Automating Crop Damage Assessments with Unmanned Aerial Systems and Machine Learning

Agricultural pests are responsible for millions of dollars in crop losses and management costs every year. In order to implement optimal site-specific treatments and reduce control costs, new methods to accurately monitor and assess pest damage need to be investigated. In this paper we explore the combination of unmanned aerial vehicles (UAV), remote sensing and machine learning techniques as a promising methodology to address this challenge. The deployment of UAVs as a sensor platform is a rapidly growing field of study for biosecurity and precision agriculture applications. In this experiment, a data collection campaign is performed over a sorghum crop severely damaged by white grubs (Coleoptera: Scarabaeidae). The larvae of these scarab beetles feed on the roots of plants, which in turn impairs root exploration of the soil profile. In the field, crop health status could be classified according to three levels: bare soil where plants were decimated, transition zones of reduced plant density and healthy canopy areas. In this study, we describe the UAV platform deployed to collect high-resolution RGB imagery as well as the image processing pipeline implemented to create an orthoimage. An unsupervised machine learning approach is formulated in order to create a meaningful partition of the image into each of the crop levels. The aim of this approach is to simplify the image analysis step by minimizing user input requirements and avoiding the manual data labelling necessary in supervised learning approaches. The implemented algorithm is based on the K-means clustering algorithm. In order to control high-frequency components present in the feature space, a neighbourhood-oriented parameter is introduced by applying Gaussian convolution kernels prior to K-means clustering. The results show the algorithm delivers consistent decision boundaries that classify the field into three clusters, one for each crop health level as shown in Figure 1. The methodology presented in this paper represents a venue for further esearch towards automated crop damage assessments and biosecurity surveillance.

Nocturnal Perching of Scarabaeine Dung Beetles (Coleoptera, Scarabaeidae) in an Australian Tropical Rain Forest

Wongabel, a northeastern Queensland tropical, wet, evergreen forest, contains 22 species of Scarabaeinae dung beetles. Five of these species were observed to perch commonly on leaves at night. Length of the beetle and the height of its perch were recorded for each of 56 1 specimens. Unlike the New World tropical dung beetle perchers, no clear evidence was found that small species perched closer to the ground than larger species. The evidence gathered, at least for the four most common perchers, supports the hypothesis that perching is one type of foraging strategy. The similarities and differences between the Australian and New World perchers are discussed.

Identification of the Sex Pheromone of Holotrichia Reynaudi

The male attractant pheromone of the scarab beetle Holotrichia reynaudi, an agricultural pest native to southern India, was extracted from abdominal glands of females with hexane and analyzed by gas chromatography– mass spectrometry. Field testing of the candidate chemicals, indole, phenol, and anisole, both alone and as binary mixtures, led us to conclude that anisole was the major component of the sex pheromone. Neither male nor female beetles were attracted to indole or phenol on their own. Similarly, when indole and anisole were combined, the attractiveness of the solution did not increase over that obtained with anisole alone. However, combination of phenol and anisole did alter the attractiveness of anisole, with fewer male beetles attracted to the binary mixture than to anisole on its own. The behavior of female beetles was not altered by any of the chemicals tested. Anisole is also the sex pheromone of H. consanguinea, making this the first known example of two melolonthine scarabs sharing the same pheromone.

Damage potential of two Scarab species on groundnut

The damage potential of two phytophagous scarab larvae on groundnut (peanut) yield was determined. Holotrichia serrata, a root and pod feeding species from southern India, was studied in microplots while the damage potential of Heteronyx piceus, a pod feeder from Queensland, Australia, was determined by analysis of on-farm chemical-rate trials. H. serrata larva reduced groundnut yield by an average of 7.52 g/ larva. In crops yielding less and more than 1900 kg ha-1, H. piceus reduced yield by 4.20 g and 1.43 g/ larva, respectively. These damage potential estimates were used to determine provisional economic injury levels (EIL). For H. piceus, the provisional EIL is 1.67 and 4.91 larvae/ row-metre in crops yielding less and more than 1900 kg/ha, respectively. For H. serrata, the provisional EIL is one H. serrata larva in 7.1 m2. As more than 70% of southern India groundnut fields have Holotrichia populations greater than 1 larva in 1.35 m2, more widespread use of chlorpyrifos seed dressing of groundnut is likely to produce regional economic benefits.

