Spatial and temporal spread of maize stem borer Busseola ... · fusca (Fuller) is the most damaging in the high potential yield areas, which include the highland tropics and moist

Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/309319204

SpatialandtemporalspreadofmaizestemborerBusseolafusca(Fuller)(Lepidoptera:Noctuidae)damageinsmallholder...

ArticleinAgricultureEcosystems&EnvironmentNovember2016

DOI:10.1016/j.agee.2016.10.013

CITATIONS

0

READS

55

6authors,including:

Someoftheauthorsofthispublicationarealsoworkingontheserelatedprojects:

INRASPE-IMPACT_PHYTOViewproject

Evolution,systematicsanddiversificationdynamicsofseed-beetlesViewproject

NdjomatchouaThomasFrank

UniversityofYaoundeI

5PUBLICATIONS3CITATIONS

SEEPROFILE

HenriE.Z.Tonnang


SEEPROFILE

ChristophePlantamp

AgenceNationaledeScuritSanitairedel'A


SEEPROFILE

BrunoPierreLeRu

InstituteofResearchforDevelopment

307PUBLICATIONS1,428CITATIONS

SEEPROFILE

AllcontentfollowingthispagewasuploadedbyChristophePlantampon11January2017.

Theuserhasrequestedenhancementofthedownloadedfile.Allin-textreferencesunderlinedinblueareaddedtotheoriginaldocumentandarelinkedtopublicationsonResearchGate,lettingyouaccessandreadthemimmediately.

https://www.researchgate.net/publication/309319204_Spatial_and_temporal_spread_of_maize_stem_borer_Busseola_fusca_Fuller_Lepidoptera_Noctuidae_damage_in_smallholder_farms?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_2&_esc=publicationCoverPdfhttps://www.researchgate.net/project/INRA-SPE-IMPACT-PHYTO?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_9&_esc=publicationCoverPdfhttps://www.researchgate.net/project/Evolution-systematics-and-diversification-dynamics-of-seed-beetles?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_9&_esc=publicationCoverPdfhttps://www.researchgate.net/?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_1&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Ndjomatchoua_Thomas_Frank?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_4&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Ndjomatchoua_Thomas_Frank?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_5&_esc=publicationCoverPdfhttps://www.researchgate.net/institution/University_of_Yaounde_I?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_6&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Ndjomatchoua_Thomas_Frank?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_7&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Henri_Tonnang2?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_4&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Henri_Tonnang2?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_5&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Henri_Tonnang2?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_7&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Christophe_Plantamp?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_4&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Christophe_Plantamp?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_5&_esc=publicationCoverPdfhttps://www.researchgate.net/institution/Agence_Nationale_de_Securite_Sanitaire_de_lAlimentation_de_lEnvironnement_et_du_Travail?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_6&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Christophe_Plantamp?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_7&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Bruno_Le_Ru?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_4&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Bruno_Le_Ru?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_5&_esc=publicationCoverPdfhttps://www.researchgate.net/institution/Institute_of_Research_for_Development?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_6&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Bruno_Le_Ru?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_7&_esc=publicationCoverPdfhttps://www.researchgate.net/profile/Christophe_Plantamp?enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg%3D%3D&el=1_x_10&_esc=publicationCoverPdf

Agriculture, Ecosystems and Environment 235 (2016) 105118

Spatial and temporal spread of maize stem borer Busseola fusca (Fuller)(Lepidoptera: Noctuidae) damage in smallholder farms

Frank T. Ndjomatchouaa,b,*, Henri E.Z. Tonnanga,c, Christophe Plantampd,Pascal Campagnee, Clment Tchawouab, Bruno P. Le Rua,f,g

a icipe African Insect Science for Food and Health, P. O. Box 30772-00100, Nairobi, Kenyab Laboratory of Mechanics, Department of Physics, Faculty of Sciences, University of Yaound 1, P. O. Box 812 Yaound, Cameroonc International Maize and Wheat Improvement Center (CIMMYT) ICRAF House, Off United Nation, Avenue, Gigiri, P. O. Box 1041, Village Market, 00621,Nairobi, KenyadUniversit de Lyon, 69000 Lyon, Universit Lyon 1, Laboratoire Biomtrie et Biologie Evolutive. Universit Claude Bernard Lyon 1 Btiment Gregor Mendel,43 bd du 11 novembre 1918, 6922 Villeurbanne Cedex, FranceeDepartment of Evolution, Ecology and Behavior, Institute of Integrative Biology, Biosciences Building, University of Liverpool, Crown Street, Liverpool L697ZB, United Kingdomf IRD/CNRS UMR IRD 247 EGCE, Laboratoire Evolution Gnomes Comportement et Ecologie, CNRS, Bat. 13, 1 Avenue de la Terrasse, 91198 Gif sur Yvette Cedex,FrancegUniversit Paris-Sud 11, 91405 Orsay Cedex, France

A R T I C L E I N F O

Article history:Received 30 May 2016Received in revised form 10 October 2016Accepted 12 October 2016Available online 20 October 2016

Keywords:Busseola fuscaSmallholder maize farmsDamage spreadTemporal patternSpatial pattern

A B S T R A C T

The main purpose of this study was to understand the spatio-temporal spread of the maize stem borerBusseola fusca (Fuller) (Lepidoptera: Noctuidae) in smallholder maize farms. The analysis carried outallowed the establishment of complementary sampling scheme and analysis that can be applied toinvestigate the propagation of stem borer damages and extended to other insect pests. This approachrequires consideration of all plants point locations, the knowledge on the level of damage and itscharacterization. Results showed that there was a two-week interval between occurrence of the peaks ofleaf damage and male adult moth abundance. The prior role of leaf damages in the farm infestation by B.fusca is revealed, and an estimate of the mean transition time between different damage types isprovided. Furthermore, damaged plants exhibited a local spatial autocorrelation within a range ofdependence of 0-10 meters; and the spatio-temporal pattern of B. fusca damage spread evolves as a spiralaround an initial patch of damaged plants. By assuming a neighbor configuration of distribution ofdamaged plants nearby non-damaged, we showed that the inner plants are likely to become damagedwithin a time period of a week; thus, B. fusca infests farms in a systematic fashion. Overall, these resultshave useful implications for improving and optimizing existing field sampling methods for insect pestdamages. The approaches used in carrying out the analysis further provided a deep understandinghelpful to improve integrated pest management (IPM) strategies against stem borers, and offer IPMpractitioners the opportunity to design, develop, and implement optimum control methods against B.fusca, an important pest of maize in Africa.

2016 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment

journal homepage: www.elsev ier .com/locate /agee

1. Introduction

The relation between ecological processes (such as individualdispersal, habitat selection, and spatial damage distribution

* Corresponding author at: icipe African Insect Science for Food and Health, P. O.Box 30772-00100, Nairobi, Kenya.

