Bug Counts Research Articles

Software defect prediction using learning to rank approach

Software defect prediction (SDP) plays a significant role in detecting the most likely defective software modules and optimizing the allocation of testing resources. In practice, though, project managers must not only identify defective modules, but also rank them in a specific order to optimize the resource allocation and minimize testing costs, especially for projects with limited budgets. This vital task can be accomplished using Learning to Rank (LTR) algorithm. This algorithm is a type of machine learning methodology that pursues two important tasks: prediction and learning. Although this algorithm is commonly used in information retrieval, it also presents high efficiency for other problems, like SDP. The LTR approach is mainly used in defect prediction to predict and rank the most likely buggy modules based on their bug count or bug density. This research paper conducts a comprehensive comparison study on the behavior of eight selected LTR models using two target variables: bug count and bug density. It also studies the effect of using imbalance learning and feature selection on the employed LTR models. The models are empirically evaluated using Fault Percentile Average. Our results show that using bug count as ranking criteria produces higher scores and more stable results across multiple experiment settings. Moreover, using imbalance learning has a positive impact for bug density, but on the other hand it leads to a negative impact for bug count. Lastly, using the feature selection does not show significant improvement for bug density, while there is no impact when bug count is used. Therefore, we conclude that using feature selection and imbalance learning with LTR does not come up with superior or significant results.

The Four Horsemen of an Ailing Software Project

KV has talked about various measures of software quality in past columns, but perhaps falling software quality is one of the most objective measures that a team is failing. This Pestilence, brought about by the low morale engendered in the team by War and Famine, is a clear sign that something is wrong. In the real world, a diseased animal can be culled so that disease does not spread and become a pestilence over the land. Increasing bug counts, especially in the absence of increased functionality, is a sure sign of a coming project apocalypse.

Comparative Efficacy of Superheated Dry Steam Application and Insecticide Spray Against Common Bed Bugs Under Simulated Field Conditions.

The common bed bug, Cimex lectularius L., is a difficult urban pest to control. A simulated field study was conducted to compare the efficacy of steam application and an insecticide mixture spray (0.05% acetamiprid and 0.06% bifenthrin mixture) against C. lectularius. Three types of furniture (desk chair, upholstered armchair, and wooden table) were treated in the laboratory. The efficacy of the treatments was evaluated by visual inspection and placement of interceptor traps under the legs of the furniture. One hundred mixed stages of an insecticide-resistant population of C. lectularius were released onto each furniture item. After a 10-day acclimation period, each furniture item received steam treatment, insecticide spray, or no treatment. The second application of treatment was conducted 14 d later. Bed bug counts from interceptors and visual inspections were recorded at 13 d and 28 d after the initial treatment. At 28 d, the mean (± SE) live bed bug count in the steam, spray, and control group was 1 ± 0, 2 ± 1, and 83 ± 10, respectively. Both treatment methods were highly effective in controlling bed bugs on furniture. The mean bed bug count from interceptors in the steam, spray, and control groups were 0.3 ± 0.2, 11 ± 7, and 47 ± 9, respectively. There was no significant difference in the efficacy between steam and spray treatments based on either visual inspection or bed bug counts from interceptors. However, based on interceptor counts, the steam treatment caused faster bed bug population reduction than insecticide sprays.

Efficacy of Lemon Extracts in Controlling Mealy Bugs of Stored Yam (Dioscorea rotundata)

Background: The use of plant extracts in controlling mealy bugs is widespread especially on stored products. Despite this, research on lemon extracts, which is known to be very effective against some insect pests are limited. We assessed the repellent ability of lemon extracts in controlling mealy bug infestation and observed if there was any added benefit of the extract in retarding sprouting of yams in storage. Methods: We used the complete randomized design with three treatments and three replications. The treatments consisted of lemon peels, lemon leaf extracts and a control. Ten tubers of Pona white yam variety were rubbed on the periderm with each treatment and placed on a shelf. Two yam tubers naturally infested with mealy bugs were placed at the edges of each replication to serve as sources of infestation. Result: Results showed that yam tubers treated with lemon peel extracts recorded relatively low mealy bug counts compared to untreated ones and, tubers treated with lemon extracts recorded relatively higher weights compared to those without any treatment. Furthermore, there were no significant differences in neither counted mealy bugs nor yam weight per treatment between extracts from leaves or peels. Lemon extract treatments did not affect sprouting and shelf life of tubers in storage. We conclude that lemon extracts are effective repellents of mealy bugs and can be promoted as a safe bio pesticide for stored yam against mealy bug infestation.

