 DAP). The amount of collected airborne spores was positively correlated to disease severity, leaf wetness, and, precipitation along both soybean growing seasons. However, during a 24h-day evaluation, the major amount of hourly collected spores was negatively correlated to the leaf wetness and air relative humidity.


INTRODUCTION
The Asian rust (Phakopsora pachyrhizi Syd.& P. Syd.) of soybean (Glycine max L.) is a destructive foliar disease found in many soybeanproducing regions (YORINORI et al., 2005;ISARD et al., 2006;JARVIE 2009;ROSA et al., 2015;KELLY et al., 2015;MAUMARY et al., 2016;GODOY et al., 2016).Weather conditions influence directly in the cycle of the pathogen, which favors the development of rust epidemics on soybean (ISARD et al. 2005).Rainfall for 12-15 days is one of the main factors that explain the variations of disease severity in field condition (DEL PONTE et al., 2006;DUFAULT et al., 2010a).Marchetti et al. (1976), demonstrated that 6 hours of leaf wetness is the minimum time to ensure infection of soybean by the fungus.It is known that 18-26 o C and leaf wetness duration determine the process of infection of urediniospores of P. pachyrhizi (ALVES et al., 2007).
The survey of airborne fungi spores by variable means has been done (LUO et al., 2007;OLIVEIRA et al., 2009a;KUDO et al., 2011;DAS;GUPTA-BHATTACHARYA, 2012;ALMAGUER et al., 2013;RIEUX et al., 2014;AHER et al., 2015).The assessment of air inoculum can contribute to understanding the potential risk of epidemics of diseases during the plant cycle (WAKEHAM; KENNEDY, 2010;WEST;KIMBER 2015;IGARASHI et al., 2016).Airborne spores can be sampled in many ways; one of them is the stationary Burkard volumetric spore trap (WEST; KIMBER, 2015;WEST et al., 2017).This trap has been used for detection of fungi and pollen of plants (INCH et al., 2005;PEEL et al., 2014).
However, studies concerning the quantification of the air dispersed urediniospores of P. pachyrhizi are limited (DEL PONTE et al., 2006;BARNES et al., 2009;ISARD et al., 2011;NASCIMENTO et al., 2012).Iamamoto (2008) using a spore trap in soybean field in central Brazil, stated that was possible to anticipate the chemical control of the soybean Asian rust in 14 to 15 d.This author reported that air weekly collected spores from 100 to 600 could result in a rust outbreak.
In some places, qualitative studies are made, and the most common airborne fungi spore assessed are anamorphic fungi and a much smaller amount are Uredinales-like spore (AHER et al., 2015).Other studies, beyond qualitative evaluations, make quantitative monthly assessments, and, sometimes, hourly collected spore quantifications (OLIVEIRA et al., 2009b;ALMAGUER et al., 2013).The highest concentrations of daily collected spores are around 9:00 am and 7:00 pm (GRANKE et al., 2014;ALMAGUER et al., 2013;FALL et al., 2016).Thus, the objective of this work was to quantify the hourly and daily collected airborne urediniospores of P. pachyrhizi for two growing soybean seasons.

MATERIAL AND METHODS
The experimental trials were conducted at the Experimental Station of Biology of the 'Universidade de Brasília', Brasília, DF, Brazil.Local coordinates are 15º44´07.59"South latitude and 47º52´56.75"West longitude and an average altitude of 1009 m.Before planting, the acidity of the soil was corrected with dolomitic limestone applications, following technical recommendations (EMBRAPA, 2006).The tillage was done with a disk plowing and twice levelling harrow in the area.The mineral fertilizer used was the 4-30-16 (NPK) following technical recommendations (EMBRAPA 2006).The seeds were planted manually to a depth of 4 (±1) cm.The population of soybean was 250,000 plants ha -1 (10 plants m -1 ), and the spacing between rows of 40 cm.Seed were treated with thiamethoxan (Cruiser TM -35 g a.i. 100 kg -1 of seed.)+ (metalaxyl+fludioxinil) (Maxim XL TM -1.0 + 2.5 g a.i. 100 kg -1 of seed) and inoculated with 6 x 10 5 CFU of Bradyrhizobium japonicum seed -1 .Respectively, in the first and second experiments were used the fungicides Tetraconazole (Domark 100 CE) and Epoxiconazole+Pyraclostrobin (Opera®) (0,5 L ha - 1 ).The application of fungicides was made with a backpack sprayer (20L -Kingfisher PS4020 20 Liter Backpack Sprayer -Yellow).Fungicide application started with the observation of first symptoms of disease (V3/V4 -29/11/06; V7/V8 -27/12/07) (Table 1).To evaluate the variation of urediniospores and the progress of the rust on the field, a two-year work has been made in a 600m 2 (30x20m) soybean (cv.MSoy-8001) planting area.This area was subdivided 24 plots of 25m 2 (5x5m plot -1 ) for disease evaluation.
The Burkard 7-day-spore-trap (Burkard Scientific Ltd PO Box 55, Uxbridge, Middlesex, UB8 2RT, UK) was set in the central portion of the experimental area.The trap has been set on metal support, and the aspiration orifice (2x14 mm) was positioned the 1.90 m in height from the ground.The air suction rate was 10 L/min (14.4 m 3 day -1 ).Inside the trap there was and aluminum drum rotating at a speed of 2 mm h -1 , which contained a clear plastic tape (Melinex, Burkard Ltd., Rickmansworth Hertfordshire, England) covered with a thin layer of silicon grease that was placed around it to capture the spores.The tape set to trap drum was replaced weekly.
Daily and hourly readings of the number of captured were taken every seven days, with seven slides for microscopy (one for each day of the week), with 48 mm tape piece per slide.In each 2mm tape (1 hour of collection) was counted the number of spores with an optical microscope (400x).Therefore, was quantified the number of spores collected per hour and day of the week and the daily concentration of urediniospores m -3 of air, as described by Lima et al. (2009).
The severity of the rust was estimated by the percentage of diseased leaf area (% RLA) of each leaflet examined visually with a stereomicroscope.Three fully developed leaves from the inferior plantthird were randomly collected per plot in each evaluation date.In the center of the planting area was installed a weather station (Metos Compact, Pessel Instruments) with climatic sensors: air temperature (°C) and relative humidity (%) (Temperature and Relative Humidity Probe Model A660610), which recorded the average, maximum and minimum hourly temperatures and humidity on the air; precipitation (mm) (Tipping Bucket Rain Model M523CD); leaf wetness (minutes) (leaf wetness sensors Model IM521CD).The sensors were connected to the automatic data collection station.The data collection was done by transferring the information contained in the data platform for a laptop.Wind speed (m s -1 ) data were collected from a Meteorological Station 200m far from the experiment.Correlation tests were carried out among the temporal collected urediniospores, disease severity, and climatic variables, using the Spearman rank correlation coefficient.The number of spores from a day was correlated to the percentage of damaged leaf area climatic and to the climatic variables the same day, as well as to the values taken up to 10 days before the rust evaluation.