E-mail addresses: [email protected], [email protected](F.T. Ndjomatchoua).

http://dx.doi.org/10.1016/j.agee.2016.10.0130167-8809/ 2016 Elsevier B.V. All rights reserved.

pattern) is of primary importance in ecology and in agro-ecologicalsystems, to allow efficient control measures against insect pests(Mazzi and Dorn, 2012; Vinatier et al., 2011). Yield losses in cropsare a consequence of the spatial and temporal dispersal of theinsect pests (Caldiz et al., 2002; Ferguson et al., 2003; Hughes,1996; van Leeuwen et al., 2000; Winder et al., 2013). Therefore, fordecades, spatial and temporal dispersion information about theseharmful organisms has gained great relevance for plant protectionspecialists and agricultural entomologists (Aukema et al., 2006;Cocu et al., 2005; Diaz et al., 2012; Emmen et al., 2004;

http://crossmark.crossref.org/dialog/?doi=10.1016/j.agee.2016.10.013&domain=pdfmailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.agee.2016.10.013http://dx.doi.org/10.1016/j.agee.2016.10.013http://www.sciencedirect.com/science/journal/01678809www.elsevier.com/locate/ageehttps://www.researchgate.net/publication/223684562_Agro-ecological_zoning_at_the_regional_level_spatio-temporal_variation_in_potential_yield_of_the_potato_crop_in_the_Argentinian_patagonia?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/221790632_Interplant_movement_and_spatial_distribution_of_alate_and_apterous_morphs_of_Nasonovia_ribisnigri_Homoptera_Aphididae_on_lettuce?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/223221039_Spatial_distribution_of_pest_insects_in_oilseed_rape_Implications_for_integrated_pest_management?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/222356419_Incorporating_spatial_pattern_of_harmful_organisms_into_crop_loss_models?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/222356419_Incorporating_spatial_pattern_of_harmful_organisms_into_crop_loss_models?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/227636721_Landscape_level_analysis_of_mountain_pine_beetle_in_British_Columbia_Canada_Spatiotemporal_development_and_spatial_synchrony_within_the_present_outbreak?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/262910296_Movement_of_insect_pests_in_agricultural_landscapes?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/236017515_The_spatial_distribution_of_canopy-resident_and_ground-resident_cereal_aphids_Sitobion_avenae_and_Metopolophium_dirhodum_in_winter_wheat?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/229874627_Spatial_autocorrelation_for_identifying_the_geographical_patterns_of_aphid_annual_abundance?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/227663817_Factors_and_mechanisms_explaining_spatial_heterogeneity_A_review_of_methods_for_insect_populations?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/40792120_Yield_loss_in_apple_caused_by_Monilinia_fructigena_Aderh_Rhul_Honey_and_spatio-temporal_dynamics_of_disease_development?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/289409404_Temporal_and_spatial_dynamics_of_Empoasca_fabae_Harris_Homoptera_Cicadellidae_in_alfalfa?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==

106 F.T. Ndjomatchoua et al. / Agriculture, Ecosystems and Environment 235 (2016) 105118

Holland et al., 2005; Ishaaya and Horowitz, 2004; Kautz et al., 2011;Keasar et al., 2005; Kim et al., 2007; Lausch et al., 2013; MoralGarca, 2006; Perry, 1994; Reay-Jones, 2010; Reay-Jones et al.,2007; Smith et al., 2004; Thomas et al., 2001; Wang et al., 2009;Wright et al., 2002; Zimmerman et al., 2004).

Several methodologies are employed to describe the spatial andtemporal distribution of insect pest damage. The most commonlyused approaches are: tracking the evolution of the centroid ofinfestation distribution (Diaz et al., 2012; Lausch et al., 2013; Reay-Jones et al., 2010), analyzing the spatial autocorrelation ofinfestation (Blackshaw and Hicks, 2013; Bone et al., 2013; Cocuet al., 2005; Smith et al., 2004), carrying out spatial evaluation bythe estimate of distance indices (Cocco et al., 2015; Dder et al.,2015; Holland et al., 2005; Kim et al., 2007; Li et al., 2012; Reinekeet al., 2011; Schumann et al., 2014; Thomas et al., 2001), conductingMarkovian chain analysis (Zimmerman et al., 2004), deployinggeostatistical tools (Keasar et al., 2005; Moral Garca, 2006; Reay-Jones et al., 2010; Rogers et al., 2015; Wright et al., 2002), usingspatial mapping (Emmen et al., 2004; Fernandes et al., 2015; Kautzet al., 2011; Massoud et al., 2012; Zhang et al., 2016) and usingcluster analysis (Aukema et al., 2006). However, results obtainedfrom these approaches only specified the type of spatial andtemporal pattern used by the pests during the damage spread. Ithas been mostly revealed that such pattern is random, with eitheraggregated or regular structures (Begon et al., 1996; Vinatier et al.,2011). Although some of the above-mentioned methodologies forspatial analysis have been applied to various types of insect pestsand crops (Aukema et al., 2006; Blackshaw and Hicks, 2013; Boneet al., 2013; Cocco et al., 2013; Cocu et al., 2005; Dder et al., 2015;Diaz et al., 2012; Emmen et al., 2004; Fernandes et al., 2015;Holland et al., 2005; Ishaaya and Horowitz, 2004; Kautz et al., 2011;Keasar et al., 2005; Kim et al., 2007; Lausch et al., 2013; MoralGarca, 2006; Perry, 1994; Reay-Jones, 2010; Reay-Jones et al.,2007; Reineke et al., 2011; Rogers et al., 2015; Schumann et al.,2014; Smith et al., 2004; Thomas et al., 2001; Wang et al., 2009;Wright et al., 2002; Zhang et al., 2016; Zimmerman et al., 2004), nostudies have investigated the spatio-temporal damage patterns oflepidopteran stem borers using experimental data from small-holder maize farms.

Maize (Zea mays L.) is the most important staple food in sub-Saharan Africa, particularly in East Africa (De Groote et al., 2005;Kipkoech et al., 2006). However, biotic (stem borers, gray leaf spot,maize streak virus) and abiotic (drought, low soil fertility) factorsconstrain maize production (Stevens, 2008). Lepidopteran stemborers are considered to be the most damaging insect pests ofmaize in Africa (Overholt et al., 2001). In East Africa, the noctuidBusseola fusca (Fuller) is the most damaging in the high potentialyield areas, which include the highland tropics and moisttransitional zones (De Groote, 2002; Ongamo et al., 2006). FemaleB. fusca moths lay several hundred eggs (in batches of 30100inserted between the sheath and the stem) (Harris and Nwanze,1992). Larvae hatch after one week and disperse over neighboringplants using silk strands (Harris and Nwanze, 1992). After passingthrough five larval instars in 3045 days, all the while eating theleaves and stems, they pupate in tunnels inside the plant, oftenafter excavating emergence windows to facilitate the exit of adultmoths (Overholt et al., 2001). Adults emerge 1020 days afterpupation. The life cycle is completed in 78 weeks. In its earlylarval stage, this insect causes foliar damage during the plant whorlstage which gives an array of small holes when unfolded; thisdamage is called leaf damage (LD) (Harris and Nwanze, 1992).Destruction of the meristematic tissues causes dead heart (DH)damage (Bosque-Prez and Mareck, 1998). Furthermore, stemborers may cause exit holes (EH) damage at the periphery of thestem, to facilitate the exits of adult moths (Harris and Nwanze,1992). Damages caused by stem borers at larval stage are

ultimately stunting plant growth and plant death (D) (Brenire,1971). In high potential areas of Kenya, yield losses of maize due tostem borer infestations are estimated at between 12%50% of thetotal production, as a result of leaf feeding, dead heart, stemtunneling, direct damage to grain, and secondary infection by stalkrots and lodging (De Groote et al., 2003; Kfir et al., 2002; Polaszek,1998).

Research on lepidopteran stem borer pests has been carried outin sub-Saharan Africa for decades (Calatayud et al., 2014; Harrisand Nwanze, 1992; Kfir et al., 2002). Although the biology andecology of lepidopteran stem borers have been extensively studiedduring a long period of time (Calatayud et al., 2014; Harris andNwanze, 1992; Kfir et al., 2002; Polaszek, 1998), to our knowledge,there are no downscaled studies from smallholder maize farms toassess spatio-temporal infestation dynamics of B. fusca. Theavailable information is related only to regional distribution,agro-climatic preferences and phylogeography over wide scales(Dupas et al., 2014; Guofa et al., 2001; Harris and Nwanze, 1992;Hauptfleisch et al., 2014; Le Ru et al., 2006; Mwalusepo et al., 2015;Ongamo et al., 2006; Overholt et al., 2001; Sezonlin et al., 2006).Understanding the spatial and temporal dynamics of damagespread constitutes the basic information for future development ofappropriate pest management strategies. Most studies concernedwith smallholder farming have focused on the density and rate ofinfestation at field level, by selecting randomly a few plants perfield and looking at the within-plant insect distribution, withoutmaking emphasis on the level of damage, its characterization andthe insect pest distribution in the field (Amoako-Atta et al., 1983;Ndemah et al., 2001; Overholt et al., 1994; Van Rensburg andPringle, 1989). Such sampling approach presents a considerablebias, because of the probable failure in capturing the infestationspread in the whole farm. Selecting only a small number of plantsfailed to reveal the spatial and temporal dynamic of infestation ofthe pest.

