Received: 27 September 2022 Revised: 30 January 2023 Accepted: 12 February 2023 Published on: 13 March 2023 DOI: 10.1111/aab.12831 ORIGINAL ARTICLE Airborne inoculum dynamics of Polystigma amygdalinum and progression of almond red leaf blotch disease in Catalonia, NE Spain Gemma Pons-Solé 1,2 | Xavier Miarnau 3 | Laura Torguet 3 | Elena Lázaro 4 | Antonio Vicent 4 | Jordi Luque 2 1 Plant Physiology Laboratory, Universitat Autònoma de Barcelona (UAB), Bellaterra, Abstract Spain The dynamics of airborne ascospores and disease progress of red leaf blotch (RLB) of 2 Sustainable Plant Protection, Institut de Recerca i Tecnologia Agroalimentàries (IRTA) almond, caused by Polystigma amygdalinum, and their correlations with weather vari- Cabrils, Cabrils, Spain ables were studied from 2019 to 2021 in two almond orchards located in Lleida, NE 3 Fruit Production Program, Institut de Recerca Spain. Airborne ascospores were detected and quantified by real-time qPCR using i Tecnologia Agroalimentàries (IRTA) Fruitcentre, PCiTAL, Park of Gardeny, species-specific primers for P. amygdalinum. Ascospores were detected mainly from Lleida, Spain April to June, with a high variability between the yearly cumulative concentrations. 4 Centre de Proteccio  Vegetal i Biotecnologia, Institut Valencià d'Investigacions Agràries Positive significant correlations were found between the weekly proportion of airborne (IVIA), Moncada, Spain ascospores and the number of wet and mild days—either combined or separated— Correspondence accumulated rainfall, number of rainy days, accumulated low temperatures on wet Gemma Pons-Solé, Plant Physiology days, and mean and maximum relative humidity. In contrast, several thermal variables Laboratory, Universitat Autònoma de (maximum temperature, VPD, and number of warm days) were negatively correlated Barcelona (UAB), 08193 Bellaterra, Spain. Email: gemmapons1992@gmail.com with ascospore catches. Positive significant correlations were found between the cumulative proportion of ascospores and RLB incidence and severity. Weekly varia- Funding information Agencia Estatal de Investigacion, Grant/Award tions in RLB incidence and severity showed significant positive correlations with the Number: PID2020-114648RR-C31; Instituto Nacional de Investigacion y Tecnología Agraria number of warm days while negative with the number of mild days. Severity was also y Alimentaria (INIA), Spain, Grant/Award positively correlated with several thermal variables (mean, maximum, and minimum Number: RTA2017-00009-C04-01 temperature, and VPD), and negatively correlated with the number of cold days and wet and mild days. Stronger correlations were generally found with ascospore catches or disease progress when using concurrent weekly weather data. Gompertz, monomo- lecular, and logistic growth models were evaluated to describe RLB disease progress. KEYWORDS almond, disease progress, epidemiology, Polystigma amygdalinum, Prunus dulcis, quantitative PCR, red leaf blotch disease 1 | I N T RO DU CT I O N almond trees (Prunus dulcis (Mill.) D.A. Webb) in the Mediterranean Basin and Middle East regions (Cannon, 1996; Farr & Rossman, 2022). Red leaf blotch (RLB), caused by the fungus Polystigma amygdalinum The disease is endemic in these regions and so far not known in other P.F. Cannon, is one of the most important foliar diseases affecting almond-growing areas, such as the United States or Australia (Farr & This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2023 The Authors. Annals of Applied Biology published by John Wiley & Sons Ltd on behalf of Association of Applied Biologists. Ann Appl Biol. 2023;183:33–42. wileyonlinelibrary.com/journal/aab 33 17447348, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12831 by Readcube (Labtiva Inc.), Wiley Online Library on [18/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 34  ET AL. PONS-SOLE Rossman, 2022). RLB is widespread in Spain (Tuset & Portilla, 1987) Saad & Masannat, 1997; Zúñiga et al., 2020), ascospore germination and its incidence has worryingly increased in recent years (Miarnau (Banihashemi, 1990; Habibi & Banihashemi, 2015; Zúñiga et al., 2020) et al., 2021). This increase has been associated to the expansion of and release (Banihashemi, 1990; Saad & Masannat, 1997; Sahragard almond-growing areas to inland regions with climatic conditions more et al., 2007), the subsequent infection and latency incubation periods suitable for the disease development, the change towards intensive (Banihashemi, 1990; Sahragard et al., 2007; Zúñiga et al., 2020), and plantations with irrigation and high tree densities, and the use of some the overall disease progress in different almond cultivars (Miarnau late-flowering cultivars (especially ‘Guara’) that are more susceptible et al., 2021). Most of these works were conducted under field condi- to RLB than traditional ones (Almacellas, 2014; Miarnau et al., 2021; tions and explored the relationships between weather variables and Ollero-Lara et al., 2019). the pathogen-related variables under study. Polystigma amygdalinum is a biotrophic ascomycete specific to Regarding the inoculum dynamics of P. amygdalinum, ascospore almond, which was firstly described in 1843 in France as Septoria production and release have been studied in Iran (Banihashemi, 1990; rubra var. amygdali Desm. from almond leaves (Desmazières, 1843). Ghazanfari & Banihashemi, 1976; Sahragard et al., 2007) and Lebanon The fungus was reclassified within the genus Polystigma and named (Saad & Masannat, 1997). According to Ghazanfari and Banihashemi for decades as P. ochraceum (Fr.) Sacc. (=P. fulvum (Pers.) DC.), until its (1976), perithecia maturation start at around 10 C in winter. identity and currently valid name were established by Cannon (1996) Banihashemi (1990) showed that ascospore release began at blooming as P. amygdalinum. This species differs from other Polystigma species (early