Impact of resistance on the efficacy of binary combinations of spinosad, chlorpyrifos-methyl and s-methoprene against five stored-grain beetles

Laboratory experiments were conducted to determine the efficacy of spinosad (a biopesticide), chlorpyrifos-methyl (an organophosphorus compound (OP)) and s-methoprene (a juvenile hormone analogue) applied alone and in binary combinations against five stored-grain beetles in wheat. There were three strains of Rhyzopertha dominica, and one strain each of Sitophilus oryzae, Tribolium castaneum, Oryzaephilus surinamensis and Cryptolestes ferrugineus. These strains were chosen to represent a range of possible resistant genotypes, exhibiting resistance to organophosphates, pyrethroids or methoprene. Treatments were applied at rates that are registered or likely to be registered in Australia. Adults were exposed to freshly treated wheat for 2 weeks, and the effects of treatments on mortality and reproduction were determined. No single protectant or protectant combination controlled all insect strains, based on the criterion of >99% reduction in the number of live F1 adults relative to the control. The most effective combinations were spinosad at 1 mg kg-1+chlorpyrifos-methyl at 10 mg kg-1 which controlled all strains except for OP-resistant O. surinamensis, and chlorpyrifos-methyl at 10 mg kg-1+s-methoprene at 0.6 mg kg-1 which controlled all strains except for methoprene-resistant R. dominica. The results of this study demonstrate the difficulty in Australia, and potentially other countries which use protectants, of finding protectant treatments to control a broad range of pest species in the face of resistance development.

Survival of Campylobacter spp. in darkling beetles (Alphitobius diaperinus) and their larvae in Australia

Campylobacter infection is the most frequently reported notifiable food-borne disease in humans in Australia. Our studies investigated the persistence of Campylobacter spp. in or on darkling beetles (Alphitobius diaperinus) and their larvae. Our results in analyses with chickens confirm that, unless very short turnaround times are used (<72 h), beetles colonized in one production cycle (i.e., one batch of chickens) are most unlikely to still be colonized during the next cycle of chickens.

Paropsine beetles (Coleoptera: Chrysomelidae) in south-eastern Queensland hardwood plantations: identifying potential pest species

Paropsine chrysomelid beetles are significant defoliators of Australian eucalypts. In Queensland, the relatively recent expansion of hardwood plantations has resulted in the emergence of new pest species. Here I identify paropsine beetles collected from Eucalyptus cloeziana Muell. and E. dunnii Maiden, two of the major Eucalyptus species grown in plantations in south-eastern Queensland, and estimate the relative abundance of each paropsine species. Although I was unable to identify all taxa to species level, at least 17 paropsine species were collected, about one-third of which have not been previously associated with hardwood plantations. Paropsis atomaria Olivier and P. charybdis Stål were the most abundant species collected in E. cloeziana plantations, while Chrysophtharta cloelia (Stål) and P. atomaria were most commonly collected from E. dunnii. Occasional collections from Corymbia citriodora (Hook.) Hill and Johns, ssp. variegata revealed an additional four species implicated in plantation damage. Abundance and voltinism varied between species and sites. I predict which paropsine species are likely to threaten plantation productivity.

Sublethal Exposure to Phosphine Decreases Offspring Production in Strongly Phosphine Resistant Female Red Flour Beetles, Tribolium castaneum (Herbst)

The red flour beetle is a cosmopolitan pest of stored grain and stored grain products. The pest has developed resistance to phosphine, the primary chemical used for its control. The reproductive output of survivors from a phosphine treatment is an important element of resistance development but experimental data are lacking. We exposed mated resistant female beetles to 0.135 mg/L of phosphine for 48 h at 25°C. Following one week of recovery we provided two non-exposed males to half of the phosphine exposed females and to half of the non-exposed control females. Females that had been exposed produced significantly fewer offspring than non-exposed females. Females that remained isolated produced significantly fewer offspring than both exposed females with access to males and non-exposed controls (P<0.05). Some females were permanently damaged from exposure to phosphine and did not reproduce even when given access to males. We also examined the additional effects of starvation prior to phosphine exposure on offspring production. Non-exposed starved females experienced a small reduction in mean offspring production in the week following starvation, followed by a recovery in the second week. Females that were starved and exposed to phosphine demonstrated a very significant reduction in offspring production in the first week following exposure which remained significantly lower than that of starved non-exposed females (P<0.05). These results demonstrate a clear sublethal effect of phosphine acting on the female reproductive system and in some individuals this can lead to permanent reproductive damage. Pest population rebound after a fumigation may be slower than expected which may reduce the rate of phosphine resistance development. The results presented strongly suggest that phosphine resistance models should include sublethal effects. © 2012 Ridley et al.