The main purpose of this study is to understand the spatio-temporal spread of B. fusca damage at smallholder maize farmlevel. Specific objectives are to: find a relationship between thetemporal evolution of infestations rate and the flight dynamicpattern; estimate the probability of infestations and understandthe transition time between damage types; analyze the spatialdistribution of the infestation; and identify the spread pattern ofthe damage in the farm.

2. Materials and methods

2.1. Study site and sampling procedure

The farms selected for the study are located in Naivasha, in theRift Valley region, Northwest of Nairobi. The geographicalcoordinates are: latitude 04300000 S, longitude 362600900 E,2086 m a.s.l. Busseola fusca is the predominant lepidopteran maizestem borer at this elevation (Ongamo et al., 2006). The study wascarried out on private land, after the owners gave permission toconduct the study on these sites. Six maize plots (Table 1) withmonoculture farming were selected. The maize plants wereplanted at regular row/within row space intervals, to facilitatecounting. The planting date and crop management practices wereidentical in all selected plots. Data collection consisted of visualchecking of all plants damaged by B. fusca, and was conductedweekly during 13 weeks, from the 23 November, 2010 to the 10February, 2011. Coordinates of all sampled maize plants with/without damages were recorded. Four plant damage types wereconsidered: LD, DH, EH and D. Furthermore, we placed twopheromone insect traps around each field to follow the flightdynamic of B. fusca males. For each plant, we assigned an ordinalnumber ranging from 1 to the total number of plants within the

https://www.researchgate.net/publication/231760233_Influence_of_maize_cowpea_and_sorghum_intercropping_systems_on_stem-pod-borer_infestations?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/230819342_Distribution_pest_status_and_agro-dimatic_preferences_of_lepidopteran_stem_borers_of_maize_in_Kenya?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/230819342_Distribution_pest_status_and_agro-dimatic_preferences_of_lepidopteran_stem_borers_of_maize_in_Kenya?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/261107535_Phyloeography_in_continuous_space_Coupling_species_distribution_models_and_circuit_theory_to_assess_the_effect_of_contiguous_migration_at_different_climatic_periods_on_genetic_differentiation_in_Busse?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/232006474_Changes_in_the_distribution_of_Lepidopteran_maize_stemborers_in_Kenya_from_the_1950s_to_1990s?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/230819333_Geographic_distribution_and_host_plant_ranges_of_East_African_noctuid_stem_borers?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/274509631_Ecology_of_the_African_Maize_Stalk_Borer_Busseola_fusca_Lepidoptera_Noctuidae_with_Special_Reference_to_Insect-Plant_Interactions?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/274509631_Ecology_of_the_African_Maize_Stalk_Borer_Busseola_fusca_Lepidoptera_Noctuidae_with_Special_Reference_to_Insect-Plant_Interactions?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/11627000_Biology_and_Management_of_Economically_Important_Lepidopteran_Cereal_Stem_Borers_in_Africa?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/11627000_Biology_and_Management_of_Economically_Important_Lepidopteran_Cereal_Stem_Borers_in_Africa?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/11627000_Biology_and_Management_of_Economically_Important_Lepidopteran_Cereal_Stem_Borers_in_Africa?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/278677039_Predicting_the_Impact_of_Temperature_Change_on_the_Future_Distribution_of_Maize_Stem_Borers_and_Their_Natural_Enemies_along_East_African_Mountain_Gradients_Using_Phenology_Models?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/7326391_Phylogeography_and_population_genetics_of_the_maize_stalk_borer_Busseola_fusca_Lepidoptera_Noctuidae_in_sub-Saharan_Africa?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/32997959_Les_problemes_de_lepidopteres_foreurs_des_graminees_en_Afrique_de_l'Ouest?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/32997959_Les_problemes_de_lepidopteres_foreurs_des_graminees_en_Afrique_de_l'Ouest?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==

Table 1Description of plots used for the experiment. Distances are given in meters. The dimensions are the maximal repartition range of plants inside the plot. The six plots covered azone of 7 kilometer square.

Plot Row length Row spacing Dimensions Total number of maize plants GPS coordinateslatitude, longitude

1 55.00 0.81 51.03 53.93 6547 0.7701, 36.49592 31.54 1.86 150.66 31.57 3914 0.7747, 36.47863 53.10 1.56 123.24 52.14 8146 0.7819, 36.47114 30.50 1.49 120.69 30.50 4974 0.7713, 36.47665 46.80 1.40 74.20 46.80 4884 0.7826, 36.47046 27.82 1.05 153.30 27.82 7314 0.7765, 36.4791

F.T. Ndjomatchoua et al. / Agriculture, Ecosystems and Environment 235 (2016) 105118 107

field. In addition, for each of the 4 damage types, we assigned foreach plant an integer value indicating the week of the observeddamage. If there was no B. fusca damage observed during thecollection period, the value given was 0.

2.2. Observation of infestation rate variation and insect trap catches

To get the weekly variation of LD rate, the differences betweennumbers of damaged plants during two consecutive weeks wereevaluated in each plot. Results were normalized so that theassociated values ranged from 0 to 1. As well, the weekly meanvalues of adult B. fusca males caught in insect traps were computedand then, the temporal dynamics of LD rate and the abundance ofadult males were compared.

2.3. Conditional probabilities for plant infestation type

Because maize was planted at regular row/within row spaceand the planting date and crop management were identical as wellas the date of data collection, we decided to compute theprobability that a randomly selected plant is found dead (D) oncedamage i has occurred, where i = LD, EH or DH. Next, theconditional probability that D occurs once i has occurred wascomputed.

2.4. Meantime transition between infestation types

The estimation of the average time span between damage typeswas done. First, for plant P we recorded the damage type i (i = LD,EH, DH or D), and the corresponding week of observation. Second,we checked and recorded the week of occurrence for otherdifferent damage from i, which we called j. Third, we computed thedifference between the two weeks to get the time duration for thetransition Pi ! Pj. Fourth, we repeated the same process for othermaize plants to obtain a mean time transition from a damage typeto another.

2.5. Spatial distribution

2.5.1. Spatial autocorrelationObservations of damage in plants with different geographical

coordinates may not be uncorrelated. Spatial autocorrelation maybe positive or negative symbolizing how similar or dissimilardamage occurs close by. The farm in this context had a latticestructure made of discontinuous spatial repartition of maizeplants. The first step in this analysis was to define the Euclideandistance matrix from x- and y-coordinates of two individual plantsi and j (dij). Secondly, the spatial relationship between damagedplants was quantified using the spatial weight matrix W in whichelements represent the strength of the spatial structure betweenunits (Anselin et al., 2004; Cliff and Ord, 1973). This matrix wasused to evaluate the level of spatial autocorrelation. There are

various ways to define W and the choice of a particular method toother is subjective. The easiest option is to construct a binarycontiguity matrix (made of 0 and 1) by specifying the units that areadjacent (1) and those that are not (0) (Cliff and Ord, 1981).Therefore, the spatial weight matrix wrij is equal to 1 for dij less thana certain critical distance r and, it is equal to 0 otherwise. Thevariable r is the radius of proximity. The Morans I coefficient Ir isused to quantify the degree of spatial correlation betweenneighboring infested plants. The formula used to calculate MoransI can be found in the literature (Cocu et al., 2005; Jumars et al.,1977; Moran, 1950; Zuur et al., 2007). In this analysis, the valueassigned to each plant is a binary index of LD, EH, DH or D (0 fornon-infested and 1 for infested plants). The interpretation ofMorans I is similar to the correlation coefficient (Zuur et al., 2007).If the Morans I is null, then the spatial link between infested plantsat distant locations is null, and there is no spatial autocorrelation(SA). If the Morans I is positive, then the contagiousness of infestedplants at distinct locations is considerable, and the SA is positive. Ifthe Morans I is negative, then the majority of the infested plantsare not next to each other, and the SA is negative. The graphicalrepresentation of the Morans I function of r is called variogram (orcorrelogram). For the Morans I computation, data were reorgan-ized as follow: for each maize plant we recorded the spatialcoordinates. Values of LD, EH, DH and D, were 0 or 1 (1 is forinfested and 0 for non-infested). The process was repeatedcumulatively for subsequent weeks to obtain a temporal vario-gram.