March in the studied cultivar) and continued for 4–5 weeks, due to its host specificity, and stromal and ascospore morphological reaching the maximum at petal fall. In Lebanon, ascospore release characteristics (Cannon, 1996). occurs between February and mid-May (Saad & Masannat, 1997). Lit- First RLB symptoms usually appear in spring as pale yellowish tle information is available on P. amygdalinum ascospore production spots on both leaf sides, turning into orange-reddish and finally becom- and release in Spain. Zúñiga et al. (2020) reported on the seasonal ing dark brown. Spots size increases with time and may cover almost evolution of ascocarp maturation and production of ascospores, but the whole leaf surface in late summer. In the case of severe infections, no information on ascospore release and dispersal is available. There- early defoliation can be observed in summer (Cannon, 1996; Habibi & fore, quantitative information on the dynamics of the airborne inocu- Banihashemi, 2016; Shabi, 1997), with a subsequent decrease in lum of P. amygdalinum remains to be investigated. tree photosynthetic activity and a reduction in yield in following Thus, the main objectives of this study were: (i) to describe the  pez-Lo seasons (Lo  pez et al., 2016). The RLB disease is monocyclic dynamics of P. amygdalinum airborne ascospores over a 3-year period and the only inoculum sources are the perithecia located in the stro- in NE Spain using spore traps coupled with qPCR-based methods, mata of fallen overwintered leaves (Banihashemi, 1990; Ghazanfari & (ii) to explore the correlations between ascospore catches and Banihashemi, 1976). The sexual stage is developed in winter and asco- weather variables, (iii) to model the RLB disease progress over time, spores are released and air-spread mainly in spring to eventually infect and (iv) to explore the correlations of disease progress with weather new almond leaves under suitable weather conditions. Following infec- variables and ascospore catches. tions, leaf spots appear after a relatively long incubation period of 30– 40 days (Banihashemi, 1990; Saad & Masannat, 1997), and eventually up to 10 weeks (Zúñiga et al., 2020). The occurrence of secondary 2 | M A T E R I A L S A N D M ET H O D S infections has not been confirmed since conidia—mostly produced in summer—are not infective and presumedly act as spermatia in the sex- 2.1 | Experimental orchards ual reproduction of P. amygdalinum (Habibi & Banihashemi, 2016; Saad & Masannat, 1997). Experiments were conducted from 2019 to 2021 in two almond RLB control is based on three main strategies: cultural practices orchards located in Lleida region, NE Spain: Les Borges Blanques to reduce the inoculum source, the use of tolerant cultivars, and the (Borges hereafter; UTM coordinates: WGS84 Datum, 31T application of fungicides. The most applied cultural practices are to x = 320,870, y = 4,597,530), and Vilagrassa (x = 341,313, remove or burn the leaf litter in autumn or winter, or to favour its y = 4,612,125). The orchard at Borges is an experimental plot decomposition through urea applications (Almacellas, 2014; planted in 2009 as bare root trees with 21 cultivars grafted onto  pez-Moral et al., 2022). Regarding RLB control with Cannon, 1996; Lo ‘INRA GF-677’ rootstock, pruned as a central axis, and with a tree fungicides, in-season applications from petal fall to the end of summer spacing of 4 m  2 m. The orchard at Vilagrassa is a commercial plot have been proved to reduce infections, while evaluations on products of ‘Tarraco’, a late-blooming cultivar (Vargas et al., 2008), grafted efficacy have been conducted mostly in Iran (Banihashemi, 1990; Bayt onto ‘INRA GF 677’ rootstock, planted in 2007, pruned as open- Tork et al., 2014) and Spain (Torguet et al., 2022). vase, and with a tree spacing of 7 m  6 m. In both orchards, RLB Research on the biology of P. amygdalinum and the RLB epidemi- disease occurs naturally (Miarnau et al., 2021). Orchards were drip ology has been conducted since the early 1970 decade in several irrigated, and pruning, soil management, and fertilization were based countries including Iran, Israel, Lebanon, and Spain, among others. The on the Spanish Integrated Production Management practices main research topics included the study of the ascocarp development (BOE, 2002). No fungicide treatments were applied during the and maturation (Banihashemi, 1990; Ghazanfari & Banihashemi, 1976; experimental period in either of the two orchards. 17447348, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12831 by Readcube (Labtiva Inc.), Wiley Online Library on [18/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License  ET AL. PONS-SOLE 35 2.2 | Airborne inoculum dynamics 2.4 | Weather data A Hirst 7-day recording volumetric spore trap (Burkard Manufacturing Daily weather data (accumulated rainfall, mean and maximum rela- Co. Ltd., Rickmansworth, Hertfordshire, England) was placed in a cen- tive humidity, and mean, minimum, and maximum temperature) tral position in each orchard, under ‘Tarraco’ trees, a highly suscepti- were monitored during the experimental period in both orchards ble cultivar to RLB (Miarnau et al., 2021). Spore traps operated with by automatic weather stations belonging to the Meteorological Ser- 1 an airflow of 10 L min from mid-February (weeks 7–8) to mid- vice of Catalonia (MeteoCat; https://ruralcat.gencat.cat/web/guest/ September (weeks 36–37) 2019–2021. This experimental period cor- agrometeo.estacions). The meteorological station at Borges was responds, in the case of ‘Tarraco’ cultivar, from BBCH phenological located about 250 m away from the orchard. The nearest weather stage 51–81 (Sakar et al., 2019). station to the Vilagrassa orchard was located about 6 km away, in Airborne particles were captured on a Melinex ® 200 gauge Tàrrega (x = 347,015, y = 4,614,430). Additional derived variables (TEKRA, New Berlin, WI, USA) clear plastic tape, rotating at were calculated, coming to a total of 14 weather variables (Table 1). 