Effects of native biodiversity on grape loss of four …...documented to control arthropod pests in...

Web Ecol 18 15ndash27 2018httpsdoiorg105194we-18-15-2018copy Author(s) 2018 This work is distributed underthe Creative Commons Attribution 40 License

Effects of native biodiversity on grape loss of fourcastes testing the biotic resistance hypothesis

Mauro Nereu12 Ruben H Heleno2 Francisco Lopez-Nuacutentildeez2 Maacuterio Agostinho3 and Jaime A Ramos1

1MARE ndash Marine and Environmental Sciences Centre Department of Life SciencesUniversity of Coimbra Calccedilada Martim de Freitas 3000-517 Coimbra Portugal

2CFE ndash Centre for Functional Ecology Department of Life SciencesUniversity of Coimbra Calccedilada Martim de Freitas 3000-456 Coimbra Portugal3Sinergiae Ambiente Lda (Environmental Consultancy) Coimbra Portugal Eiras

Rua da Liberdade Lote 5 Loja No 1 3020-112 Coimbra Portugal

Correspondence Jaime A Ramos (jramosucpt)

Received 13 July 2017 ndash Revised 30 December 2017 ndash Accepted 10 January 2018 ndash Published 20 February 2018

Abstract Management of agricultural landscapes can influence the biodiversity and the ecological servicesprovided by these ecosystems such as natural biological pest control Viticulture is a very important economicactivity in most countries with Mediterranean climate often shaping their landscapes and culture Grape produc-tion is affected by a number of pests and diseases and farmers use prophylactic and response-driven pesticidesto control these pests Here we quantified the main biotic causes of crop losses in four grape castes two red(Touriga Nacional and Baga) and two white (Arinto and Chardonnay) and evaluated the potential effect of na-tive biodiversity to provide biotic resistance to pest outbreaks and grape losses Specifically the diversity andabundance of bird and insect communities in these vineyards were quantified and divided into functional guilds(pest neutral or auxiliary) to test whether these natural communities hold the potential to naturally controlgrape pests (biotic resistance hypothesis) under normal vineyard management (including pesticide applicationregimes) A potential association between distance to the vineyard edge and grape losses was also evaluatedWe recorded a very small proportion of grape losses (mean = 06 max= 75 ) with insect pests showing apreference for the castes Baga (red) and Chardonnay (white) while bird pests avoided the caste Arinto (white)Grape color did not influence losses caused by insect pests but birds showed a preference for red castes Thecaste Baga was also more vulnerable to losses caused by fungi Despite their low impact on grape productionmost insects and birds detected in the six vineyards were pests which entails a potentially low level of bioticresistance in this highly managed agricultural ecosystem Further research is necessary to fully evaluate the roleof functional biodiversity in vineyards particularly if alternative production processes such as organic farmingcan increase the potential of native biodiversity to protect against grape losses from pests under lower regimesof chemical spraying

1 Introduction

For agriculture to be profitable it is critical to keep pest out-breaks under control Modern agricultural systems ndash charac-terized by the use of pesticides monoculture practices andintensive use of heavy machinery ndash have been shown tonegatively affect biodiversity as well as the resistance andresilience of agro-forestry systems (Altieri 1999) In turnhigh biodiversity is critical for ecosystem resilience includ-

ing agricultural systems and is responsible for many ecosys-tem services such as pollination nutrient cycling seed dis-persal and natural pest control Agricultural areas are notstatic systems they are characterized by extreme fluctuationsin the densities of organisms and frequent outbreaks of newspecies that can rapidly become pests with relevant reduc-tions in productivity (Gurr et al 2012) In recent years thepotential role of predators parasites and pathogens to con-

Published by Copernicus Publications on behalf of the European Ecological Federation (EEF)

16 M Nereu et al Effects of native biodiversity on grape loss of four castes

trol such pests has attracted increasing attention (Gurr et al2000 Koh 2008 Johnson et al 2010 Loacutepez-Nuacutentildeez et al2017) including the evaluation of biological control agentsas a management tool to maintain pest outbreaks in vine-yards below thresholds (De Bach 1964) Alternatively natu-ral populations have also been suggested to hold an intrinsicpotential to control pest outbreaks without the intentional re-lease of specific biocontrol agents which is known as thebiotic resistance hypothesis (Heide-Jorgensen 2011) Herewe evaluate the potential of natural populations of birds andinsects to provide biotic resistance to vineyards one of themost important agricultural systems of southern Europe withlarge economic cultural and ecological relevance

Biotic resistance is the capacity of natural resident speciesto reduce the success of pests and invasive species (Levineat al 2004 Flower et al 2014) Agro-forestry ecosystemshave been recognized as important systems to test this hy-pothesis For example the role of natural enemies in the con-trol of aphid pests has been evaluated in Citrus plantations(Michaud 1999) and birds have been shown to reduce thedamages of the emerald ash borer (Agrilus planipennis) inash forests (Flower et al 2014) These effects have also beenevaluated in California vineyards where an experimental in-crease in the density of breeding birds leads to a higher con-trol of plastic sentinel larvae (Jedlicka et al 2011) howeverthe real effect of grape production has not been evaluatedand the real potential of biotic resistance in agro-forestryecosystems remains underevaluated (Buumlrgi et al 2015)

Portugal has a strong tradition in wine production (Cunhaet al 2009) and contains great concentrations of au-tochthonous grape castes more than 290 (Boumlehm et al2007) About 22 of the countryrsquos land area is occupiedby vineyards which places Portugal in first position re-garding the relative importance of this habitat in relation tothe countryrsquos area (International Organization of Vine andWine ndash OIV httpwwwoivintendatabases-and-statisticsdatabasebdd=IG) The wine sector represents an impor-tant product for the Portuguese economy estimated to beover EUR 727 million annually (httpwwwivvgovptnp4estatistica) Despite this importance the grape productivityand consequently the wine industry is largely vulnerable toa large number of pests and diseases including viruses bac-teria arthropods birds and fungi (Delaunois et al 2014)To combat these pests most wine producers worldwide relyheavily on chemical treatments (pesticides) and particularlyfungicides (Delaunois et al 2014) which can be sprayedmore than 10 times per year (Corio-Costet et al 2011) Somestudies estimate that some French vineyards possibly receiveup to 93 000 tons of fungicide per year (Viel et al 1998Niccolucci et al 2008) to control losses in productivity andwine quality due to fungi (Hocking et al 2007)