2.5.2. Tracking the center of plant infestationObserving directly the spatial and temporal evolution of plant

infestation is difficult in a maize farm having a considerablenumber of stems; thus we opted to compute the iso-barycentre (IB)coordinates of each occurring damage type. First, we selected thecoordinates of plants with leaf damage, and then we calculated thecoordinates of the weekly IB without taking into account resultsobtained from the previous weeks computations. We obtained theIB coordinates by averaging x- and y-coordinates corresponding tothe set of newly infested plants observed at each week, whichallowed us to track the position of the center of patches formed byinfested plants with time. The IB coordinates at a week t xt ; yt wascomputed as follows:

xt ; yt Xnti1

xtint;Xnti1

ytint

!; 1

where xti ; yti are the spatial coordinates of a damaged plant i atweek t, and nt is the total number of damaged plants at week t.

2.5.3. Model-based cluster analysis: spatial clusteringTo follow the evolution of the initial shape, density and number

of clusters of infested plants formed with time the model-basedcluster analysis was used. For a spatial classification of clusters, the

https://www.researchgate.net/publication/227260508_Detecting_two-dimensional_spatial_structure_in_biological_data?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/227260508_Detecting_two-dimensional_spatial_structure_in_biological_data?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/229874627_Spatial_autocorrelation_for_identifying_the_geographical_patterns_of_aphid_annual_abundance?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/8295498_Notes_on_Continuous_Stochastic_Phenomena?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/40792120_Yield_loss_in_apple_caused_by_Monilinia_fructigena_Aderh_Rhul_Honey_and_spatio-temporal_dynamics_of_disease_development?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/40792120_Yield_loss_in_apple_caused_by_Monilinia_fructigena_Aderh_Rhul_Honey_and_spatio-temporal_dynamics_of_disease_development?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==

Fig. 1. Neighborhood plant configurations (black color) around the central plant(white color). (a) Representation of the extended Moore neighborhood types thecentral plant possessing the label number 1 the rest are labeled randomly from 2 to25. (b) Representation of the Moore neighborhood type. We consider the centralplant possessing the label number 1, the rest are labeled randomly from 2 to 9.


following procedure was adopted: (i) positions of damaged plantswere assumed implicitly generated by a mixture of probabilitydistribution function in which different components representgroups or clusters; (ii) based on the framework developed in(Banfield and Raftery, 1993), we used Gaussian and non-Gaussianprobability density (PDF) functions for the clustering. In addition,orientation, volume, and shape of the cluster were determinedusing the models summarized in Table 2 (Fraley et al., 2007). Thesethree features are characteristics of the bottom of the two-dimensional PDF. The likelihood of each model is estimated via theexpectation-maximization algorithm (Dempster et al., 1977),which allows us to assign a Bayesian information criterion (BIC)(Akaike, 1974; Schawrz, 1978) for selecting the best model. Thissection of the analysis was carried out using the statistical packagemclust of the software R (R Core Team, 2013).

2.5.4. Spatial and temporal contagion patterns: identification ofcellular automata rules for the propagation of infestations

Cellular automata (CA) are spatially and temporally discretesystem characterized by local interaction and synchronousdynamical evolution (Von Neumann, 1957). Usually, CA is usedas a modeling approach to mimic the behavior and pattern of thespread of an infestation. However, our objective was to find thelikely geometrical configuration (rule) of the spread of infestationto uninfested maize plants by considering each maize plant as anelement, which is attributed a state (infested/non-infested), whichwas assumed to change depending on the plants state and thestates of the plants in its vicinity. Hence, CA is useful forinvestigating the spatial pattern of the nearest neighboring plantsaround a safe central plant that is most likely the source of thecontagion. The minimal neighborhood for the CA is estimated viathe application of the algorithm described in (Sun et al., 2011). Themethodology was structured as follows: (i) the state of each plantand its neighbors are collected in a maximal neighborhood radius(initially fixed), then the primary neighborhood of a plant isformed by 24 maize plants (Fig.1); (ii) the initial neighborhood wasreduced by selecting the configuration that minimized thevariance between the CA results and the data; (iii) in the obtainedneighborhood, the plants without any effect on the state of thecentral plant are removed; (iv) the BIC is used to determine theneighboring configuration that has the most significant impact onthe change of state of the central maize plant. This approach wasimplemented in Matrix Laboratory (MATLAB, 2010).

3. Results

Due to climatic conditions in Naivasha, B. fusca was theexclusive lepidopteran maize stem borer found during the surveyperiods. Subsequently the results presented here only focus on thisspecie.

Table 2Model identifiers use three letters encoding the geometric characteristics: Volume-Shape-Orientation. E means equal, V means varying across clusters and I refers toidentical orientation (Fraley et al., 2007).

Model identifier distribution

EII SphericalVII SphericalEEI DiagonalVEI DiagonalEVI DiagonalVVI DiagonalEEE EllipsoidalEEV EllipsoidalVEV EllipsoidalVVV Ellipsoidal

3.1. Dynamics of infestation

Variation of the number of damaged plants and damage typesper week for each plot are shown in Fig. 2. It is observed that thephenomenon is not linear. A detailed observation of LD in plots 3, 5and 6, displayed two notable phases of the occurrence of damages,especially around weeks 45 and weeks 78. During weeks 45 asudden increase of damaged plants in plots 3, 5 and 6 is detectedalthough no damage was previously noticed. At weeks 7 and 8,another sudden peak of damage is noted before starting togradually decrease, as the maturity of the plants approached. InFig. 3, we showed the variation of the average number of insectscaught and the number of LD in the six plots. Two peaks of maleflying activities can be clearly seen during weeks 2 and 6.Subsequent appearances of LD peaks during weeks 4 and 8 can alsobe noticed. Linking Figs. 2 and 3, we estimated that the time lagbetween the peak of B. fusca male captured and the observed peakof LD is approximately 2 weeks.

3.2. Probability and transition time between damages

The conditional probability for a randomly selected plant to diefollowing different situations is given in Table 3. In plots 1 and 2,the probability for a plant to die given that it has LD is the highest.For plots 36 those probabilities are null. Table 4 shows the meantransition time between different types of damage. The time spansare similar for plots 1 and 2 and, longer for plot 5. In plot 4, thetransitions are faster compared to the other plots. Plots 3, 4 and 6have a shorter transition time between damages. The transitionLD ! DH seems to be the fastest. Overall, if all the plots areconsidered as a unique field, the results would demonstrate thatthe transition time between different types of damage is notuniform, which means the phenomenon is stochastic.

