Modelling the impact of climate change on the interaction between grapevine and its pests and pathogens: European grapevine moth and powdery mildew
Highlights
► We model climate-driven changes in the interaction between grapevine and its pests. ► We combine and apply phenological models to climate change scenarios. ► Temperature affects synchrony between Lobesia botrana and Lobesia-resistant phenology. ► Future damage due to more pest generations is limited by shorter grapevine cycle. ► Combining host and pest phenology for a realistic picture of climate change impacts.
Introduction
Major shifts in temperature and changes in the seasonal pattern of rainfall distribution are currently affecting most of the world. Climatic projections suggest that these trends will continue in the coming decades, affecting both mean and extreme values of these variables (Easterling et al., 2007). In the latest report of the Intergovernmental Panel on Climate Change (IPCC), mean global temperature is estimated to increase between 1.8 and 4.0 °C (with a likely range of 1.1–6.4 °C), by the end of the present century, depending on the greenhouse gas emission scenario (Easterling et al., 2007). The combination of climate change, associated disturbances and other global change drivers is expected to exceed the resilience of many agro-ecosystems. As a consequence, climate change could substantially impact agriculture and food production (Olesen and Bindi, 2002, Fuhrer, 2003, Maracchi et al., 2005, Kang et al., 2009). The result of climatic change should not be always seen as a threat to farm productivity, especially where water is not a limiting factor. However, Olesen et al. (2011) pointed out that the perceived outcomes of climate change expected by European farmers remain mostly negative, and in particular, interviewed farmers disclosed the feeling that the risk from pests and diseases for grapevine will increase in the Alps (both north and south).
However, consequences of climate-driven changes are not easily predictable in complex agro-ecosystems, as the biology of pests and pathogens and that of their host plants are interdependent. For example, many pests/pathogens affect their host plant only during specific vulnerable periods of the plant life-cycle. This is the case of the pathogens that infect plants through their flowers, as the bacterium Erwinia amylovora, which can penetrate its hosts (e.g., apple and pear) during flowering (Thomson, 2000). Other pests might be able to attack their host throughout their growth season but cause higher damage during specific growth stages. For example, the larvae of European grapevine moth (Lobesia botrana) are less harmful during flowering (Gabel and Roehrich, 1995), but produce more damage in the post-veraison period, when they influence grey mould (Botrytis cinerea) infections (Moschos et al., 2004). Many plant species progressively increase their resistance to pests and pathogens as they age by developing “ontogenic” resistance, which may be active on the whole plant or in specific organs or tissues (Panter and Jones, 2002, Gadoury et al., 2003). For example, grape berries are reported to be susceptible to Erysiphe necator (the powdery mildew fungus) infections until soluble solids levels reach 8% (8°Brix), and the established fungal colonies are reported to sporulate until soluble solids levels reach 15% (15°Brix) (Delp, 1954, Chellemi and Marois, 1991).
Thus, in presence of pests, infestations will occur only under specific environmental conditions and only if the host plant is in a susceptible growth stage (Chakraborty et al., 2000). For pathogens, this interaction has been frequently represented by the “plant–disease triangle”, which is made up by the three elements required for the infection to develop: a susceptible host, the presence of the pathogen, and a conducive environment (Chakraborty et al., 2000).
In addition to the above described “susceptibility windows”, one should consider the duration of the productive cycle of each crop. In fact, while higher temperatures might favour the development of certain pests, they could also shorten the length of crop cycles, thus balancing out a potential increase in pest pressure. For example, higher temperatures might cause an increase in the number of generations of insect species that are able to produce several broods per year (multivoltine species). This would imply an increase in the number of reproductive events per year, leading to an increase in population, and increased levels of infestation (Yamamura and Yokozawa, 2002, Dukes et al., 2009). However, if the last generations emerge after a crop is harvested, they cannot impact crop yield, and pest population might decrease in size due to the absence of suitable food. Some pathogens are able to infect its hosts when the plants are in certain developmental stages. This means that in order to maximize their chance of infection, the life cycle of pathogen populations must be in synchrony with host development. Since climate change can influence the rate of both host and pathogen development, it could affect the development and impact of plant diseases. Some pathogen species may be able to maintain their synchrony with target host tissue, and others may become out of synchrony (Garrett et al., 2009).
While it is clear that all these factors respond to climatic variables, they might be controlled by different combinations of driving factors, or respond to their change at different rates. In order to separate these effects, we need to better assess the dynamical interactions of the host-pest/pathogen system. Indeed, meaningful projections of climate change impacts on disease/infestation pressure can be obtained only by coupling host phenology with patterns of pest development and infestation (Grulke, 2011). At present, only a few modelling studies have considered these interactions for the projection of climate change impacts on agriculture (Baker et al., 2005, Calonnec et al., 2008, Ponti et al., 2009, Gutierrez et al., 2009). In fact, most research has concentrated on the effects of climate change on either the physiology/phenology of single crops (see, for example Webb et al., 2008, Hall and Jones, 2008, Eccel et al., 2009, Caffarra and Eccel, 2011) or pests alone (see, for example Porter et al., 1991, Woiwod et al., 1997, Bale et al., 2002, Salinari et al., 2006, Estay et al., 2009). Whereas the pressure of pest/pathogen on their host plant will probably also depend on factors other than the direct effect of temperature on their development, such as genetic adaptation, the simulation of the effects of climate change on their phenological interaction is nonetheless useful to highlight possible trends in future disease/infestation patterns.