Birds and insects can have a double function regardingwine production they can directly consume the grapes thusacting as pests (Canavelli et al 2014) or they can consumespecies that attack the grapes in which case they act as bio-

control agents (Ceia and Ramos 2014 Barbaro et al 2016)For example frugivorous birds have been recorded to causesignificant economic losses to vineyards (Tracey et al 2007Canavelli et al 2014) while insectivorous birds have beendocumented to control arthropod pests in agricultural con-texts (Wenny et al 2011 Ceia and Ramos 2014) Similarlywhile insects are among the most relevant grape pests someinsects have also been shown to be beneficial for vineyardproductivity ie the auxiliary insects (Bournier 1976 Jons-son et al 2008)

Vineyards attract several pests capable of attacking all or-gans of the plant (Bournier 1976) Among the most dam-aging grape pests are the root pest Daktulosphaira vitifo-liae (Hemiptera Phylloxeridae) and the fruit-eating cater-pillars Lobesia botrana and Eupoecilia ambiguella (Lepi-doptera Tortricidae) (Bournier 1976) To control these andother arthropod pests most producers spray their vineyardswith insecticides which depress the pests along with otheradventive biodiversity potentially reducing natural biocon-trol agents

In this study we analyze whether natural biotic resistancecan reduce grape losses in the wine region of Bairrada cen-tral Portugal under the usual vineyard management (includ-ing legal pesticide application regimes) The Bairrada wineregion is one of the most emblematic regions in Portugalknown for a large diversity of soil characteristics that resultin several distinct wines including famous Champagne-likewines The Bairrada region is characterized by a relativelymoist Mediterranean climate and consists of a fragmented ru-ral landscape which often results in small vineyards largelyinfluenced by field margins and the contiguous habitats suchas stream banks and forested areas The most common redcastes in the region are the Baga Touriga Nacional and Jaenand the most common white castes are Arinto Maria Gomes(also known as Fernatildeo Pires) Bical Cerceal and Chardonnay(wwwinfovinicom) Apart from their color each caste hasunique characteristics such as grape size number of grapesper bunch sugar content or acidity (Varandas et al 2004Keller 2010)

Specifically we investigate the potential role of naturallyoccurring birds and insects both as pests and natural biocon-trol agents of four main grape castes of Bairrada differingin their morphological and physiological characteristics twored (Touriga Nacional and Baga) and two white (Arinto andChardonnay) Finally we evaluated whether the causes andmagnitude of grape losses were affected by the distance tothe vineyard edges as pests and auxiliary biodiversity mightnot use the landscape matrix equally

2 Methods

Six vineyards were selected across the Bairrada regionnamely Quinta da Aveleda Caves de Satildeo Joatildeo Caves Mes-sias Colinas de Satildeo Lourenccedilo Estaccedilatildeo Vitiviniacutecola da Bair-

Web Ecol 18 15ndash27 2018 wwwweb-ecolnet18152018

M Nereu et al Effects of native biodiversity on grape loss of four castes 17

Figure 1 Location of the sampling areas within the Bairrada re-gion the circles represent the sampling vineyards Vineyard codes1 ndash Aveleda 2 ndash Caves Messias 3 ndash Colinas de Satildeo Lourenccedilo 4 ndashCaves de Satildeo Joatildeo 5 ndash Boas Quintas and 6 ndash Estaccedilatildeo Vitiviniacuteculada Bairrada Caste codes B ndash Baga (red) T ndash Touriga Nacional(red) A ndash Arinto (white) C ndash Chardonnay (white)

rada and Boas Quintas (Fig 1) We focused on four typesof castes (ie grape varieties) very common in the regiontwo red (Touriga Nacional and Baga) and two white (Ar-into and Chardonnay) In each of the six vineyards we se-lected 10 vines from uniform patches of each of the availablecastes namely 50 Touriga Nacional (five sites) 40 Arinto(four sites) 20 Chardonnay (two sites) and 20 Baga (twosites) (Fig 1) This sampling design allowed the character-ization of 60 white vines and 70 red vines All selected vineswere separated by a minimum distance of 30 m so that lossescaused by insects and birds could be considered largely inde-pendent (Williams and Martinson 2000) Insects generallypresent a very restricted distribution and birds are also fairlyrestricted in their range when they attack a specific group ofvines (Somers and Morris 2002 Barbaro and Battisti 2011)

21 Exclusion experiment

In order to analyze the effect of natural biodiversity on grapelosses we performed exclusion experiments on the selectedvines In each vine three bunches were selected and randomlyallocated to each of the three treatments (1) exclusion ofbirds by installing a net with a broad mesh of 19times 19 mm(2) exclusion of birds and insects by installing a net with a

fine mesh of 19times 19 mm and (3) control vines accessibleto all species (no exclusion)

All nets were installed in early June when the unripe fruitswere already formed but before they started to ripen At thisstage the grapes were sufficiently robust to endure the netinstallation and were not yet attacked by insects or birds Theinitial number of grapes per bunch was recorded as well asthe linear distance from each vine to the nearest vineyardedge

By the end of August all nets were removed and the num-ber of grapes lost due to each pest type (birds insects andfungi) was scored by visual inspection of the marks left onthe grapes according to field guides and the farmerrsquos expe-rience (Isaacs et al 2003 Carisse et al 2006 Hahn andWold-burkness 2008 Hoover et al 2011 Mani et al 2014)

22 Bird census

Bird abundance in each vineyard was evaluated with fourcensuses per month in June July and August 2016 the periodof grape development Censuses were performed between0700 and 1000 at two sites with good visibility at oppo-site edges in each vineyard Each census lasted 5 min dur-ing which the horizontal distance to all birds seen or heardwas estimated Only birds within a radius of 50 m from thecensus point were used in the analyses Bird species were di-vided into three groups ndash auxiliary neutral and pest ndash accord-ing to Cramp and Perrins (1993) and our previous experiencewith the local bird diets (Cruz et al 2013 Costa et al 2014da Silva et al 2017) These categories were adjusted everymonth in order to reflect bird feeding habits in relation to theavailable resources For instance most species are largely in-sectivorous during the breeding season and consume largeamount of fruits in the end of summer