3.3. Spatial scattering of leaf damage with time

The weekly evolution of spatial and temporal autocorrelation(Morans I) for the LD is depicted in Fig. 4. For plots 2, 3, 5, and 6 theinfestation is strongly spatially correlated for a very low radius ofproximity between infested plants, but over time, the spatial linkbetween damaged plants became more and more considerable.The spatial correlation distance for damaged plants inside thesefour plots did not exceed 10 m. A considerable spatial autocorrela-tion was observed between damaged plants within a radius ofproximity exceeding 10 m in plot 1 and 4. These results help us to

https://www.researchgate.net/publication/265505693_Model-Based_Gaussian_and_Non-Gaussian_Clustering?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/5142972_Model-Based_Methods_of_Classification_Using_the_Mclust_Software_in_Chemometrics?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/5142972_Model-Based_Methods_of_Classification_Using_the_Mclust_Software_in_Chemometrics?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/49661344_Fast_Rule_Identification_and_Neighborhood_Selection_for_Cellular_Automata?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/221995817_Maximum_Likelihood_from_Incomplete_Data_Via_EM_Algorithm?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/271196714_New_Look_at_Statistical-Model_Identification?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==

Fig. 2. Differences in infested plants densities between two consecutive weeks. The computation has been done for the four infestation types, leaf damages (LD), death hearth(DH), exit hole (EH) and death (D). For plot 4 see the annex.


understand the level of similarity and dependence between thepoint locations of damaged plants. Fig. 5 shows the weekly trackingof spatial iso-barycentre (IB) for LD plants. The initial IB position israndom; the others are located around the former IB. After the firstcycle of infestation, we observed a general trend towards an IBfurther from the previous recorded during the first cycle. Indeed, a

spatio-temporal pattern of infestation is clearly visible: a spiral-like pattern of LD at the beginning, which goes farther from theorigin point with time.

Spatial clustering analysis was carried only on plots 3, 5 and 6where the damage was at the initial stage of plants growth. Fig. 6displays the initial spatial classification of LD plants. The spatial

Fig. 3. (a) Mean values of number of adult B. fusca caught weekly with pheromonetraps at. (b) Mean values of plants with the leaf damage (LD) infestation recordedweekly. (1a) and (2a) represent the peaks observed before theapparition of theinfestation peaks (1b) and (2b) respectively. The bars represent the standarddeviation error of the mean.

Table 4Mean time in weeks for transition between infestation types. It is given in theformat: mean time standard error of the mean (SEM). The symbol () means thatsuch transition was not observed. For the plot 4 and 5 see the annex.

Mean time SEMTransition Plot 1 Plot 2 Plot 3 Plot 6LD ! DH 6.34 0.12 6.42 0.17 6.75 1.60 6.49 0.11DH ! D 7.14 0.30 7.45 0.31 7.03 0.53 LD ! D 7.17 0.28 7.07 0.31 7.03 0.53 LD ! EH 7.23 0.08 8.28 0.16 7.77 0.20


clusters generated by patches of infested plants are dissimilar interms of the number of damaged plants at the start of observation;however the gap reduced as it approaches the last weeks. In plot 3,we initially observed a small number of damaged and scatteredplants; whereas plot 5 and 6 engorged big number of damagedplants. The cluster shape made by damaged plants is typicallyellipsoidal. With time, two phases were observed; during the firstphase, the number of clusters remained almost the same but theshape of the initial clusters and their respective size (number ofplants) changed gradually. During the second phase, new clustersarose possibly from the fragmentation of the initial clusters ratherthan the creation of isolated new entities.

3.4. Understanding plant contagion patterns

According to Table 5, the LD dynamics in plot 1, 2 and 4 followeda cellular automata law. It is demonstrated that, if four plants areinfested following the Moore neighborhood of contagion pattern,the plant at the central position is most likely to be infested duringthe subsequent week. However, in plot 3, 5 and 6 the algorithmreduced the initial neighborhood to the central cell, indicating thefailure to estimate the neighborhood in these cases. Such resultsare justified by the non-linear and perhaps chaotic behavior of thespread of the infestations of pests within plants at field level. Asummary of all the spatial results are provided in Table 6.

Table 3Conditional probability for a plant to die following different cases. P(D|i) is aprobability that a plant died, given that it has an infestation i. LD = leaf damage,EH = exit hole, DH = dead heart and D = dead.The symbol () means null probabilityand the symbol [ stands of or.

Probabilities

Events Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6

PDjLD 0.0116 0.0337 0 0 0 0PDjEH[ DH 0.0082 0.0303 0 0 0 0PDjEH[ DH[ LD 0.0102 0.0330 0 0 0 0

4. Discussions

4.1. Sampling scheme

Developed sampling schemes for assessing lepidopteran stemborer pest infestations in maize farms are often focusing onchecking/collecting at random a few plants infested/uninfested bythe larvae without taking into consideration the precise pointlocation of the plants within the field (Amoako-Atta et al., 1983;Ndemah et al., 2001; Overholt et al., 1994; Van Rensburg andPringle, 1989). In our study, all the plants had a precise geo-referenced position in each field. We did not used destructivesampling, as the dissection of plants can interfere with the spreadof the larvae and the oviposition distribution of the females in theplots. In addition, farmers perceive destructive sampling asunacceptable, and it is time consuming and expensive (Nyropet al., 1999). The presence/absence infestation data collected fromvisual inspection of external and comprehensive signs of insectpest damages in the plants may be more appropriate. Particularattention was given to leaf damage because young larvae are oftenresponsible for this type of damage and it occurs mainly on youngmaize plants; therefore making it easy to be used as a proxy forrelating the spread of the infestation. In addition, leaf damage hasalso been reported as an important factor in contributing to yieldlosses in maize farms (Kfir et al., 2002; Reddy and Sum, 1991).

4.2. Infestation rate dynamic

Considering that B. fusca egg development is completed during810 days at 2520 C (Khadioli et al., 2014), the first flight periodstarted at least 10 days before the first set of leaf damage, andended in all plots at least 10 days before the end of the 5th week.This suggests an absence of B. fusca female oviposition during days1015 days around week 56 as no damage was recorded in week 7in all plots. The second peak of damages is observed on all plotsbetween weeks 7 and 8. No exit holes were noticed in all plotsduring the second peak; which suggest that majority of eggs laidwithin this period of time was caused by females from otherlocalities. The observation of a fixed time lag between theoccurrence of peaks of LD and adult male abundance in pheromonetraps is in accordance with previous studies (Coop et al., 1992;Hillier et al., 2004; Masetti et al., 2015; Millar et al., 2002; Moriet al., 2014; Qureshi and Ahmed, 1991; Riedl and Croft, 1974;Thming et al., 2011; Tobin and Whitmire, 2005). The abundance ofmales in pheromone-baited traps was significantly correlated toobserved damages in studies (Coop et al., 1992; Hillier et al., 2004;Masetti et al., 2015; Millar et al., 2002; Mori et al., 2014; Qureshiand Ahmed,1991; Riedl and Croft,1974; Thming et al., 2011; Tobinand Whitmire, 2005). Additionally, the spacing of two weeks is theB. fusca egg development time, after which it emerges to larvastage and starts damage. It has been reported that B. fusca prefer tolay eggs in young maize plants (Calatayud et al., 2014; Harris andNwanze, 1992; Kfir et al., 2002); however, the continuous