An inordinate fondness for Fusarium: Phylogenetic diversity of fusaria cultivated by ambrosia beetles in the genus Euwallacea on avocado and other plant hosts

Ambrosia beetle fungiculture represents one of the most ecologically and evolutionarily successful symbioses, as evidenced by the 11 independent origins and 3500 species of ambrosia beetles. Here we document the evolution of a clade within Fusarium associated with ambrosia beetles in the genus Euwallacea (Coleoptera: Scolytinae). Ambrosia Fusarium Clade (AFC) symbionts are unusual in that some are plant pathogens that cause significant damage in naive natural and cultivated ecosystems, and currently threaten avocado production in the United States, Israel and Australia. Most AFC fusaria produce unusual clavate macroconidia that serve as a putative food source for their insect mutualists. AFC symbionts were abundant in the heads of four Euwallacea spp., which suggests that they are transported within and from the natal gallery in mandibular mycangia. In a four-locus phylogenetic analysis, the AFC was resolved in a strongly supported monophyletic group within the previously described Cade 3 of the Fusarium solani species complex (FSSC). Divergence-time estimates place the origin of the AFC in the early Miocene similar to 21.2 Mya, which coincides with the hypothesized adaptive radiation of the Xyleborini. Two strongly supported clades within the AFC (Clades A and B) were identified that include nine species lineages associated with ambrosia beetles, eight with Euwallacea spp. and one reportedly with Xyleborus ferrugineus, and two lineages with no known beetle association. More derived lineages within the AFC showed fixation of the clavate (club-shaped) macroconidial trait, while basal lineages showed a mix of clavate and more typical fusiform macroconidia. AFC lineages consisted mostly of genetically identical individuals associated with specific insect hosts in defined geographic locations, with at least three interspecific hybridization events inferred based on discordant placement in individual gene genealogies and detection of recombinant loci. Overall, these data are consistent with a strong evolutionary trend toward obligate symbiosis coupled with secondary contact and interspecific hybridization. (C) 2013 Elsevier Inc. All rights reserved.

Gut-inhabiting yeasts from the gut of Australian passalid beetles

Passalidae (Polyphaga, Coleoptera) is a family of beetles with approximately 960 species that are distributed worldwide. Preliminary studies that characterized ascomycete and basidiomycete yeasts in the gut of these wood-eating beetles from the USA, Guatemala, and Thailand, demonstrated associations between certain yeast taxa and passalids. We extended the study to include yeasts and beetles from tropical forests near Cairns and Brisbane, Queensland, Australia. We isolated more than 1000 yeast strains from about 150 beetles belonging to 10 species. LSU and ITS rRNA markers were used to identify a subset of 250 yeast strains, which revealed that the gut of Australian passalids contained undescribed ascomycetes in the Debaryomyces, Dipodascus, Kazachstania, Ogataea, Scheffersomyces, Sugiyamaella, Spathaspora, Torulaspora, and Zygowilliopsis clades, as well as basidiomycetes in the genera Cryptococcus and Trichosporon. A close relative of Candida subhashii (Spathaspora clade) and the xylose-fermenting yeast Scheffersomyces stipitis were the most common species isolated in Queensland. These results agree with those of previous studies that showed a common association of xylose-fermenting yeasts in the gut of lignicolous insects. Species and higher taxa of yeasts, however, vary between Queensland passalids and those previously collected in distant regions.