The aim of this study was to refine current assessments of climate change impacts on pest/pathogen occurrence on grapevines by simulating pest/pathogen–host interactions. This research (i) combines detailed phenological models of grapevine with phenological models of one of its key pests (Lobesia botrana, Den. and Schiff., Lepidoptera: Tortricidae) and one of its key pathogens [E. necator, (Schw.) Burr.], (ii) apply the models to climate change scenarios for a selected study area (in the eastern Italian Alps (Fig. 1), and (iii) consider potential changes in the interactions in these two systems.
The European Grapevine moth (Lobesia botrana) is one of the most noxious vineyard-pests in the European and Mediterranean areas (Delbac et al., 2010). Its larvae feed on grapevine flowers and berries, with a facultative diapause and a variable number of generations per year, depending on temperature and photoperiod (Pavan et al., 2010). It is usually reported as being trivoltine in Mediterranean areas although, in the warmest years, a fourth partial generation has been reported (Torres-Vila et al., 2004). The first adults of L. botrana appear in the spring and are shortly followed by the first generation of larvae which feed on inflorescences and buds; in Northern Italy this occurs between May and June. Subsequent generations feed on berries and usually cause considerable damage (Moschos et al., 2004). However, the sensitivity of grapevines to infestation by this pest varies during the grape growing season (Gabel and Roehrich, 1995, Pavan et al., 2009). Gabel and Roehrich (1995) compared the damage produced by larvae infestation at different growth stages on different grapevine cultivars and observed for all stages a period in which fructiferous organs (flowers and berries) were unsuitable for infestation by freshly hatched larvae, i.e. flowering and fruit set. During this “resistant” phenological window, the level of damage caused by larvae was significantly lower compared to earlier and later growth stages.
Powdery mildew (E. necator) is one of the major diseases in grapevine (Gadoury et al., 2003, Bendek et al., 2007, Caffi et al., 2011). It affects green leaves and fruit and reduces the yield of grapes and the quality of must and wine (Gadoury et al., 2001, Campbell et al., 2006). This pathogen undergoes sexual and asexual cycles during the year, overwinters as mycelium in infected buds or chasmothecia in the bark of vines (Gadoury and Pearson, 1988, Cortesi et al., 1995). The primary infections are usually caused by the ascospores; afterwards, the disease progresses during the season by asexual, secondary infection cycles driven by conidia (Carisse et al., 2009). The severity of the disease is related to the number of disease cycles per season, by air humidity (a moderately high air humidity promotes the germination of conidia) and rainfall (rain prevents germination of conidia) (Carroll and Wilcox, 2003, Bendek et al., 2007). Asexual reproduction and rate of epidemic development of powdery mildew are mainly controlled by temperature (Delp, 1954, Sall, 1980, Chellemi and Marois, 1991). The length of the “latency period” of each cycle (i.e., the time period between spore deposition of the plant surface and sporulation of the resultant colony) is the main driver of the number of disease cycles, which is in turn affected by temperature (Calonnec et al., 2008). This latency period is minimum when temperatures are within the optimal range between 20 and 28 °C (Caffi et al., 2011).
Section snippets
Study area and sites
The study area is in the pre-alpine and alpine viticultural regions along the Adige River in the central-eastern Italian Alps, and includes part of the provinces of Verona, Trento and Bolzano (Fig. 1). This mountainous area consists of a system of minor valleys converging to the largest and longest of them, the Adige Valley.
Viticulture is an important source of revenue for this region and forms part of its historical and social identity, with evidence of vine-growing dating back to the Roman
Validation of pest and host models
In the study area flowering usually occurs at the beginning of June and lasts around 10–14 days (period between beginning of flowering and fruit set) depending on the elevation. Ripening occurs between the end of August (valley floor, on warm years) and the end of September (mountain slopes, on cool years) (Table 1).
The predicting performance of the sub-models for grapevine phenological susceptibility was considered satisfactory as they yielded Mean Absolute Errors (MAEs) that were lower than
European grapevine moth
There is widespread concern that the predicted future warming will increase the pressure of insect pests and diseases on crops (Porter et al., 1991, Estay et al., 2009, Olesen et al., 2011). Increased temperatures and earlier onset of the growing season may reduce winter mortality, increase the rate of insect metabolism and development (Bale et al., 2002), and increase the number of generations of multivoltine species (Laštůvka, 2009). Recent works show that these changes are already taking
Acknowledgements
The authors wish to thank Gianfranco Anfora, Fabio Zottele, Len Coop, Friedrich Menke, Francesco Penner, Maurizio Bottura, Francesco Fellin for data supply, advice and proofreading.
References (80)
- et al.
Climate change: potential impact on plant diseases
Env. Poll.
(2000) - et al.
Larval instars determination for the European Grapevine Moth (Lepidoptera: Tortricidae) based on the frequency distribution of head-capsule widths
Crop Prot.
(2010) Agro-ecosystem response to combinations of elevated CO2, ozone, and global climate change
Agric. Ecosyst. Environ.
(2003)- et al.
Adaptive amplification: an inducible chromosomal instability mechanism
Cell
(2000) - et al.
Climate change impacts on crop yield, crop water productivity and food security – A review
Prog. Nat. Sci.
(2009) - et al.
Consequences of climate change for European agricultural productivity, land use and policy
Eur. J. Agron.
(2002) - et al.
Impacts and adaptation of European crop production systems to climate change
Eur. J. Agron.
(2011) - et al.
Age-related resistance to plant pathogens
Adv. Bot. Res.
(2002) - et al.
The potential effects of climatic change on agricultural insect pest
Agric. For. Meteorol.
(1991) Obtaining of sub-models for modeling the entire life cycle of a pathogen
J. Plant Dis. Prot.
(1980)