https://www.researchgate.net/publication/231760233_Influence_of_maize_cowpea_and_sorghum_intercropping_systems_on_stem-pod-borer_infestations?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/23408867_Sampling_for_IPM_Decision_Making_Where_Should_We_Invest_Time_and_Resources?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/23408867_Sampling_for_IPM_Decision_Making_Where_Should_We_Invest_Time_and_Resources?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/266564145_Effect_of_temperature_on_the_life_history_parameters_of_noctuid_lepidopteran_stem_borers_Busseola_fusca_and_Sesamia_calamistis?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/240840184_A_study_of_pheromone_trap_catches_in_relation_to_codling_moth_Lepidoptera_Olethreutidae_damage?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/240840184_A_study_of_pheromone_trap_catches_in_relation_to_codling_moth_Lepidoptera_Olethreutidae_damage?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/233637897_Relationship_of_Corn_Earworm_Lepidoptera_Noctuidae_Pheromone_Catch_and_Silking_to_Infestation_Levels_in_Oregon_Sweet_Corn?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/233637897_Relationship_of_Corn_Earworm_Lepidoptera_Noctuidae_Pheromone_Catch_and_Silking_to_Infestation_Levels_in_Oregon_Sweet_Corn?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/228069084_Trap_Captures_of_Male_Grapholita_libertina_Lepidoptera_Tortricidae_Moths_Relationship_to_Larval_Numbers_and_Damage_in_Wild_Lingonberry1SUP?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/228069084_Trap_Captures_of_Male_Grapholita_libertina_Lepidoptera_Tortricidae_Moths_Relationship_to_Larval_Numbers_and_Damage_in_Wild_Lingonberry1SUP?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/266972355_Area-wide_monitoring_of_potato_tuberworm_Phthorimaea_operculella_by_pheromone_trapping_in_Northern_Italy_Phenology_spatial_distribution_and_relationships_between_catches_and_tuber_damage?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/266972355_Area-wide_monitoring_of_potato_tuberworm_Phthorimaea_operculella_by_pheromone_trapping_in_Northern_Italy_Phenology_spatial_distribution_and_relationships_between_catches_and_tuber_damage?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/11172071_Development_and_Optimization_of_Methods_for_Using_Sex_Pheromone_for_Monitoring_the_Mealybug_Planococcus_ficus_Homoptera_Pseudococcidae_in_California_Vineyards?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/11172071_Development_and_Optimization_of_Methods_for_Using_Sex_Pheromone_for_Monitoring_the_Mealybug_Planococcus_ficus_Homoptera_Pseudococcidae_in_California_Vineyards?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/264717413_Relationships_among_male_Coleophora_deauratella_Lepidoptera_Coleophoridae_pheromone-baited_trap_capture_larval_abundance_damage_and_flight_phenology?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/264717413_Relationships_among_male_Coleophora_deauratella_Lepidoptera_Coleophoridae_pheromone-baited_trap_capture_larval_abundance_damage_and_flight_phenology?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/264717413_Relationships_among_male_Coleophora_deauratella_Lepidoptera_Coleophoridae_pheromone-baited_trap_capture_larval_abundance_damage_and_flight_phenology?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/249449990_Monitoring_seasonal_population_fluctuation_of_spotted_and_spiny_bollworms_by_synthetic_sex_pheromones_and_its_relationships_to_boll_infestation_in_cotton?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/249449990_Monitoring_seasonal_population_fluctuation_of_spotted_and_spiny_bollworms_by_synthetic_sex_pheromones_and_its_relationships_to_boll_infestation_in_cotton?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/249449990_Monitoring_seasonal_population_fluctuation_of_spotted_and_spiny_bollworms_by_synthetic_sex_pheromones_and_its_relationships_to_boll_infestation_in_cotton?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/262002147_Risk_assessment_of_pea_moth_Cydia_nigricana_infestation_in_organic_green_peas_based_on_spatio-temporal_distribution_and_phenology_of_host_plant?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/262002147_Risk_assessment_of_pea_moth_Cydia_nigricana_infestation_in_organic_green_peas_based_on_spatio-temporal_distribution_and_phenology_of_host_plant?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/274509631_Ecology_of_the_African_Maize_Stalk_Borer_Busseola_fusca_Lepidoptera_Noctuidae_with_Special_Reference_to_Insect-Plant_Interactions?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/11627000_Biology_and_Management_of_Economically_Important_Lepidopteran_Cereal_Stem_Borers_in_Africa?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/11627000_Biology_and_Management_of_Economically_Important_Lepidopteran_Cereal_Stem_Borers_in_Africa?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/283593398_Spread_of_gypsy_moth_Lepidoptera_Lymantriidae_and_its_relationship_to_defoliation?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/283593398_Spread_of_gypsy_moth_Lepidoptera_Lymantriidae_and_its_relationship_to_defoliation?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/283593398_Spread_of_gypsy_moth_Lepidoptera_Lymantriidae_and_its_relationship_to_defoliation?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/259379438_Determination_of_Economic_Injury_Level_of_the_Stem_Borer_Chilo_Partellus_Swinhoe_in_Maize_Zea_Mays_L?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==

Fig. 4. Spatial and temporal autocorrelation for leaf damage (LD) on plants. The computation has been done for a radius of proximity (r) from 1 m to 20 m. The color bar on theright side of each figure represents the spatial autocorrelation level. For plot 1 and 4 see the annex.


infestation pattern observed in this study suggests that ovipositionon older plants is possible.

Although the main goal of this study was to examine the spatialand temporal spread of maize stem borer B. fusca under fieldconditions, we though necessary to comprehend the sequence ofplants infestations by estimating the conditional probabilities ofeach damage type causing death of the plant. A higher mortality ofmaize plants with LD was observed. However, the stem tunnelingwhich is logically expected to occur after LD and before DH, D, EHwas not considered here. Moreover, the data collection protocol ofthe present study was unsuitable for assessing stem tunneling,which effect is a major cause of grain yield reduction in maize(Ajala and Saxena, 1994). It might be convenient to sample anddissect a few plants to assess and account for the incidence oftunneling.

The mean transition time from a type of damage to anotherdamage type was different from one plot to another. This suggestseither a variability of the phenological stage of the plants due to thedifference in sowing time in each plot and/or the heterogeneouscharacteristics of the soil and nutrients managements, which weredifferent from one plot to another. Variability might be also due tothe phenological stage of the stem borer larvae and the differencein nutritional quality of the plants. Temperatures difference may aswell play a part in the discrepancy.

4.3. Spatial and temporal scattering of infestation

An increase of the spatial correlation with time implies anoverall trend toward a single gradient. The results of independenceof damaged plants noticed via negative spatial autocorrelations insome plots are likely an artifact and not a biological phenomenon.

The variogram is expected to depict no spatial autocorrelation forsome distances (Blackshaw and Hicks, 2013; Bone et al., 2013; Cocuet al., 2005; Smith et al., 2004). During data collection, indepen-dent samples must be taken at random at different locations sothat each sample has an equal chance to be selected (Pedigo, 1999).However, infested plants separated by a small radius of proximity(less than 10 m) were considerably correlated. Therefore, suchevidence of a spatial link between infested plants leads to theconclusion that sequential plant samples may not be a properrepresentative of a statistically independent sample procedure,thus violating the assumptions of random sampling in the field(Pedigo,1999). In improving traditional sampling schemes devotedto lepidopteran stem borer infestation through a selection ofsamples far enough from each other to ensure total independence,this finding should be taken into account.

The movements of the focal area (centroid) associated withinfestations might reflect shifts in the spatial and temporalpatterns of adult and larval dynamics during insect damageprocess. In literature, centroid infestation tracking was used tomonitor the movement of Eoreuma loftini with pheromones baitedtraps throughout the Texas rice belt (Reay-Jones et al., 2007), tocheck the spatial evolution of Nasonavia ribisnigri after release offew adults at the center of a lettuce field (Diaz et al., 2012), andquantifying spatio-temporal infestation patterns of Ips typogra-phus in the Bavarian Forest National Park through dead woodpatches surveillance (Lausch et al., 2013). These studies providedan approximate speed of leading edge of rice borer infestationmovement (Reay-Jones et al., 2007), ability of the aphid to spreadfrom a source plant of release over time in a green house(Diaz et al., 2012) and spatio-temporal dispersion patterns of barkbeetle damages on trees on a wide area scale over years

https://www.researchgate.net/publication/290791657_Interrelationship_among_Chilo_partellus_Swinhoe_damage_parameters_and_their_contribution_to_grain_yield_reduction_in_maize_Zea_mays_L?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/257496322_Distribution_of_adult_stages_of_soil_insect_pests_across_an_agricultural_landscape?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/236455471_A_GIS-Based_Risk_Rating_of_Forest_Insect_Outbreaks_Using_Aerial_Overview_Surveys_and_the_Local_Moran's_I_Statistic?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/221790632_Interplant_movement_and_spatial_distribution_of_alate_and_apterous_morphs_of_Nasonovia_ribisnigri_Homoptera_Aphididae_on_lettuce?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/221790632_Interplant_movement_and_spatial_distribution_of_alate_and_apterous_morphs_of_Nasonovia_ribisnigri_Homoptera_Aphididae_on_lettuce?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/43260929_Dispersal_and_Spatiotemporal_Dynamics_of_Asian_Longhorned_Beetle_Coleoptera_Cerambycidae_in_China?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/229874627_Spatial_autocorrelation_for_identifying_the_geographical_patterns_of_aphid_annual_abundance?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/229874627_Spatial_autocorrelation_for_identifying_the_geographical_patterns_of_aphid_annual_abundance?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/236216904_Spatio-temporal_infestation_patterns_of_Ips_typographus_L_in_the_Bavarian_Forest_National_Park_Germany?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/6436329_Movement_of_Mexican_Rice_Borer_Lepidoptera_Crambidae_Through_the_Texas_Rice_Belt?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/6436329_Movement_of_Mexican_Rice_Borer_Lepidoptera_Crambidae_Through_the_Texas_Rice_Belt?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==

Fig. 5. Temporal evolution of the isobarycentre (IB) for leaf damaged infested plants. The integer values represent the corresponding week. The position of the number is thespatial position of the IB. The IB from the first and second cycle are colored in blue and red respectively. For plot 4 and 5 see the annex. (For interpretation of the references tocolour in this figure legend, the reader is referred to the web version of this article.)