1 2 mm h , which was previously coated with a layer of silicone solu- These variables were summarized on a weekly basis (Table S1). tion (Lanzoni, Bologna, Italy). Tapes were replaced weekly and cut into 1-day sections to estimate the daily airborne ascospore concen- tration of P. amygdalinum, according to the molecular methods 2.5 | Data analysis described by Zúñiga et al. (2018). Total DNA was extracted from each daily tape section using the E.Z.N.A® Plant DNA Kit (Omega The correlations of current and past weather conditions with the Bio-Tek, Norcross, GA, USA). Next, a real-time quantitative PCR ascospore catches of P. amygdalinum in the study period 2019–2021 (qPCR) in a Stepone™ Real-Time PCR System thermal cycler (Life were evaluated. Spearman's correlation coefficients (ρ) were calcu- Technologies, Carlsbad, CA, USA) was conducted, using the lated between the weekly proportion of ascospores caught by the P. amygdalinum specific primer pair PamyI2F4/PamyI2R2 and the spore trap and the values of the 14 weather variables for the same optimized reaction conditions described by Zúñiga et al. (2018). concurrent week, as well as for 1 and 2 weeks before. Spearman's Three technical replicates for each biological sample were analysed, correlation coefficients were also calculated between these weather and a sterile HPLC-quality water template was used as a negative variables and the weekly increase in RLB incidence and severity in control. The ascospore quantification was done by fitting the 2021, as well as between RLB incidence and severity levels and the obtained Cq values to a standard curve formula (with inter- cumulative proportion of ascospores. Additionally, Spearman's corre- cept = 34.26 and slope = 3.349), obtained previously by Zúñiga lation coefficients were calculated among weather variables et al. (2018) from a dilution series of a P. amygdalinum ascospore sus- pension placed on plastic tapes. Daily concentration of ascospores TABLE 1 Name and description of the studied weather variables. was then calculated considering the daily air volume sunk by the spore trap and expressed as ascospores m 3 day 1 . For the statisti- Variable Description cal analysis, we calculated the weekly proportion (0–1) of ascospores Temp.mean Mean daily temperature ( C) based on the cumulative concentration of ascospores caught by the Temp.max Mean maximum daily temperature ( C) spore traps in each orchard and year. Temp.min Mean minimum daily temperature ( C) Rh.mean Mean daily relative humidity (%) Rh.max Mean maximum daily relative humidity (%) 2.3 | Disease progress Rain Accumulated rainfall (mm) Rain.days Number of rainy days (with Rain ≥0.2 mm) Disease symptom progression was evaluated on three ‘Tarraco’ trees VPD Accumulated vapour pressure deficit (hPa), located nearby the spore trap in each orchard. RLB incidence and calculated as described by Rossi et al. (2009): severity were evaluated on a weekly basis, from March to September VPD = (1 Rh.mean/100)  6.11  exp 2021, corresponding from BBCH phenological stage 59 to 81 in the [(17.47  Temp.mean)/(239 + Temp.mean)] studied cultivar. On each tree, 50 fully expanded leaves were arbi- Wet.days Number of wet days (with VPD ≤0.4 hPa or trarily selected from new shoots, at different heights and orientations Rain ≥0.2 mm) on the outer canopy, and visually assessed. Disease incidence and Low.temp.wet Accumulated low temperatures in wet days, measured as the sum of (50 – Temp.mean) only in severity were evaluated according to procedures described by wet days Miarnau et al. (2021). RLB incidence was recorded as the percentage Cold.days Number of cold days (with Temp.mean < 10 C) of leaves showing at least one identifiable RLB lesion regardless of Mild.days Number of mild days (with Temp.mean ≥ 10 C its size, whereas RLB severity was estimated from the mean percent- and < 20 C) age of affected leaf surface, by classifying leaves into the following Warm.days Number of warm days (with Temp.mean ≥ 20 C) category classes: class 0 (0% affected leaf surface), 1 (1%–10%), Wet.mild.days Number of days both wet and mild 2 (11%–20%), 3 (21%–50%), and 4 (>50%). 17447348, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12831 by Readcube (Labtiva Inc.), Wiley Online Library on [18/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 36  ET AL. PONS-SOLE F I G U R E 1 Weekly quantifications of airborne ascospores of Polystigma amygdalinum (black bars) in Vilagrassa and Borges orchards from 2019 to 2021. Grey bars, solid lines, and dashed lines depict the weekly accumulated rainfall, mean temperature, and mean relative humidity, respectively. (Figure S1). All correlation analyses were implemented using the yt = K/[1 + y0  exp( r  t)] for logistic, in which yt is disease inci- function rcorr of the ‘Hmisc’ package in the R software version 4.0.2 dence at a given time, y0 is initial disease incidence, r is disease (R Core Team, 2020). increase rate, K is disease maximum incidence, and t is time in days. Disease growth models (namely, Gompertz, monomolecular, and To check models' goodness-of-fit, the pseudo-coefficient of determi- logistic) were fitted separately for each orchard using nonlinear nation (pseudo-R2), the residual standard error (RSE), and Lin's concor- regression for RLB incidence mean data against days after disease dance correlation coefficient (CCC) (Lin, 1989) were calculated. All onset. Models' equations were yt = K  exp[ y0  exp( r  t)] analyses were implemented using the function fit_nlin2 of the R pack- for Gompertz, yt = K[1 y0  exp( r  t)] for monomolecular, and age ‘epifitter’ (Alves & Del Ponte, 2021). 