23 Insect sampling

Insects in each vineyard were sampled with 10times 25 cmyellow sticky traps (commercially available from KoppertHoriver) a broad sampling method particularly useful forflying insects (Thomson et al 2004) The traps were stickyon both sides and were suspended from the lower wire thatconnects the different vines along straight lines Five trapswere placed in each vineyard approximately 30ndash50 m apartTraps were operated once per month during the duration ofthe experiment (JunendashAugust) on each occasion they wereremoved after 5 days and kept at 4ndash5 C until insect identi-fication All insects were later extracted from the traps witha solvent identified to the family level sorted according totheir morphotypes and counted Only three families of micro-hymenoptera (Platygastridae Diapriidae and Proctotrupidae)were grouped together due to their similar morphology Likebirds all insects were divided into three functional groupsregarding their main relation with agriculture crops ie aux-iliary neutral and pests (Mani et al 2014 Bostanian et al

wwwweb-ecolnet18152018 Web Ecol 18 15ndash27 2018

Figure 2 Bird records per 5 min census in each vineyard between June and August Error bars represent the standard error The few barsthat do not have an error flag correspond to bird groups represented by a single species at that site during that period

2015) This classification was temporally flexible (ie vari-able across months) in order to reflect the changing roles ofinsects in relation to their life cycle and food availability

24 Statistical analysis

Grape losses were quantified in terms of the percentage oflosses in relation to the initial number of grapes present ineach bunch (ie losses=N grapesfinalminusN grapesinitial N

grapesinitialtimes 100)To evaluate whether grape loss by birds and insects dif-

fered among grape caste and color we used generalized lin-ear mixed models (GLMMs) with nested effects with colorand caste nested within color as fixed effects vineyard as arandom factor and distance to the edge as a co-variable We

evaluated the proportion of grape loss due to insects and birdsin the control treatment (no exclusion) with two GLMMs foreach loss type namely one to look for differences betweenthe four castes and one to look for differences between grapecolor

Significant differences between castes were explored witha general linear hypothesis followed by a Tukey multiple-comparison test This test uses a Tukey multiple-comparisontest to create confidence intervals for all pairwise differencesbetween factor-level means while controlling for the familyerror rate

To evaluate the potential of natural biodiversity to controlgrape losses we performed linear regressions between thepercentage of fruit losses per bunch (response variable) with

Figure 3 Abundance of insects per yellow sticky trap in each vineyard during the study period Error bars represent the standard error

the abundance of each insect and bird functional guilds (pestauxiliary and neutral) as explanatory variables

All analyses were performed in R 305 (R Core team2016) using packages ggplot2 Rmisc lmerTest (Kuznetsovaet al 2017) and multcomp

3 Results

31 Vineyard biodiversity

Bird abundance ndash there were no differences in bird abun-dance across vineyards (F563 = 1025 p = 0411 Fig 2) oracross months (F2177 = 0834 p = 044 Fig 2) In termsof functional diversity most birds were neutral for the grapes(ie they were not pests or auxiliary agents) The abundanceof neutral birds per census was higher in July than in the

other two months (F2171 = 759 p =lt 0001 Fig 2) Theabundance of pest and auxiliary birds did not vary signif-icantly throughout the season (F280 = 050 p = 060 andF228 = 027 p = 076 respectively) (Fig 2)

Insect abundance ndash insect abundance differed among vine-yards (F5333 = 3165 p = 0008 Fig 3) the vineyard withthe most insects was Boas Quintas (162 individuals per trap)and the vineyard with the least insects was Messias (30 in-sects per trap) There was a small non-significant trend forinsect abundance to increase over the season (F21093 = 138p = 0252) (Fig 3) In terms of functional diversity most in-sects were pests and only a few insects were neutral Theabundance of insect pests was much higher than that of theother two functional groups in the three months (F21091 =

17835 p =lt 00001)

Figure 4 Mean percentage of fruit loss due to bird fungi and insect activity per vineyard Error bars represent the standard error

32 Grape losses

Overall grape losses were low or negligible at most vine-yards (mean= 059 ) The vineyard with the highest per-centage of fruit losses was by far Messias (750 ) wherethe percentage of fruit losses by fungi reached 700 (Fig 3) All other vineyards had very few grape losses par-ticularly those of Satildeo Lourenccedilo and Aveleda where therewere no documented losses due to birds or fungi and only100 of the fruits were lost due to insect activity (Fig 4)The overall mean grape losses across all vineyards were201 due to fungi 008 due to birds and 024 due toinsects

Regarding the differences between grape castes Chardon-nay had the highest proportion of insect losses (043 )and Touriga Nacional the lowest (008 ) (Fig 5a) Whitegrapes (Arinto and Chardonnay) had slightly more losses dueto insects (070 ) than red grapes (Baga and Touriga Na-cional) (050 ) though these differences were not signifi-cant (F1119 = 098 p = 0337)

The nested GLMM showed that grape loss by insects in thecontrol bunches differed only among castes (F2119 = 974p =lt 0001) with Chardonnay and Baga showing slightlyhigher losses than both Arinto and Touriga Nacional (Fig 5aTable 1a) In relation to birds grape loss differed marginallyamong castes (F2117 = 302 p = 005) and significantly be-tween caste colors (F1119 = 838 p = 0004) with a prefer-ence for red castes (Fig 5b Table 1b)

Similarly the vulnerability to fungi differed among castes(F3365 = 6706 p = 0003) with Baga being the caste withthe greatest losses due to fungi and Touriga Nacional thecaste with the fewest losses (Fig 5c Table 1c)