Pheromone-trapping of Carpophilus spp. (Coleoptera: Nitidulidae) in stone fruit orchards near Gosford, New South Wales: Fauna, seasonality and effect of insecticides

Traps baited with synthetic aggregation pheromones of Carpophilus hemipterus (L.), Carpophilus mutilatus Erichson and Carpophilus davidsoni Dobson and fermenting bread dough were used to identify the fauna and monitor the seasonal abundance of Carpophilus spp. in insecticide treated peach and nectarine orchards in the Gosford area of coastal New South Wales. In four orchards 67 178 beetles were trapped during 1994–1995, with C. davidsoni (82%) and Carpophilus gaveni (Dobson) (12.2%) dominating catches. Five species (C. hemipterus, C. mutilatus, Carpophilus marginellus Motschulsky, Carpophilus humeralis (F.) and an unidentified species) each accounted for 0.2–3.2% of trapped beetles. Carpophilus davidsoni was most abundant during late September–early October but numbers declined rapidly during October, usually before insecticides were applied. Spring populations of Carpophilus spp. were very large in 1994–1995 (1843–2588 per trap per week). However, despite a preharvest population decline of approximately 95% and 2–11 applications of insecticide, 14–545 beetles per trap per week (above the arbitrary fruit damage threshold of 10 beetles per trap per week) were recorded during the harvest period and fruit damage occurred at three of the four orchards. Lower preharvest populations in 1995–1996 (< 600 per trap per week) and up to six applications of insecticide resulted in < 10 beetles per trap per week during most of the harvest period and minimal or no fruit damage. The implications of these results for the integrated management of Carpophilus spp. in coastal and inland areas of southeastern Australia are discussed.

An improved coattractant for pheromone trapping of Carpophilus spp. (Coleoptera: Nitidulidae)

Fermenting apple juice (FAJ) contained within polyacrylamide granules was an effective pheromone coattractant for Carpophilus davidsoni in trapping experiments conducted in stone fruit orchards in southern New South Wales. Fermenting apple juice-baited traps captured as many beetles as traps baited with the 'standard' coattractant fermenting bread dough (FBD), either alone or in combination with aggregation pheromone. Increasing the interval of FAJ replacement to 2 weeks instead of 1 week, as is necessary for FBD, did not reduce trapping efficiency. Replacement of FAJ every three weeks did not affect captures of C. davidsoni in one experiment but did reduce captures of Carpophilus mutilatus. In a second experiment, captures of C. davidsoni were also reduced. Fermenting apple juice contained within polyacrylamide granules replaced at fortnightly intervals is an effective, convenient and practical pheromone coattractant for Carpophilus spp.

Phenology of Carpophilus spp. (Coleoptera: Nitidulidae)in stone fruit orchards as determined by pheromone trapping: Implications for prediction of crop damage

Traps baited with synthetic aggregation pheromone and fermenting bread dough were used to monitor seasonal incidence and abundance of the ripening fruit pests, Carpophilus hemipterus (L.), C. mutilatus Erichson and C. davidsoni Dobson in stone fruit orchards in the Leeton district of southern New South Wales during five seasons (1991-96). Adult beetles were trapped from September-May, but abundance varied considerably between years with the amount of rainfall in December-January having a major influence on population size and damage potential during the canning peach harvest (late February-March). Below average rainfall in December-January was associated with mean trap catches of < 10 beetles/trap/week in low dose pheromone traps during the harvest period in 1991/92 and 1993/94 and no reported damage to ripening fruit. Rainfall in December-January 1992/93 was more than double the average and mean trap catches ranged from 8-27 beetles/week during the harvest period with substantial damage to the peach crop. December-January rainfall was also above average in 1994/95 and 1995/96 and means of 50-300 beetles/trap/week were recorded in high dose pheromone traps during harvest periods. Carpophilus spp. caused economic damage to peach crops in both seasons. These data indicate that it may be possible to predict the likelihood of Carpophilus beetle damage to ripening stone fruit in inland areas of southern Australia, by routine pheromone-based monitoring of beetle populations and summer temperatures and rainfall.

Trap Design Effect on the Capture of Carpophilus spp. (Coleoptera: Nitidulidae) Using Synthetic Aggregation Pheromones and a Coattractant

In a series of experiments conducted in stone fruit orchards in southern Australia, water-based funnel-type traps baited with synthetic aggregation pheromone and fermenting bread dough, trapped 3- to 7-fold as many Carpophihus beetles (primarily C. dauidsoni) than wind-oriented pipe traps or dry funnel traps. The efficacy of dry funnel traps but not pipe traps, appeared to be improved by using water-filled collecting bottles. The potential for using water-based funnel traps in population suppression of Carpophilus spp. in stone fruit orchards through mass trapping is discussed.