DISCUSSION
Aerial spores of P. pachyrhizi might have a significant role in this fungal spread and distribution (HARTMAN;HAUDENSHIELD, 2009).The number of collected airborne spores varies from place to place and due to the hour of the day (OLIVEIRA et al., 2009a;2009b).Most of these seasonal variations are associated to meteorological variables.Isard et al. (2005) has affirmed the role of environmental factors such as daylight, temperature, and humidity may affect the timing of spore release.Nascimento et al. (2012) stated that the number of airborne spores of P. pachyrhizi was related to the presence of soybean field crop, and there were positive correlations among the number of urediniospores, cumulative rainfall, disease intensity and favorable days for Asian rust (SAR).In addition, those authors defined the favorability to epidemic of SAR was associated to 18-26°C and 90-100% relative humidity.
However, in a study made in Manitoba, Canada, monitoring airborne-ascospores of Gibberella zeae, the maximum amount was collected at 9:00 pm (INCH et al., 2005).Correia & Costa (2005), noted that the period of spore release Lasiodiplodia theobromae occurred during the hours of 6:00 am to 10:00 am.Schuh (1993) reported that Cercospora kikuchii released spores between 4:00 am and 8:00 am and between 6:00 pm and 9:00 pm.These, daily and hourly variations of collected spores are affected by environmental and type of the fungal spore (OLIVEIRA et al., 2009a;GRANKE et al., 2014).Fitt et al. (1989), informed that the release of spores that are scattered in dry air shows a diurnal periodicity, often, the spores are collected around 12:00 pm when the wind speed is higher, the temperature is higher, and the relative humidity is lower.Conversely, Correia & Costa (2005) observed that, with rain precipitations above 80 mm, the conidia of L. theobromae in coconut tree began to be precipitated from the air, and consequently resulted in the decrease of spores captured.
From 9:00 pm to 8:00 am, few spores of P. pachyrhizi were collected; this fact might be associated to the moisture on leaf (BECK et al., 2006).Therefore, the spore release would not start before leaf wetness dried out (BECK et al., 2006).Boudrot et al. (2016) reported that a high frequency of Hemileia vastatrix urediniospores capture occurred from 7:00 am to 7:00pm, and very little from 8:00 pm to 6:00 am.These authors also showed a strong effect of the hour of the day on the dispersal of urediniospores, and, this diurnal pattern of pathogen dispersal where the highest dispersal occurred late in the morning and in the early afternoon was linked to the decrease in relative humidity and the highest recorded wind speed.Vuorinen & Helander (1995) found that airborne collected urediniospores of Melampsoridium betulinum showed a circadian variation with a peak was observed from12:00 am to 4:00 pm.As in the present study, these authors also found that a high wind speed (Figure 4) significantly increased airborne urediniospores concentration; likewise, temperature affected the spore concentration positively.

CONCLUSIONS
The maximum amount of urediniospores collected in a single was around 15,000 in 1/25/2007) and, around 12,000 in 2/13/2008.
In both soybean growing seasons most of the hourly urediniospores was collected from 10:00 am through 6:00 pm.
The highest amount of airborne-collectedurediniospores from December to March, was at 3:00 pm.
The amount of airborne spores was positively correlated to disease severity, leaf wetness, and, precipitation along both soybean growing seasons.
During a 24h-day evaluation, the major amount of hourly collected spores was negatively correlated to the leaf wetness and air relative humidity.
). Beginning seed -seeds are 3 mm long on one of the top four nodes.