33 Biodiversity and grape losses

Grape losses by insects were positively correlated with in-sect abundance (rs = 089 p = 003 Fig 6a) but not withbird abundance (rs = 054 p = 02) abundance of insectpests (rs = 009 p = 092) or auxiliary insects (rs =minus02p = 071) Grape losses by birds were correlated neither withoverall abundance of birds (rs =minus046 p = 035) or in-sects (Fig 6b rs = 002 p = 1) nor with the abundance ofpest birds (rs = 066 p = 018) or auxiliary birds (rs = 002p = 1) Finally the abundance of auxiliary birds was not sig-nificantly correlated with insect abundance (rs = 026 p =041)

The proportion of grape losses due to insects birds andfungi was independent from the distance to the vineyardedge (rs = 008 p = 038 rs =minus012 p = 020 and rs =minus006 p = 050 respectively)

4 Discussion

This study shows that most grape losses in vineyards of theBairrada region were due to fungi while losses due to birdsand insects were almost negligible in all vineyards Contraryto birds which showed a preference for red castes the in-sects did not show an overall preference for grape color butinsects caused more losses on the castes Chardonnay (white)and Baga (red) and birds avoided the consumption of Arinto(white) The preference for certain castes should be takeninto account particularly to understand why insects and birdsare more attracted and cause more damage to certain vine-yards Unfortunately organically grown vineyards were notavailable in the region and therefore chemical spraying wasused in all vineyards sampled hindering the detection of arelevant biotic resistance effect We expect that such poten-tial effects of biotic resistance conferred by natural biodiver-

Table 1 Summary of the general linear hypothesis models testing the effect of grape caste and color on fruit loss by insects birds and fungiThe models compare all caste color pairs Significant results an α le 005 are highlighted in bold

Estimate Std error z value Pr (gt |z|)

(a) Loss due to insects BagandashArinto 2124 0603 3520 0002ChardonnayndashArinto 1107 0434 2550 0049Touriga NacionalndashArinto minus0769 0512 minus1503 0422ChardonnayndashBaga minus1017 0660 minus1541 0400Touriga NacionalndashBaga minus2893 0764 minus3787 0001Touriga NacionalndashChardonnay minus1876 0588 minus3190 0007

(b) Loss due to birds BagandashArinto 2378 0831 2859 0021ChardonnayndashArinto minus0159 1027 minus0154 1000Touriga NacionalndashArinto 0537 0870 0618 0924ChardonnayndashBaga minus2534 1202 minus2108 0147Touriga NacionalndashBaga minus1839 0907 minus2028 0173Touriga NacionalndashChardonnay 0696 1021 0681 0902

(c) Loss due to fungi Baga ndash Arinto 0729 0075 9717 lt 0001ChardonnayndashArinto 0194 0151 1280 0547Touriga NacionalndashArinto minus1284 0165 minus7770 lt 0001ChardonnayndashBaga minus0535 0169 minus3164 0007Touriga NacionalndashBaga minus2012 0165 minus12161 lt 0001Touriga NacionalndashChardonnay minus1478 0224 minus6595 lt 0001

sity will be more important in the absence of pesticide treat-ments Our results suggest that the chemical use in the regionis highly effective as losses by insects and birds were neg-ligible and losses with fungi were relatively low Howeverthis form of agriculture is known to affect local biodiversityand might threaten ecosystem functions and services partic-ularly that of natural biocontrol (Geiger et al 2010) Emerg-ing agricultural practices are now aiming to reduce the envi-ronmental impacts without compromising production Thesenew practices include integrated production protocols (Periniand Susi 2004) that specifically take into account the role ofbiodiversity in agricultural areas (Mccracken 2011) Recentstudies showed that a 42 reduction in the use of pesticidesdid not compromise the production of French wine on 77 of the farms and 59 of the vineyards actually become moreprofitable after such reduction (Lechenet et al 2017)

41 Grape losses

We observed a very low proportion of grape losses bybirds and insects but other studies report a higher percent-age of grape loss attributed to these pests In an experi-ment in New Zealand Kross et al (2012) registered 35 of grapes damaged by birds and Bournier (1976) calcu-lated that crop losses in California can reach up to 15 chiefly due to Platynota stultana (Lepidoptera Tortricidae)(Bournier 1976 Kross et al 2012)

The fact that the percentage of grape losses attributed tofungi was much higher than those attributed to birds andinsects was largely influenced by the vineyards of CavesMessias Such high vulnerability of these vineyards to fungi

could be due to specific microclimatic conditions (eg highrelative humidity)

Similarly to previous studies we registered a significantinfluence of caste color on grape losses due to birds Thepreference for the red color may be explained by its attrac-tiveness to birds (Whitney 2005 Gagetti et al 2016) Thefact that insects did not show a preference for the red color inour study goes against previous studies (Takahara and Taka-hashi 2016) and may be explained by the high effectivenessof the chemical treatments in our study region

The biochemical composition of grapes is another fac-tor that influences the preference of grapes by birds and in-sects indeed our results show a significant preference forChardonnay and Baga grape castes over Arinto and TourigaNacional by insects and Baga by birds The chemical andtactile characteristics of the different castes can affect theirvulnerability to several bird and insect pests (Varandas et al2004 Belliacute et al 2007 Saxton et al 2009) but unfortu-nately we did not quantify grape composition in the presentstudy

Few studies evaluated grape losses by birds however thefew studies that tried to quantify this problem suggest thatchanges are relatively small namely lower than 5 in SouthAfrica (Dignon 2013) and lower than 9 in North America(Anderson et al 2013)

There are more studies evaluating grape losses due to in-sects although most of them focus on the effects of a singleinsect pest (Hoffman and Dennehy 1987 Moschos 2005)while others focus on losses in general like a study in Brazilwhich registered 4 of grape losses in the whole coun-

Figure 5 Percentage of fruit loss by insects (a) birds (b) andfungi (c) per caste Bars with the same letters do not differ signifi-cantly Error bars represents the standard error

try (Oliveira et al 2014) Other studies analyzed the lossescaused by Lobesia botrana in grapes and authors registered57 losses in some years (Hoffman and Dennehy 1987) Inour study insect and birds losses (142 and 008 respec-tively) were slightly lower than those reported in previousstudies which is normal considering the multiple factors af-fecting grape losses such as year localization age of thevineyard and castes