17447348, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12831 by Readcube (Labtiva Inc.), Wiley Online Library on [18/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License  ET AL. PONS-SOLE 37 3 | RESULTS 2021; and accumulated rainfall in the experimental period was 162.7 mm in 2019, 372.3 mm in 2020, and 183.6 mm in 2021 3.1 | Airborne inoculum dynamics (Table S1). We generally found stronger correlations with the weekly propor- We detected and quantified P. amygdalinum airborne ascospores in tion of ascospores when considering the concurrent weather data, as both orchards for the three monitored years (Figure 1). The earliest compared to those involving past weather data from 1 or 2 weeks seasonal detections were on 17 February 2021 in both orchards. before (Figure 3a and Table S2). We observed that ascospore catches The latest seasonal detections were on 15 September 2019 and were significantly correlated ( p < .05) with 11 of the 14 studied 3 September 2021 for Borges and Vilagrassa, respectively. In Bor- weather variables with concurrent data. Positive significant correla- ges, the yearly cumulative concentrations of ascospores in 2019, tions were found between the weekly proportion of airborne asco- 3 2020, and 2021 were 1042.5, 505.9, and 29.5 ascospores m , spores and the number of wet and mild days (either combined respectively; while in Vilagrassa were 6043.8, 580.5, and 110.5 or separated), the accumulated rainfall, the number of rainy days, 3 ascospores m . Considering the weekly cumulative proportion of accumulated low temperatures on wet days, and the relative ascospores (Figure 2), in Vilagrassa most ascospores (>75%) were humidity (mean and maximum), in decreasing order of strength collected between weeks 20 and 28, that is, from May to July, in (0.46 ≥ ρ(df = 172) ≥ 0.26, all p < .01). Negative significant correla- the monitored years. In Borges, the period of major ascospore tions were found for accumulated vapour pressure deficit (VPD), the catches was more extended than in Vilagrassa, and the 75% of asco- number of warm days, and maximum temperature, in decreasing order spore catches was reached between weeks 28 and 35 (from July to of strength ( 0.24 ≤ ρ(172) ≤ 0.17, all p < .05). Several thermal vari- August) in 2019 and 2020. However, in Borges in 2021, 85% of ables (mean, maximum, and minimum temperature, VPD, and number ascospores were already caught in week 13, corresponding to the of warm days) were highly positively intercorrelated (Figure S1). end of March. In Borges, mean weekly temperatures of the experimental period each year ranged from 7.5 to 27.5 C in 2019, from 7.7 to 3.2 | Disease progress 27.9 C in 2020, and from 9.2 to 26.8 C in 2021; mean weekly rela- tive humidity ranged from 44.3% to 72.6% in 2019, from 52.4% to First RLB symptoms in 2021 were observed on 2 June (week 23, 84.3% in 2020, and from 50.9% to 87.9% in 2021; and accumulated Figure 4) in both orchards. The main increase in disease incidence rainfall in the experimental period was 110.3 mm in 2019, 263.9 mm was observed between weeks 25 and 29 (mid-June to mid-July, in 2020, and 143.8 mm in 2021. In Vilagrassa, mean weekly temper- BBCH phenological stage 81). RLB incidence and severity reached atures of the experimental period each year ranged from 9.0 to higher values in Vilagrassa, with maximum values of 95% and 28.7 C in 2019, from 7.9 to 29.0 C in 2020, and from 8.3 to 27.6 C 46% respectively, than in Borges, with maximum values of 83% and in 2021; mean weekly relative humidity ranged from 34.0% to 70.3% 24%. In both orchards, we occasionally registered a decrease in in 2019, from 41.6% to 83.6% in 2020, and from 42.0% to 85.7% in RLB incidence and severity from mid-July on. Positive significant F I G U R E 2 Weekly cumulative proportion of Polystigma amygdalinum airborne ascospores in Borges and Vilagrassa orchards within the season's growth periods of 2019–2021. 17447348, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12831 by Readcube (Labtiva Inc.), Wiley Online Library on [18/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 38  ET AL. PONS-SOLE F I G U R E 3 Spearman's rank correlation coefficients between the values of 14 weather variables for the current week, 1 week before (Previous 1), and 2 weeks before (Previous 2) and (a) the weekly proportion of Polystigma amygdalinum ascospores caught within season's growth periods of 2019–2021, (b) the weekly increase in the red leaf blotch incidence, and (c) severity in 2021. ** and * indicate significance at p < .01 and 0.05, respectively; ns indicates no significance. correlations were found between the cumulative proportion of ascospores, and RLB incidence and severity (ρ(62) = 0.81 and ρ(62) = 0.82, respectively, both p < .01). Regarding the association between either current and past weather conditions and the weekly increase in RLB incidence (Figure 3b and Table S2) or severity (Figure 3c, Table S2), we gener- ally found stronger correlations when using the concurrent weather data in the calculations. We observed that the weekly increase in RLB incidence and severity showed significant positive correlations with the number of warm days (ρ(62) = 0.31 and ρ(62) = 0.48, respectively, all p < .05), and negative correlations with the number of mild days (ρ(62) = 0.29 and ρ(62) = 0.46, respectively, all p < .05). Significant correlations were stronger for the increase in RLB severity than for RLB incidence. Additionally, the weekly F I G U R E 4 Seasonal disease progress of the almond red leaf increase in RLB severity was positively correlated with temperature blotch from June (detection of the first symptoms) to September 2021 in Borges and Vilagrassa, expressed as the mean incidence and (minimum, mean, and maximum) and accumulated VPD, in decreas- severity values (±standard errors). ing order of strength (0.44 ≥ ρ(62) ≥ 0.35, all p < .01), being all 17447348, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12831 by Readcube (Labtiva Inc.), Wiley Online Library on [18/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License  