(Lausch et al., 2013), respectively. A similar method was appliedhere, but with regular monitoring of all plants within the fields andprecise damage types were captured. Such consistency in theexperiment helped us to detect that spatial and temporal patternof B. fusca infestations had a geometrical design under a form of aspiral around initial patches of damaged plants. Identifying suchan infestation pattern can be difficult or impossible with previousapproaches (Diaz et al., 2012; Lausch et al., 2013; Reay-Jones et al.,2007).

It has been reported that some females insects such as Deliaradicum on cabbage plants (Baur et al., 1996), Eurosta solidaginis onlate golden rod plants (Craig et al., 2000), Lygocoris pabulinus onpotato plants (Groot et al., 2003) and Anthocoris nemorum onapple/pear trees (Sigsgaard, 2004), oviposit where conspecificshave already oviposited. The contrary has been reported forTrichoplusia ni on cotton plants (Landolt, 1993), Narnia femorata oncactus plants (Miller et al., 2013), Heliothis virescens on tobacco

plants (De Moraes et al., 2001) and Pieris rapae on crucifer plant(Sato et al., 1999). In the present study, results of spatial clusteringsuggest that B. fusca females tend to oviposition patches of maizeplant already infested; and thus, exhibit the fisrt behavorialcharacteristic. The homogenization of distribution inside eachcluster might be due to overcrowding during plant host selection(Schoonhoven et al., 2005; Thompson and Pellmyr, 1991).Furthermore, a strong instability (drastic changes in clustersshapes and number of plants inside each spatial cluster) during theevolution of initial damage distributions might be due to acombination of adult moth oviposition and larval interplantmovements (Calatayud et al., 2014; Harris and Nwanze, 1992;Van Rensburg et al., 1987).

Spatial pattern dynamics that insect pests generated can beused to improve sampling strategy, site-specific pest management,and sowing distribution. However, when agricultural practices areundertaken, the precise spatial and temporal dynamics of the

https://www.researchgate.net/publication/221790632_Interplant_movement_and_spatial_distribution_of_alate_and_apterous_morphs_of_Nasonovia_ribisnigri_Homoptera_Aphididae_on_lettuce?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/227142786_Preference_for_plants_damaged_by_conspecific_larvae_in_ovipositing_cabbage_root_flies_Influence_of_stimuli_from_leaf_surface_and_roots?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/12053020_Caterpillar-induced_nocturnal_plant_volatiles_repel_conspecific_females?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/230333390_Oviposition_preference_of_Lygocoris_pabulinus_Het_Miridae_in_relation_to_plants_and_conspecifics?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/281771434_Pieris_rapae_Lepidoptera_Pieridae_females_avoid_oviposition_on_Rorippa_indica_plants_infested_by_conspecific_larvae?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/274509631_Ecology_of_the_African_Maize_Stalk_Borer_Busseola_fusca_Lepidoptera_Noctuidae_with_Special_Reference_to_Insect-Plant_Interactions?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/249432697_The_influence_of_host_plant_variation_and_intraspecific_competition_on_oviposition_preference_in_the_host_races_of_Eurosta_solidaginis?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/236216904_Spatio-temporal_infestation_patterns_of_Ips_typographus_L_in_the_Bavarian_Forest_National_Park_Germany?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/236216904_Spatio-temporal_infestation_patterns_of_Ips_typographus_L_in_the_Bavarian_Forest_National_Park_Germany?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/227838144_Oviposition_preference_of_Anthocoris_nemorum_and_A_nemoralis_for_apple_and_pear?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/296353059_Evolution_of_oviposition_behavior_and_host_preference_in_Lepidoptera?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/226441022_Effects_of_host_plant_leaf_damage_on_Cabbage_Looper_moth_attraction_and_oviposition?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/231750377_Ecology_of_the_maize_stalk_borer_Busseola_fusca_Fuller_Lepidoptera_Noctuidae?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/249967859_Conspecific_and_Heterospecific_Cues_Override_Resource_Quality_to_Influence_Offspring_Production?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/6436329_Movement_of_Mexican_Rice_Borer_Lepidoptera_Crambidae_Through_the_Texas_Rice_Belt?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/6436329_Movement_of_Mexican_Rice_Borer_Lepidoptera_Crambidae_Through_the_Texas_Rice_Belt?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==

Fig. 6. Spatial clustering of plants with LD infestation for plot 6 at week 4 (a), 9 (b), 11 (c) and 13 (d). Each color represents a spatial cluster. The ellipsoids/circles represent thebottom of two-dimensional probability density functions (PDF). The shapes are chosen according to the Bayesian information criteria. The centers represent the centroid ofthe cluster distributions. The dot outside the circular represents plants positions with weak probabilities compared to the center of the PDF. For plot 3 and 5 see the annex.


damages are usually not well known (Wang et al., 2009). Moreover,most of the existing statistical techniques used in similar agro-ecological frameworks than the present study are not accurateenough to detect a precise plant-to-plant contagion patterns(Aukema et al., 2006; Blackshaw and Hicks, 2013; Bone et al., 2013;

Table 5Results obtained after estimations of the neighborhood configuration of the cellular auto3.7. The number represents the number of plants in the neighborhood (including the c

Plots Initial number ofneighbors

Number after reduction made duringstep (ii)

Numberstep (iii)

Plot 1 25 25 4 Plot 2 25 25 4 Plot 3 25 25 1 Plot 4 25 25 4 Plot 5 25 25 1 Plot 6 25 25 1

Cocco et al., 2013; Cocu et al., 2005; Dder et al., 2015; Diaz et al.,2012; Emmen et al., 2004; Fernandes et al., 2015; Holland et al.,2005; Ishaaya and Horowitz, 2004; Kautz et al., 2011; Keasar et al.,2005; Kim et al., 2007; Lausch et al., 2013; Moral Garca, 2006;Perry, 1994; Reay-Jones, 2010; Reay-Jones et al., 2007; Reineke

mata (CA). Step (ii), (iii) and (iv) are briefly describe in the text, in the methodologyentral cell).

after reduction made during Maximal Bayesian information criteria after step(iv)