In summary there is some evidence that insects select cer-tain grape varieties due to their physiological and morpho-logical characteristics ndash such as skin thickness color and nu-trient content (Galvan et al 2008) ndash and this needs to betaken into account in order to manage vineyards more ef-fectively for example by improving the conditions for thereproduction of insectivorous birds in areas with more vul-nerable castes Other evidence is the preference of birds byred caste colors and a non-preference for Arinto This can becorrelated because Arinto is a white caste and therefore notvisually appealing to birds But as the percentage of losses

caused by birds was low we cannot draw great conclusionsabout the non-preference for Arinto

In an effort to reduce costs and the environmental footprintwithout jeopardizing productivity alternate strategies havebeen developed focusing on the economic level threshold forpesticide application and habitat management ie integratedpest management integrated production and biological pro-duction strategies (Perini and Susi 2004 Mccracken 2011)

42 Relation between losses and natural biodiversity

In contrast with our expectations we could not detect anymeasurable effect of natural biodiversity on pest suppressionWe showed that vineyards with more insects had more lossesas most insects sampled in the vineyards were pests (745 )Interestingly the proportion of auxiliary insects was alsovery low likely due to the use of pesticides which is likely tohinder their potential role as biocontrol agents Such a neg-ative relationship between insecticide toxicity and the abun-dance of biocontrol agents (spiders lacewings carabids andparasitoids) has also been observed by Thomson and Hoff-mann (2006) in Australia It should be noted that the stickytraps that we used in our study were not appropriate to sam-ple spiders which might also be relevant as biocontrol agentsin our study area

Our bird census revealed that the abundance of birdpests and auxiliary birds was similar Bird feeding behaviorchanges throughout the season namely by consuming mostlyinsects early in the season which corresponds to the breedingseason when insects are crucial dietary items for their off-spring (Herrmann and Anderson 2007) The breeding sea-son of most bird species in the study area corresponded tothe first months of our experiment (JunendashJuly) while lateron (AugustndashSeptember) they began feeding on grapes whichhad ripened In our dataset vineyards with a higher densityof bird pests had considerably more grape losses howeverthis relationship was not statistically significant due to thehigh heterogeneity of the dataset the overall low effect ofbirds and the small number of vineyards sampled Such alow impact of frugivorous birds may be a common trait ofthe vineyards of Bairrada region which are highly embed-ded within a complex landscape matrix that provide shelterbreeding sites and alternative feeding areas for many birds(Pithon et al 2016) This might contrast with the damagesdocumented in extensive vineyards in more simplified land-scapes such as those in California and Alentejo (southernPortugal) where large flocks of birds such as starlings (Stur-nus sp) can have considerably higher impacts (Stevensonand Virgo 1971 Curtis et al 1994)

43 Edge effect

We did not find any effect of the distance to the edge ofthe vineyard on the proportion of grape losses Howevermost losses were recorded in the first 100 m from the vine-

Figure 6 Relationships between the total number of insects (a) and birds (b) and the mean fruit loss caused by birds and insects pervineyard The solid regression lines represent significant relationships

yard edge In previous studies grape losses caused by birdstended to decline with increasing distances from the edge(Somers and Morris 2002) Most avian species only visit thevineyards occasionally for feeding and they rapidly look forcover in the edge habitats (Pithon et al 2016) Accordinglywe can expect that the vines closest to the edge are visitedmore often by frugivorous birds (Stevenson and Virgo 1971Somers and Morris 2002 Saxton et al 2004) thus morelosses are predicted in this area Another factor that mayaffect grape losses is type of edge which might influencethe abundance and diversity of frugivorous and insectivorousbirds thus potentially affecting grape losses Unfortunatelyfor the reasons discussed above we could not detect sucheffects in the present study

Likewise grape losses caused by insects tended to declinewith increasing distances from the edge although not signif-icantly (Hoffman and Dennehy 1987) Grape losses are pre-sumably dependent on the ratio of auxiliary and pest insectswhich might both be inflated near the edge due to the greaterhabitat complexity thus canceling any positive or negativeeffect on grape losses (Nicholls et al 2001 Williamson andJohnson 2005 Sciarretta and Trematerra 2014 Steel et al2017)

This study constitutes an important first step to evaluatethe potential role of natural biodiversity on grape productiv-ity in Portugal Nevertheless we failed to detect an apprecia-ble effect of biodiversity on grape losses due to insects fungior birds mostly because these losses were already stronglylimited by intensive pesticide spraying which is commonpractice in the region Further studies comparing alternativemanagement actions such as integrated production and bi-ological production strategies will likely shed new light onthe real potential of natural communities to suppress pest out-breaks thus providing a valuable test to the biotic resistancehypothesis

Data availability Raw data are available at httpsfigsharecomsad1245915ab725b15c78

Appendix A

Table A1 Bird species sampled in each month according to theirfunction in the ecosystem

Bird species Function in the ecosystem

JuneJuly August

Motacilla alba Auxiliary AuxiliaryHirundo rustica Auxiliary AuxiliaryDelichon urbicum Auxiliary AuxiliaryApus apus Auxiliary AuxiliaryTroglodytes troglodytes Auxiliary AuxiliarySaxicola rubicola Auxiliary AuxiliaryOenanthe hispanica Auxiliary AuxiliarySerinus serinus Neutral NeutralCyanistes caeruleus Auxiliary AuxiliaryAegithalos caudatus Auxiliary AuxiliaryParus major Auxiliary AuxiliaryLullula arborea Neutral AuxiliarySturnus unicolor Pest PestRegulus ignicapilla Auxiliary PestCorvus corone Auxiliary PestTurdus merula Auxiliary PestSylvia communis Pest PestOriolus oriolus Pest PestFicedula hypoleuca Auxiliary AuxiliaryPasser domesticus Neutral NeutralPica pica Auxiliary AuxiliaryAlectoris rufa Neutral PestLinaria cannabina Neutral NeutralCarduelis carduelis Neutral NeutralErithacus rubecula Auxiliary AuxiliaryColumba livia Neutral NeutralPhoenicurus ochruros Auxiliary PestStreptopelia turtur Neutral NeutralStreptopelia decaocto Neutral NeutralFringilla coelebs Neutral NeutralSylvia atricapilla Auxiliary PestSitta europaea Auxiliary AuxiliaryCerthia brachydactyla Auxiliary AuxiliaryCarduelis chloris Neutral Neutral