ET AL. PONS-SOLE 39 T A B L E 2 Parameter values and goodness-of-fit indexes of the models used to describe the progress of the red leaf blotch disease incidence in Borges and Vilagrassa almond orchards in 2021. Orchard Models Estimated parametersa Goodness of fitb y0 r K CCC pseudo-R2 RSE Borges Gompertz 0.105 (0.034) 0.038 (0.007) 0.825 (0.046) 0.972 0.946 0.061 Monomolecular 0.071 (0.045) 0.018 (0.005) 0.938 (0.096) 0.974 0.948 0.060 Logistic 0.205 (0.044) 0.031 (0.008) 1 (0.152) 0.945 0.899 0.085 Vilagrassa Gompertz 0.000 (0.000) 0.217 (0.049) 0.881 (0.016) 0.987 0.977 0.055 Monomolecular 0.061 (0.080) 0.047 (0.009) 0.930 (0.040) 0.961 0.925 0.090 Logistic 0.096 (0.054) 0.107 (0.030) 1 (0.042) 0.939 0.931 0.120 a y0 : initial disease incidence; r: rate of disease increase; K: maximum disease incidence. Standard errors of the estimated parameters are in parentheses. b CCC: Lin's concordance correlation coefficient; pseudo-R2: pseudo-coefficient of determination; RSE: residual standard error. F I G U R E 5 Gompertz, logistic and monomolecular curves (lines) fitted to the red leaf blotch disease progress data (dots) after disease onset in Borges and Vilagrassa in 2021. these thermal variables highly intercorrelated (Figure S1); and 4 | DI SCU SSION negatively correlated with the number of cold days and the number of wet and mild days (ρ(62) = 0.31 and ρ(62) = 0.25, respectively, We have studied the dynamics of airborne ascospores of both p < .05). P. amygdalinum and the red leaf blotch disease progression in north- Regarding disease growth models, parameter values and model fit eastern Spain. To the best of our knowledge, this is the first time that statistics are shown in Table 2, whereas disease progress and equation the specific qPCR protocol developed by Zúñiga et al. (2018) to detect fitting are plotted in Figure 5. We found that Gompertz, monomolecu- and quantify P. amygdalinum ascospores is implemented in a field lar, and logistic growth models presented high goodness-of-fit when study using volumetric air samplers. This molecular method offers describing disease progress in both orchards, with Lin's concordance advantages over the traditional microscopic morphological quantifica- correlation coefficients (CCC) higher than 0.94, residual standard tion of airborne plant pathogens, including shorter processing times errors (RSE) lower than 0.12, and pseudo-R2 higher than 0.90 in all and higher specificity and sensitivity for fungal identification and cases (Table 2). In these models, growth epidemic rate (r) was higher quantification (Dung et al., 2018; González-Domínguez et al., 2020; in Vilagrassa (0.05 ≤ r ≤ 0.22) than in Borges (0.02 ≤ r ≤ 0.04), while Marimon et al., 2020). initial disease incidence (y0) was lower in Vilagrassa than in Borges Previous data about P. amygdalinum inoculum development in ( 0.06 ≤ y0 ≤ 0.10, as compared to 0.07 ≤ y0 ≤ 0.21, respectively) Spain showed that ascospores were potentially available from January (Table 2, Figure 5). to August (Zúñiga et al., 2020). Consistent with this finding, our 17447348, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12831 by Readcube (Labtiva Inc.), Wiley Online Library on [18/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 40  ET AL. PONS-SOLE results showed that ascospores of P. amygdalinum can be released, with variables related to water availability (maximum relative humidity depending on the location and the year, from February to mid- and accumulated rainfall) and temperatures ≥20 C during the previous September. In other studies conducted in Iran and Lebanon, where fall, mainly in October, and the accumulated rainfall in January. In the ascospores were identified and counted by microscopy techniques, a present study, we found that the weekly proportion of caught asco- main period from March to April was reported for ascospore release spores was correlated with several weather variables of the concur- (Banihashemi, 1990; Saad & Masannat, 1997). Our results describe a rent week. Some thermal variables (maximum temperature, VPD, and wider period of airborne ascospore detection, which could be the number of warm days) were negatively correlated with ascospore explained by the different geographical conditions but also by the catches, while the number of mild days (with mean temperatures ≥10 more sensitive qPCR-based methods used in our study. Most of the and <20 C) and the variables related to humidity and rain were posi- ascospores caught in our study (>75%) were already detected from tively correlated. Thus, in general terms, P. amygdalinum inoculum May to July in Vilagrassa, and from July to August in Borges, which release may benefit from mild temperatures and the hydration of in both cases is later than reported in Iran (Banihashemi, 1990; fallen leaves. This is consistent with Torguet et al. (2022) observa- Saad & Masannat, 1997). The wider period of ascospore detection in tions, who found that when fungicide applications were conducted Borges, compared to Vilagrassa, could be due to the higher mean rel- after >15 mm of rain with 10–15 C as minimum temperature, their ative humidity recorded between weeks 27 and 35, corresponding efficacy against RLB was comparable to calendar-based treatments, to July and August, in the period 2019–2021 (about 10%–15% thus suggesting that these conditions are favourable for infections. As higher than in Vilagrassa), that could have extended the period for similar to our findings, mild temperatures and moisture have been inoculum depletion to the end of summer. Nevertheless, most of the pointed as beneficial for ascospore release in other monocyclic asco- ascospores in Borges in 2021 were detected before mid-April, which mycete pathogens, such as Plurivorosphaerella nawae (Hiura & Ikata) is still consistent with previous findings (Banihashemi, 1990; Saad & O. Hassan & T. Chang (Vicent et al., 2011) or Taphrina deformans Masannat, 1997). (Berk.) Tul. (Rossi et al., 2007), causal agents of the persimmon circular High variability was observed between the yearly cumulative con- leaf spot and the peach leaf curl, respectively. centration of ascospores across the studied orchards and years, with Regarding RLB disease progress, in 2021 the first symptoms were decreasing values from 2019 to 2021. This could be due to multiple observed at the beginning of June. This is consistent with previous factors, including the RLB incidence in preceding years, the amount of studies in the same region and with the same cultivar ‘Tarraco’ where, fallen infected leaves from the previous season, or the weather condi- depending on the year, first RLB symptoms developed between the tions recorded within the period from October to January in the previ- beginning of May and the first half of June (Miarnau et al., 2021), and ous season, which influence ascospore production and maturation also coincident with additional previous reports on the disease pro- (Zúñiga et al., 2020), among other factors. No information has been gress (Banihashemi, 1990; Saad & Masannat, 1997). In our study, RLB found on the relationship between the amount of P. amygdalinum incidence and severity progressively increased between mid-June to inoculum and the ascospore production and release in the following mid-July. Afterwards, disease intensity remained relatively stable, with year. Regarding the weather conditions, the accumulated rainfall in some fluctuations that could be due to an early defoliation associated 2020, from April to June, was well over the average records (146% with the disease and the emergence of new leaves (Miarnau higher than the equivalent average for the decade 2010–2019; own et al., 2021). The positive correlation (ρ(62) > 0.80) between RLB inci- elaboration from MeteoCat database at: https://ruralcat.gencat.cat/ dence/severity and the accumulated proportion of ascospores con- web/guest/agrometeo.estacions). It is known that convective rain epi- firms the relationship between airborne ascospore catches and sodes may wash-out airborne particles (Gatz & Dingle, 1970; Rodrí- disease progress. guez-Solà et al., 2022), thus resulting in lower ascospore captures in Our study showed positive correlations between the weekly 2020. This might consequently result in less infections in 2020 (not increase in RLB incidence/severity and the number of warm days evaluated in this study) and reduce the inoculum and ascospore cap- (with mean temperatures ≥20 C) in the concurrent week, while nega- tures for the next season (2021). It is clear from this study that the tive correlations were found with the number of mild days (with mean relationship between the potential inoculum (i.e. infected fallen temperatures ≥10 and <20 C). Disease severity was also positively leaves) and the airborne ascospore dynamics of the following season correlated with the thermal variables mean, maximum, and minimum should be investigated in further research. Nevertheless, in the corre- temperature, and VPD, and negatively correlated with the number of lation analyses we addressed the variability in the yearly cumulative cold days and both wet and mild days. Thus, our results suggest that concentration of ascospores by calculating the proportion (0–1) of RLB symptom expression would benefit from concurrent warm tem- ascospores at weekly intervals. peratures in the studied region, in contrast with ascospore release, Few studies have addressed the relationship between weather that would depend on mild temperatures (10 ≤ T < 20 C) and water- conditions and inoculum availability of P. amygdalinum. Ghazanfari related variables. Miarnau et al. (2021) found that the annual RLB inci- and Banihashemi (1976) reported that temperatures <10 C in fall and dence was positively correlated with accumulated rainfall in spring, winter are a prerequisite for ascocarp formation and development in while it was negatively correlated with high temperatures in spring P. amygdalinum, while Zúñiga et al. (2020) found that seasonal accu- and summer. Zúñiga et al. (2020) suggested that infections occur in mulated amounts of available ascospores were positively correlated spring and summer when mean temperatures are in the range of 10– 17447348, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12831 by Readcube (Labtiva Inc.), Wiley Online Library on [18/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License  ET AL. PONS-SOLE 41 20 C. These previous studies are consistent with our results, as we OR CID found similar correlations of the same weather variables but with con- Gemma Pons-Solé https://orcid.org/0000-0001-6582-6987 current ascospore catches, which determine subsequent infections Xavier Miarnau https://orcid.org/0000-0001-5314-6722 and consequently, after the incubation period, the disease expression. Laura Torguet https://orcid.org/0000-0002-3804-7803 Gompertz, monomolecular and logistic growth models presented Elena Lázaro https://orcid.org/0000-0003-3821-7769 all high goodness-of-fit when describing disease progress in both Antonio Vicent https://orcid.org/0000-0002-3848-0631 orchards. The monomolecular model has been often used to describe Jordi Luque https://orcid.org/0000-0003-0751-9692 monocyclic diseases (Madden et al., 2007; Pfender, 1982; Vicent et al., 2012), as RLB is. Both Gompertz and logistic models have simi- RE FE RE NCE S lar sigmoidal curve shapes and are often used to describe polycyclic Almacellas, J. (2014). Síntomas, daños y métodos de control de la mancha epidemics (Madden et al., 2007). Nevertheless, it is not possible to ocre del almendro. Vida Rural, 389, 28–32. Alves, K. S., & Del Ponte, E. M. (2021). Analysis and simulation of plant infer the nature of a disease from disease progress curve fitting disease progress curves in R: Introducing the epifitter