441411

https://www.researchgate.net/publication/257496322_Distribution_of_adult_stages_of_soil_insect_pests_across_an_agricultural_landscape?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/236455471_A_GIS-Based_Risk_Rating_of_Forest_Insect_Outbreaks_Using_Aerial_Overview_Surveys_and_the_Local_Moran's_I_Statistic?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/268452941_Control_of_insect_vectors_and_plant_viruses_in_protected_crops_by_novel_pyrethroid-treated_nets?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/221790632_Interplant_movement_and_spatial_distribution_of_alate_and_apterous_morphs_of_Nasonovia_ribisnigri_Homoptera_Aphididae_on_lettuce?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/221790632_Interplant_movement_and_spatial_distribution_of_alate_and_apterous_morphs_of_Nasonovia_ribisnigri_Homoptera_Aphididae_on_lettuce?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/225356710_Spatio-temporal_dynamics_of_Scotinophara_lurida_Hemiptera_Pentatomidae_in_rice_fields?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/227636721_Landscape_level_analysis_of_mountain_pine_beetle_in_British_Columbia_Canada_Spatiotemporal_development_and_spatial_synchrony_within_the_present_outbreak?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/277918272_A_new_and_highly_effective_sampling_plan_using_attractant-baited_traps_for_the_coffee_berry_borer_Hypothenemus_hampei?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/229858631_Farm-scale_spatiotemporal_dynamics_of_predatory_beetles_in_arable_crops?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/229858631_Farm-scale_spatiotemporal_dynamics_of_predatory_beetles_in_arable_crops?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/232413826_Analysis_of_the_Spatio-temporal_Distribution_of_Helicoverpa_armigera_Hb_in_a_Tomato_Field_using_a_Stochastic_Approach?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/235652662_Control_of_Tuta_absoluta_Meyrick_Lepidoptera_Gelechiidae_ingreenhouse_tomato_crops_using_the_mating_disruption_technique?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/229874627_Spatial_autocorrelation_for_identifying_the_geographical_patterns_of_aphid_annual_abundance?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/236216904_Spatio-temporal_infestation_patterns_of_Ips_typographus_L_in_the_Bavarian_Forest_National_Park_Germany?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/291891214_Sampling_and_applied_statistics_for_pests_and_diseases?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/44675831_Spatial_and_Temporal_Patterns_of_Stink_Bugs_Hemiptera_Pentatomidae_in_Wheat?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/241444825_Research_of_spatio-temporal_analysis_of_agricultural_pest?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/7968310_Spatial_and_Temporal_Dynamics_of_Potato_Tuberworm_Lepidoptera_Gelechiidae_Infestation_in_Field-Stored_Potatoes?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/7968310_Spatial_and_Temporal_Dynamics_of_Potato_Tuberworm_Lepidoptera_Gelechiidae_Infestation_in_Field-Stored_Potatoes?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/289409404_Temporal_and_spatial_dynamics_of_Empoasca_fabae_Harris_Homoptera_Cicadellidae_in_alfalfa?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/6436329_Movement_of_Mexican_Rice_Borer_Lepidoptera_Crambidae_Through_the_Texas_Rice_Belt?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==https://www.researchgate.net/publication/233356198_Insect_Pest_Management_-_Field_and_Protected_Crops?el=1_x_8&enrichId=rgreq-80e54525e315c07168090c2ac1354ae5-XXX&enrichSource=Y292ZXJQYWdlOzMwOTMxOTIwNDtBUzo0NDkyODQzMzczNDQ1MTJAMTQ4NDEyOTE0MTA2Mg==

Table 6A summary of the purpose of applying each spatial analysis symbols x and U stand for no and yes respectively.

Method Purpose of the method Main result

Evaluates with time thespatial link betweeninfested plants located atdifferent positions

Tracks the positions ofthe center of patchesformed by infestedplants with time

Follows the evolution of the initialshape, the density and the numberof clusters of infested plantsformed with time

Estimates the rulesby which plants getinfected throughneighbors

Morans' I U x x x The spatial correlation between infestedplants is considerable at a radius of 10 m

Centertracking

x U x x The spatial shift of the patterns created byan infested plant moves with time in a formof a spiral from an origin point to otherparts of the plot

Modelbasedclusteranalysis

x x U x The initial shape and number of spatialclusters remain stable during a certainperiod of time and, later disaggregate tocreate new clusters

Cellularautomata

x x x U A safe plants surrounded by four infectedplants is most likely to become damaged inthe next subsequent week

Symbols x and U stand for no and yes respectively.


et al., 2011; Rogers et al., 2015; Schumann et al., 2014; Smith et al.,2004; Thomas et al., 2001; Wang et al., 2009; Wright et al., 2002;Zhang et al., 2016; Zimmerman et al., 2004). The reasons reposedon the fact that the methodologies used in previous studiesfocused on two main aspects: analyzing the count-variance ofsamples without taking into account the spatial point locations(indices of dispersion, aggregation and clumping indices, Taylorpower law, quadrat-variance methods, probability distributionmodels); and/or focusing on count-variance on a few samples andtheir locations (semi-variography, spatial autocorrelation analysis,local spatial statistics) (Dale et al., 2002; Ishaaya and Horowitz,2004; Liebhold and Gurevitch, 2002; Moran, 1950; Perry et al.,2006, 2002; Vinatier et al., 2011; Wiegand and Moloney, 2014). Togain a considerable amount of information from spatial data, theapplication of several methodologies is recommended (Perry et al.,2002). Previous studies failed to report some of our findingsbecause they usually applied a single method (Aukema et al., 2006;Blackshaw and Hicks, 2013; Bone et al., 2013; Cocco et al., 2013;Cocu et al., 2005; Dder et al., 2015; Diaz et al., 2012; Emmen et al.,2004; Fernandes et al., 2015; Holland et al., 2005; Ishaaya andHorowitz, 2004; Kautz et al., 2011; Keasar et al., 2005; Kim et al.,2007; Lausch et al., 2013; Moral Garca, 2006; Perry, 1994; Reay-Jones, 2010; Reay-Jones et al., 2007; Reineke et al., 2011; Rogerset al., 2015; Schumann et al., 2014; Smith et al., 2004; Thomas et al.,2001; Wang et al., 2009; Wright et al., 2002; Zhang et al., 2016;Zimmerman et al., 2004). In choosing to use several approachesthat have not been used in the literature before (such as centroidtracking and spatial clustering in order to track easily the shift ofspatial patches and their relative stability, respectively), thecurrent study was more precise. In addition, the cellular automatamethod was used in our study to investigate the likely arrange-ment of infested plants to contaminate healthy maize plants, goesbeyond the classical analyses.

4.4. Advanced in movements ecology of B. fusca

As an attempt to advance the knowledge on movement ecology,the quantification and qualification of the delocalization pattern ofthe damages spread of B. fusca with time were studied. Weexamined whether the infestation is initiated from an origin pointfrom which it spreads to other locations to create a big cluster ofplants damages, or whether multiple damages of plants eruptedsimultaneously in different locations to create clusters, which

merged with time. Through the record of point location forindividual infested plants at each week and their respectivecentroid, it is noticed that the shift of the focal point moves as aspiral centered around the first set of damaged plants. Theformation of new spiral during the second cycle of damages couldbe explained by the ability of B. fusca to distinguish patches oflarvae/eggs and oviposit in different zones. Practically, whileprobing for more host plant and flying toward adjacent stems, thepest female is showing preference to the areas with less density ofinfection, thus spreading their offsprings more widely by creating anew spiral at a distant radius. This result corroborates thehypothesis stating that a lepidopteran is able to adopt a regularpattern during host selection (Thompson and Pellmyr, 1991).

We observed that the selection of the first plants to be infestedby the B. fusca female is likely to be random; subsequently theneighbors to the infected plants are most likely to be infected. Inaddition, it is noticed that the initial infestations occurredsimultaneously in numerous plants. The enhancement of thespatial expansion of damage clusters can be explained by the pestlarvae short-range movement ability to migrate from a maize plantto another (Calatayud et al., 2014). This displacement is consideredas a survival mechanism; because of competition, the larvae of B.fusca are likely to die in a highly infested maize stem (Ntiri et al.,2016). Although B. fusca females have the ability to fly over severalkilometers (Campagne et al., 2015; Dupas et al., 2014), it wasobserved that the number of plants infested within a clusterincrease progressively and, the shape of the cluster was relativelymaintained with time; this can be explained by the ability oflepidopteran to visit patches of plants already infested by theirconspecifics (Thompson and Pellmyr, 1991). The competitionamong B. fusca female for the same set of host plants duringoviposition may be further considered as a key factor thatinfluences the appearance of the patterns observed in this study.

4.5. Implications for pest management

4.5.1. OverviewControl of negative impa

Documents

Spatial and temporal spread of maize stem borer Busseola ... · fusca (Fuller) is the most damaging in the high potential yield areas, which include the highland tropics and moist