Table A2 Insect families sampled according to their function in theecosystem

Insects Function in the ecosystem

Acrididae NeutralAgromyzidae PestAphididae PestApidae NeutralAraneidae AuxiliaryBuprestidae PestChloropidae NeutralChrysomelidae PestChrysopidae AuxiliaryCiccadellidae PestCoccinellidae AuxiliaryCurculionidae PestDermestidae NeutralElateridae PestFormicidae AuxiliaryLygaeidae PestMicrohymenoptera AuxiliaryMiridae AuxiliaryMordellidae NeutralMuscidae NeutralHymenoptera parasitoids AuxiliaryPentatomidae PestPieridae NeutralPsocoptera NeutralSphecidae AuxiliarySyrphidae NeutralTenebrionidae NeutralTephritidae PestThysanoptera PestTipulidae PestTortricidae Pest

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements We are grateful for the kind collaborationof all wine producers who gave us access to their vineyards andshared their valuable insights about wine and vineyards withus We thank Nuno Vilela for triggering this collaboration andestablishing the bridge with land owners and Catherine OrsquoConnorfor commenting on the manuscript Ruben H Heleno was fundedby grant IF004412013 of the Portuguese Foundation for Scienceand Technology (FCT)

Edited by Jutta StadlerReviewed by Ricardo Ceia Tanja Milotic and two anonymousreferees

References

Anderson A Lindell C A Moxcey K M Siemer W FLinz G M Curtis P D Carroll J E Burrows C LBoulanger J R Steensma K M M and Shwiff S ABird damage to select fruit crops The cost of damage andthe benefits of control in five states Crop Prot 52 103ndash109httpsdoiorg101016jcropro201305019 2013

Altieri M A The ecological role of biodiversity inagroecosystems Agr Ecosyst Environ 74 19ndash31httpsdoiorg101016S0167-8809(99)00028-6 1999

Barbaro L and Battisti A Birds as predators of the pine pro-cessionary moth (Lepidoptera Notodontidae) Biol Control56 107ndash114 httpsdoiorg101016jbiocontrol2010100092011

Barbaro L Rusch A Muiruri E W Gravellier B ThieryD and Castagneyrol B Avian pest control in vineyardsis driven by interactions between bird functional diversityand landscape heterogeneity J Appl Ecol 54 500ndash508httpsdoiorg1011111365-266412740 2016

Belliacute N Mariacuten S Coronas I Sanchis V and Ramos A JSkin damage high temperature and relative humidity as detri-mental factors for Aspergillus carbonarius infection and ochra-toxin A production in grapes Food Control 18 1343ndash1349httpsdoiorg101016jfoodcont200602014 2007

Boumlehm J Antunes M T Andrade R Barroso J M CabritaM J Cardoso H Eiras-Dias J E Fernandes L FevereiroP Figueiredo A Gato Oacute Laureano O Loureiro V MartinsA Magalhatildees N Nulasco G Oliveira H Pais M S PeixeA Rego C Sequeira Oacute Sequeira C Silva J R and VelosoM M Portugal Vitiacutecola O Grande Livro das Castas enciclopeacute-dia dos vinhos de Portugal Chaves Ferreira Publicaccedilotildees Lisbon2007

Bostanian N J Vincent C and Isaacs R (Eds) Arthropod Man-agement in Vineyards Springer Netherlands Dordrecht 2015

Bournier A Grape Insects Annu Rev Entomol 22 355ndash761976

Buumlrgi L P Roltsch W J and Mills N J Allee effects and pop-ulation regulation a test for biotic resistance against an inva-sive leafroller by resident parasitoids Popul Ecol 57 215ndash225httpsdoiorg101007s10144-014-0451-4 2015

Canavelli S B Branch L C Cavallero P Gonzaacutelez C andZaccagnini M E Multi-level analysis of bird abundance anddamage to crop fields Agr Ecosyst Environ 197 128ndash136httpsdoiorg101016jagee201407024 2014

Carisse O Bacon R Lasnier J and McFadden-Smith W Ideni-fication Guide to the Major Diseases of Grapes Agriculture andAgri-Food Canada 2006

Ceia R S and Ramos J A Birds as predators ofcork and holm oak pests Agrofor Syst 90 159ndash176httpsdoiorg101007s10457-014-9749-7 2014

Corio-Costet M F Dufour M C Cigna J Abadie P and ChenW J Diversity and fitness of Plasmopara viticola isolates re-sistant to QoI fungicides Eur J Plant Pathol 129 315ndash329httpsdoiorg101007s10658-010-9711-0 2011

Costa J M Ramos J A da Silva L P Timoteo S Arauacutejo PM Felgueiras M S Rosa A Matos C Encarnaccedilatildeo P Ten-reiro P Q and Heleno R H Endozoochory largely outweighsepizoochory in migrating passerines J Avian Biol 45 59ndash64httpsdoiorg101111j1600-048X201300271x 2014

Cramp S and Perrins C M The Birds of the Western PalearcticVolume 7 Oxford Univ Press 1993

Cruz J C Ramos J A da Silva L P Tenreiro P Q and HelenoR H Seed dispersal networks in an urban novel ecosystem EurJ For Res 132 887ndash897 httpsdoiorg101007s10342-013-0722-1 2013

Cunha J Santos M T Carneiro L C Fevereiro Pand Eiras-Dias J E Portuguese traditional grapevine cul-tivars and wild vines (Vitis vinifera L) share morphologi-cal and genetic traits Genet Resour Crop Ev 56 975ndash989httpsdoiorg101007s10722-009-9416-4 2009

Curtis P D Merwin I A Pritts M P and Peterson D V Chem-ical repellents and plastic netting for reducing bird damage tosweet cherries blueberries and grapes HortScience 29 1151ndash1155 1994

da Silva L P Ramos J A Coutinho A P Tenreiro P Q andHeleno R H Flower visitation by European birds offers thefirst evidence of interaction release in continents J Biogeogr44 687ndash695 httpsdoiorg101111jbi12915 2017