package. (Pfender, 1982; Xu, 2006). In the case of the RLB, the only inoculum Phytopathology Research, 3, 22. https://doi.org/10.1186/s42483-021- sources are the perithecia located in the stromata of fallen leaves 00098-7 (Banihashemi, 1990; Zúñiga et al., 2020). Using Gompertz, monomo- Banihashemi, Z. (1990). Biology and control of Polystigma ochraceum, the cause of almond red leaf blotch. Plant Pathology, 39, 309–315. lecular or logistic equations, growth epidemic rate (r) was higher in https://doi.org/10.1111/j.1365-3059.1990.tb02508.x Vilagrassa than in Borges, while initial disease incidence (y0) was Bayt Tork, D., Taherian, M., & Divan, R. (2014). Evaluation of some fungi- lower, regardless of the fitted model. This can be explained because cides for controlling almond red leaf blotch (Polystigma amygdalinum). the first RLB register in Borges had a higher incidence than in Vila- International Journal of Advanced Biological and Biomedical Research, 4, 1011–1016. grassa, although the progression of the disease was later slower in the Bergamin-Filho, A., & Amorim, L. (2002). Doenças com período de case of Borges. Considering RLB as a monocyclic disease, we can then incubação variável em função da fenologia do hospedeiro. Fitopatolo- emphasize the importance of RLB control by managing the inoculum gia Brasileira, 27, 561–565. https://doi.org/10.1590/S0100- source (e.g. through cultural practices) or by predicting its production, 41582002000600001 release and dispersal dynamics, as the seasonal disease progress BOE. (2002). Real decreto 1201/2002, de 20 de noviembre, por el que se reg- ula la produccion integrada de productos agrícolas. Boletín Oficial del depends on tree infections coming from inoculum produced in previ- Estado. ous epidemics or in epidemics at other nearby locations (Bergamin- Cannon, P. F. (1996). Systematics and diversity of the Phyllachoraceae Filho & Amorim, 2002; Madden et al., 2007). associated with Rosaceae, with a monograph of Polystigma. Mycologi- The results of the current research increase the knowledge of cal Research, 100, 1409–1427. https://doi.org/10.1016/S0953-7562 (96)80072-4 P. amygdalinum biology and the epidemiology of RLB of almond. The Desmazières, J. B. H. J. (1843). Dixième notice sur quelques plantes cryp- results reported here will contribute to the development of a predic- togames, la plupart inédites, récemment découvertes en France, et tive model for the periods of airborne inoculum availability in the que vont paraître en nature dans la collection publiée par l'auteur. region of Catalonia (NE Spain), that could be incorporated into a deci- Annales des Sciences Naturelles, Séries, 2(19), 335–373. Dung, J. K. S., Scott, J. C., Cheng, Q., Alderman, S. C., Kaur, N., sion support system to make RLB management programs more effi- Walenta, D. L., Frost, K. E., & Hamm, P. B. (2018). Detection and quan- cient and sustainable. tification of airborne Claviceps purpurea sensu lato ascospores from Hirst-type spore traps using real-time quantitative PCR. Plant Disease, ACKNOWLEDGEMEN TS 102, 2487–2493. https://doi.org/10.1094/PDIS-02-18-0310-RE Farr, D. F., & Rossman, A. Y. (2022). Fungal databases, U.S. National Fungus The authors acknowledge the technical support from IRTA staff: Collections, ARS, USDA. https://nt.ars-grin.gov/fungaldatabases/ , Jesús Ortiz, and Dr. Neus Marimon, Guillem Martínez, Robert Oro Gatz, D. F., & Dingle, A. N. (1970). Trace substances in rain water: Concen- Lourdes Zazurca. Research was funded by Instituto Nacional tration variations during convective rains, and their interpretation. n y Tecnología Agraria y Alimentaria (INIA), Spain, de Investigacio Tellus, 23, 14–27. https://doi.org/10.3402/tellusa.v23i1.10286 Ghazanfari, J., & Banihashemi, Z. (1976). Factors influencing ascocarp with grant RTA2017-00009-C04-01, and Agencia Estatal de formation in Polystigma ochraceum. Transactions of the British n (AEI), Spain, with grant PID2020-114648RR-C31 Investigacio Mycological Society, 66, 401–406. https://doi.org/10.1016/S0007- (MCIN/AEI/10.13039/501100011033). G.P. is supported by a pre- 1536(76)80208-2 doctoral grant (PRE2018-085207) from AEI. J.L., L.T., and X.M. are González-Domínguez, E., Berlanas, C., Gramaje, D., Armengol, J., supported by the CERCA Program, Generalitat de Catalunya. Rossi, V., & Berbegal, M. (2020). Temporal dispersal patterns of Phaeo- moniella chlamydospora, causal agent of Petri disease and esca, in vine- yards. Phytopathology, 110, 1216–1225. https://doi.org/10.1094/ CONF LICT OF IN TE RE ST ST AT E MENT PHYTO-10-19-0400-R The authors declare no potential conflicts of interest. Habibi, A., & Banihashemi, Z. (2015). Ascospore germination and appresso- rium formation in vitro of Polystigma amygdalinum and its survival period. Iranian Journal of Plant Pathology, 51, 461–469. DATA AVAI LAB ILITY S TATEMENT Habibi, A., & Banihashemi, Z. (2016). Mating system and role of pycnidios- The data that support the findings of this study are openly available in pores in biology of Polystigma amygdalinum, the causal agent of CORA (Catalan Open Research Area. Repositori de Dades de Recerca) almond red leaf blotch. Phytopathologia Mediterranea, 55, 98–108. at https://doi.org/10.34810/data237 https://doi.org/10.14601/Phytopathol_Mediterr-16498 17447348, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12831 by Readcube (Labtiva Inc.), Wiley Online Library on [18/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 42  ET AL. PONS-SOLE Lin, I.