De Bach P Biological control of insect pests and weeds Biol Con-trol Insect Pests Weeds 844 pp London Chapman amp Hall Ltd1964

Delaunois B Farace G Jeandet P Cleacutement C Baillieul FDorey S and Cordelier S Elicitors as alternative strategy topesticides in grapevine Current knowledge on their mode of ac-tion from controlled conditions to vineyard Environ Sci PollutR 21 4837ndash4846 httpsdoiorg101007s11356-013-1841-42014

Dignon N Quantifying bird damage to wine Africa aquestionnaire-based approach University of Cape Town 2013

Flower C E Long L C Knight K S Rebbeck J BrownJ S Gonzalez-Meler M A and Whelan C J Native bark-foraging birds preferentially forage in infected ash (Fraxinusspp) and prove effective predators of the invasive emerald ashborer (Agrilus planipennis Fairmaire) Forest Ecol Manag 313300ndash306 httpsdoiorg101016jforeco201311030 2014

Gagetti B L Piratelli A J and Pintildea-Rodrigues F CM Fruit color preference by birds and applications toecological restoration Brazilian J Biol 76 955ndash966httpsdoiorg1015901519-698405115 2016

Galvan T L Koch R L and Hutchison W D Impact of fruitfeeding on overwintering survival of the multicolored Asian ladybeetle and the ability of this insect and paper wasps to in-jure wine grape berries Entomol Exp Appl 128 429ndash436httpsdoiorg101111j1570-7458200800731x 2008

Geiger F Bengtsson J Berendse F Weisser W W EmmersonM Morales M B Ceryngier P Liira J Tscharntke T Win-qvist C Eggers S Bommarco R Paumlrt T Bretagnolle VPlantegenest M Clement L W Dennis C Palmer C OntildeateJ J Guerrero I Hawro V Aavik T Thies C Flohre AHaumlnke S Fischer C Goedhart P W and Inchausti P Persis-tent negative effects of pesticides on biodiversity and biologicalcontrol potential on European farmland Basic Appl Ecol 1197ndash105 httpsdoiorg101016jbaae200912001 2010

Gurr G M Barlow N D Memmott J Wratten S D and Great-head D J A history of methodological theoretical and empiri-cal approaches to biological control in Biological Control Mea-sures of success 3ndash37 Springer Netherlands Dordrecht 2000

Gurr G M Wratten S D Snyder W E and Read D M Y(Eds) Biodiversity and Insect Pests John Wiley amp Sons LtdChichester UK 2012

Hahn J and Wold-burkness S Grape Insect Pestsof the Home Garden Univ Minesota available athttpswwwextensionumnedugardeninsectsfindgrape-insect-pests-of-the-home-gardendocM1236pdf (lastaccess 13 February 2018) 2008

Heide-Jorgensen H S Encyclopedia of Biological Invasions En-cycl Biol Invasions 78 504ndash510 httpsdoiorg1018900012-9658(1997)078[1612PP]20CO2 2011

Herrmann E and Anderson M D Foraging behaviour of damage-causing birds in table grape vineyards in the Orange River ValleySouth Africa South African J Enol Vitic 28 150ndash154 2007

Hocking A D Leong S L Kazi B A Emmett RW and Scott E S Fungi and mycotoxins in vineyardsand grape products Int J Food Microbiol 119 84ndash88httpsdoiorg101016jijfoodmicro200707031 2007

Hoffman C J and Dennehy T J Assessing the Risk of GrapeBerry Moth Attack in New York Vineyards New Yorkrsquos FoodLife Sci Bull Cornell University New York 1987

Hoover E Wold-Burkness S Hilton J Mollov D Burkness EGalvan T Hemstad P and Hutchison W D Grape IPM Guidefor Minnesota Producers University of Minnesota 2011

Isaacs R Schilder A Zabadal T and Weigle T Guidefor Grape IPM Scouting in the North Central available athttpmsueanrmsueduresourcesa_pocket_guide_for_grape_ipm_scouting_of_grapes_in_north_central_and_easter (lastaccess 13 February 2018) 2003

Jedlicka J A Greenberg R and Letourneau DK Avian conservation practices strengthen ecosys-tem services in California vineyards PLoS One 6httpsdoiorg101371journalpone0027347 2011

Johnson M D Kellermann J L and Stercho A M Pest re-duction services by birds in shade and sun coffee in JamaicaAnim Conserv 13 140ndash147 httpsdoiorg101111j1469-1795200900310x 2010

Jonsson M Wratten S D Landis D A and GurrG M Recent advances in conservation biological controlof arthropods by arthropods Biol Control 45 172ndash175httpsdoiorg101016jbiocontrol200801006 2008

Keller M The Science of Grapevines Anatomy and PhysiologySci Grapevines 2010 85ndash105 httpsdoiorg101016B978-0-12-374881-200012-X 2010

Koh L P Birds defend oil palms from herbivorous insects EcolAppl 18 821ndash825 httpsdoiorg10189007-16501 2008

Kross S M Tylianakis J M and Nelson X J Effects ofIntroducing Threatened Falcons into Vineyards on Abundanceof Passeriformes and Bird Damage to Grapes Conserv Biol26 142ndash149 httpsdoiorg101111j1523-1739201101756x2012

Kuznetsova A Brockhoff P B and Christensen R H BlmerTest Package Tests in Linear Mixed Effects Models J StatSoftw 1 httpsdoiorg1018637jssv082i13 2017

Lechenet M Dessaint F Py G Makowski D and Munier-Jolain N Reducing pesticide use while preserving crop pro-ductivity and profitability on arable farms Nat Plants 3 17008httpsdoiorg101038nplants20178 2017

Levine J M Adler P B and Yelenik S G A meta-analysis ofbiotic resistance to exotic plant invasions Ecol Lett 7 975ndash989 httpsdoiorg101111j1461-0248200400657x 2004

Loacutepez-Nuacutentildeez F A Heleno R H Ribeiro S Marchante H andMarchante E Four-trophic level food webs reveal the cascadingimpacts of an invasive plant targeted for biocontrol Ecology 98782ndash793 httpsdoiorg101002ecy1701 2017

Mani M Shivaraju C and Kulkarni N S The Grape Entomol-ogy Springer India 2014

Mccracken D Farmland biodiversity and the Common Agricul-tural Policy (CAP) Policy Brief 4 1ndash8 2011

Michaud J P Sources of mortality in colonies of browncitrus aphid Toxoptera citricida Biocontrol 44 347ndash367httpsdoiorg101023A1009955816396 1999

Moschos T Yield loss quantification and assessment of eco-nomic injury level for the anthophagous generation of theEuropean grapevine moth Lobesia botrana Den et Schiff(Lepidoptera Tortricidae) Int J Pest Manag 51 81ndash89httpsdoiorg10108009670870500107566 2005

Niccolucci V Galli A Kitzes J Pulselli R M Borsa S andMarchettini N Ecological Footprint analysis applied to the pro-duction of two Italian wines Agr Ecosyst Environ 128 162ndash166 httpsdoiorg101016jagee200805015 2008

Nicholls C I Parrella M and Altieri M The effects of a vegeta-tional corridor on the abundance and dispersal of insect biodiver-sity within a northern California organic vineyard Landsc EcolEng 16 41ndash42 httpsdoiorg101023A10111282228672001

Oliveira C M Auad A M Mendes S M and Friz-zas M R Crop losses and the economic impact of in-sect pests on Brazilian agriculture Crop Prot 56 50ndash54httpsdoiorg101016jcropro201310022 2014

R Core Team R A language and environment for statistical com-puting R Foundation for Statistical Computing Vienna Austria2016

Perini A and Susi A Developing a decision support system forintegrated production in agriculture Environ Modell Softw 19821ndash829 httpsdoiorg101016jenvsoft200303001 2004

Pithon J A Beaujouan V Daniel H Pain G and Val-let J Are vineyards important habitats for birds at lo-cal or landscape scales Basic Appl Ecol 17 240ndash251httpsdoiorg101016jbaae201512004 2016

Saxton V P Creasy G L Paterson A M and Trought M C TExperimental method to investigate and monitor bird behaviorand damage in vineyards Am J Enol Viticult 55 288ndash2912004

Saxton V P Creasy G L Paterson A M and Trought M CT Behavioral responses of european blackbirds and australasiansilvereyes to varying acid and sugar levels in artificial grapesAm J Enol Viticult 60 82ndash86 2009

Sciarretta A and Trematerra P Geostatistical tools for the study ofinsect spatial distribution Practical implications in the integratedmanagement of orchard and vineyard pests Plant Prot Sci 5097ndash110 2014

Somers C M and Morris R D Birds and wine grapes Foragingactivity causes small-scale damage patterns in single vineyardsJ Appl Ecol 39 511ndash523 httpsdoiorg101046j1365-2664200200725x 2002

Steel Z L Steel A E Williams J N Viers J H Marquet PA and Barbosa O Patterns of bird diversity and habitat use inmixed vineyard-matorral landscapes of Central Chile Ecol In-dic 73 345ndash357 httpsdoiorg101016jecolind2016090392017

Stevenson A B and Virgo B B Damage By Robins and Star-lings To Grapes in Ontario Can J Plant Sci 51 201ndash210httpsdoiorg104141cjps71-041 1971

Takahara B and Takahashi K H Associative learn-ing of color and firmness of oviposition substrates inDrosophila suzukii Entomol Exp Appl 162 13ndash18httpsdoiorg101111eea12521 2016

Thomson L J and Hoffmann A A Field validation oflaboratory-derived IOBC toxicity ratings for natural ene-mies in commercial vineyards Biol Control 39 507ndash515httpsdoiorg101016jbiocontrol200606009 2006

Thomson L J Neville P J and Hoffmann A A Effectivetrapping methods for assessing invertebrates in vineyards AustJ Exp Agric 44 947ndash953 httpsdoiorg101071EA032192004

Tracey J Bomford M Hart Q Saunders G and Sinclair RManaging Bird Damage to Fruit and Other Horticultural CropsBureau of Rural Sciences Canberra 2007

Varandas S Teixeira M J Marques J C Aguiar A Alves Aand Bastos M M S M Glucose and fructose levels on grapeskin Interference in Lobesia botrana behaviour Anal Chim Ac513 351ndash355 2004

Viel J-F Challier B Pitard A and Pobel D Brain Can-cer Mortality among French Farmers The Vineyard Pesti-cide Hypothesis Arch Environ Heal An Int J 53 65ndash70httpsdoiorg10108000039899809605690 1998

Wenny D G DeVault T L Johnson M D Kelly D Seker-cioglu C H Tomback D F and Whelan C J The Need toQuantify Ecosystem Services Provided by Birds Auk 128 1ndash14 httpsdoiorg101525auk201110248 2011

Whitney K D Linking frugivores to the dynamics ofa fruit color polymorphism Am J Bot 92 859ndash867httpsdoiorg103732ajb925859 2005

Williams L and Martinson T E Colonization of NewYork vineyards by Anagrus spp (Hymenoptera Mymaridae)Overwintering biology within-vineyard distribution of waspsand parasitism of grape leafhopper Erythroneura spp (Ho-moptera Cicadellidae) eggs Biol Control 18 136ndash146httpsdoiorg101006bcon20000817 2000

Williamson J R and Johnson D T Effects of grape berry mothmanagement practices and landscape on arthropod diversity ingrape vineyards in the southern United States Horttechnology15 232ndash238 2005

Abstract
Introduction
Methods
- Exclusion experiment
- Bird census
- Insect sampling
- Statistical analysis
- - Results
  - - Vineyard biodiversity
    - Grape losses
    - Biodiversity and grape losses
    - - Discussion
      - Grape losses
        
        Relation between losses and natural biodiversity
        
        Edge effect
        
        Data availability
        
        Appendix A
        
        Competing interests
        
        Acknowledgements
        
        References

Page 2: Effects of native biodiversity on grape loss of four …...documented to control arthropod pests in agricultural con-texts (Wenny et al., 2011; Ceia and Ramos, 2014). Similarly, while