-K. (1989). A concordance correlation coefficient to evaluate Sakar, E. H., El Yamani, M., Boussakouran, A., & Rharrabti, Y. (2019). Codifica- reproducibility. Biometrics, 45, 255–268. https://doi.org/10.2307/ tion and description of almond (Prunus dulcis) vegetative and reproductive 2532051 phenology according to the extended BBCH scale. Scientia Horticulturae,  pez-Lo Lo  pez, M., Caldero  n, R., González-Dugo, V., Zarco-Tejada, P., & 247, 224–234. https://doi.org/10.1016/j.scienta.2018.12.024 Fereres, E. (2016). Early detection and quantification of almond red Shabi, E. (1997). Disease management of the almond pathogens Glomerella leaf blotch using high-resolution hyperspectral and thermal imagery. cingulata, Polystigma ochraceum and Tranzschelia pruni-spinosae. EPPO Remote Sensing, 8, 276. https://doi.org/10.3390/rs8040276 Bulletin, 27, 479–480. https://doi.org/10.1111/j.1365-2338.1997.  pez-Moral, A., Agustí-Brisach, C., Ruiz-Prados, M. D., Lovera, M., Lo tb00668.x Luque, F., Arquero, O., & Trapero, A. (2022). Biological and urea treat- Torguet, L., Zazurca, L., Martínez, G., Pons-Solé, G., Luque, J., & ments reduce the primary inoculum of red leaf blotch of almond Miarnau, X. (2022). Evaluation of fungicides and application strategies caused by Polystigma amygdalinum. Plant Disease (in press). https://doi. for the management of the red leaf blotch disease of almond. Horticul- org/10.1094/PDIS-04-22-0957-RE turae, 8, 501. https://doi.org/10.3390/horticulturae8060501 Madden, L. V., Hughes, G., & van den Bosch, F. (2007). The study of plant Tuset, J. J., & Portilla, M. T. (1987). Principales alteraciones del almendro disease epidemics. American Phytopathological Society. https://doi. causadas por hongos en el área mediterránea española. Fruticultura org/10.1094/9780890545058 Profesional, 11, 13–17. Marimon, N., Eduardo, I., Leo n, M., Berbegal, M., Armengol, J., & Luque, J. Vargas, F., Romero, M., Clavé, J., Vergés, J., Santos, J., & Batlle, I. (2008). (2020). A qPCR-based method for the detection and quantification of ‘Vayro’, ‘Marinada’, ‘Constantí’, and ‘Tarraco’ almonds. HortScience, the peach powdery mildew (Podosphaera pannosa) in epidemiological 43, 535–537. https://doi.org/10.21273/HORTSCI.43.2.535 studies. European Journal of Plant Pathology, 158, 1005–1016. https:// Vicent, A., Bassimba, D. D. M., Hinarejos, C., & Mira, J. L. (2012). Inoculum doi.org/10.1007/s10658-020-02136-0 and disease dynamics of circular leaf spot of persimmon caused by Miarnau, X., Zazurca, L., Torguet, L., Zúñiga, E., Batlle, I., Alegre, S., & Mycosphaerella nawae under semi-arid conditions. European Journal of Luque, J. (2021). Cultivar susceptibility and environmental parameters Plant Pathology, 134, 289–299. https://doi.org/10.1007/s10658-012- affecting symptom expression of red leaf blotch of almond in Spain. 9989-1 Plant Disease, 105, 940–947. https://doi.org/10.1094/PDIS-04-20- Vicent, A., Bassimba, D. D. M., & Intrigliolo, D. S. (2011). Effects of temper- 0869-RE ature, water regime and irrigation system on the release of ascospores Ollero-Lara, A., Agustí-Brisach, C., Lovera, M., Roca, L. F., Arquero, O., & of Mycosphaerella nawae, causal agent of circular leaf spot of persim- Trapero, A. (2019). Field susceptibility of almond cultivars to the four mon. Plant Pathology, 60, 890–908. https://doi.org/10.1111/j.1365- most common aerial fungal diseases in southern Spain. Crop Protection, 3059.2011.02461.x 121, 18–27. https://doi.org/10.1016/j.cropro.2019.03.005 Xu, X. (2006). Modelling and interpreting disease progress in time. In B. Pfender, W. F. (1982). Monocyclic and polycyclic root diseases. Distin- Cooke, D. Jones, & B. Kaye (Eds.), The epidemiology of plant diseases guishing between the nature of the disease cycle and the shape of the (pp. 215–238). Springer. https://doi.org/10.1007/1-4020-4581-6_8 disease progress curve. Phytopathology, 72, 31–32. Zúñiga, E., Leo  n, M., Berbegal, M., Armengol, J., & Luque, J. (2018). A R Core Team. (2020). R: A language and environment for statistical comput- qPCR-based method for detection and quantification of Polystigma ing. R Foundation for Statistical Computing. amygdalinum, the cause of red leaf blotch of almond. Phytopathologia Rodríguez-Solà, R., Casas-Castillo, M. C., Zhang, J. J. H., Kirchner, R., Mediterranea, 57, 257–268. https://doi.org/10.14601/Phytopathol_ Alarco n, M., Periago, C., de Linares, C., & Belmonte, J. (2022). A study Mediterr-23190 on correlations between precipitation ETCCDI and airborne Zúñiga, E., Romero, J., Ollero-Lara, A., Lovera, M., Arquero, O., Miarnau, X., pollen/fungal spore parameters in the NE Iberian Peninsula. Interna- Torguet, L., Trapero, A., & Luque, J. (2020). Inoculum and infection tional Journal of Biometeorology, 66, 1173–1187. https://doi.org/10. dynamics of Polystigma amygdalinum in almond orchards in Spain. Plant 1007/s00484-022-02267-5 Disease, 104, 1239–1246. https://doi.org/10.1094/pdis-07-19-1406-re Rossi, V., Bolognesi, M., & Giosuè, S. (2007). Seasonal dynamics of Taph- rina deformans inoculum in peach orchards. Phytopathology, 97, 352– 358. https://doi.org/10.1094/PHYTO-97-3-0352 SUPPORTING INF ORMATION Rossi, V., Salinari, F., Pattori, E., Giosuè, S., & Bugiani, R. (2009). Predicting Additional supporting information can be found online in the Support- the dynamics of ascospore maturation of Venturia pirina based on ing Information section at the end of this article. environmental factors. Phytopathology, 99, 453–461. https://doi.org/ 10.1094/PHYTO-99-4-0453 Saad, A. T., & Masannat, K. (1997). Economic importance and cycle of Poly- stigma ochraceum, causing red leaf blotch disease of almond, in How to cite this article: Pons-Solé, G., Miarnau, X., Torguet, L., Lebanon. Bulletin OEPP/EPPO, 27, 481–485. https://doi.org/10.1111/ Lázaro, E., Vicent, A., & Luque, J. (2023). Airborne inoculum j.1365-2338.1997.tb00669.x dynamics of Polystigma amygdalinum and progression of Sahragard, N., Eshaghi, R., Aflaki, M. R., & Banihashemi, Z. A. D. (2007). almond red leaf blotch disease in Catalonia, NE Spain. Annals Time of fungicide application against Polystigma amygdalinum in almond based on ascospore discharge in Chahar Mahal va Bakhtiari of Applied Biology, 183(1), 33–42. https://doi.org/10.1111/ province. Iranian Journal of Plant Pathology, 43, 219–239. aab.12831