IMPLEMENTATION OF NOVEL APPROACHES IN BIDIRECTIONAL MODEL TRANSFORMATION: A SYSTEMATIC LITERATURE REVIEW

In the past few years, software development has seen rapid growth, and developers have adopted different methods to provide efficient procedures in software development, thus reducing the overall bug counts and time delay. Bidirectional model transformation is one such technique that encompasses the development of the object code in both directions enabling an abstract view of the software to the developer; over the year’s researchers, have been able to produce many approaches in bidirectional model transformations (bx), but the cost and best fir for effective model transformation, in particular, a quantities survey will be designed which will discuss the best possible apron in the bx. The methodology for this survey shall be made through SLR to identify around 20 different approaches proposed for bidirectional model transformation; these studies range from the year 2010 till date and are thus, rendered latest in the respective fields of our survey. The gathered results have been calculated on the specific set of parameters that are cost and time of usage time are the main aspects of these approaches, and that is the predicament that has made us produce a systematic literature review (SLR) on this very topic. Thus, this paper investigates different approaches based on their implementation cost and time delay and discusses their limitations, and the approach is implemented. Those approaches have been selected, which culminate in both of these respective parameters. The main objective of this SLR is to provide an insight into the different approaches and establish a well-balanced approach that can be used in bidirectional model transformation in software development.

Evaluation of a Non-Chemical Compared to a Non-Chemical Plus Silica Gel Approach to Bed Bug Management.

Bed bug resistance to commonly used pesticide sprays has led to exploring new pesticides and other strategies for bed bug management. Non-chemical methods are effective in bed bug management; however, they do not provide residual protection. Compared to insecticide sprays, dust formulations are considered to provide longer residual control. We evaluated two bed bug management programs in apartment buildings. A building-wide inspection was initially conducted to identify bed bug infested apartments. Selected apartments were divided into two treatment groups: non-chemical plus silica gel dust treatment (10 apartments) and non-chemical treatment (11 apartments). After initial treatment, apartments were re-visited monthly for up to 6 months. During each visit, the total bed bug count per apartment was obtained by examining interceptor traps placed in the apartments and conducting a visual inspection. Mean bed bug count was reduced by 99% and 89% in non-chemical plus silica gel dust and non-chemical treatment, respectively. Non-chemical plus silica gel dust treatment caused significantly higher bed bug count reduction than the non-chemical treatment at 6 months. Bed bugs were eradicated from 40% and 36% of apartments treated with non-chemical plus silica gel dust treatment and non-chemical treatment, respectively.

BCV-Predictor: A bug count vector predictor of a successive version of the software system

Predicting the number of bugs in a software system intensifies the software quality and turns down the testing effort required in software development. It reduces the overall cost of software development. The evolution of hardware, platform, and user requirements leads to develop the next version of a software system. In this article, we formulate a problem and its novel solution, i.e., we are considering the prediction of the bug count vector of a successive version of a software system. After predicting the bug count vector in the next version of a software, the developer team leader can adequately allocate the developers in respective fault dense modules, in a more faulty dense module, more number of developers required. We have conducted our experiment over seven PROMISE repository datasets of different versions. We build metadata using a concatenation of different versions of the same software system for conducting experiments. We proposed a novel architecture using deep learning called BCV-Predictor. BCV-Predictor predicts the bug count vector of the next version software system; it is trained using metadata. To the best of our knowledge, no such work has been done in these aspects. We also address overfitting and class imbalance problem using random oversampling method and dropout regularization techniques. We conclude that BCV-Predictor is conducive to predicting the bug count vector of the next version of the software. We found five out of seven meta datasets reaches to more than 80% accuracy. In all seven meta datasets, Mean Squared Error (MSE) lies from 0.71 to 4.715, Mean Absolute Error (MAE) lies from 0.22 to 1.679, MSE and MAE over validation set lie between 0.84 to 4.865, and 0.22 to 1.709 respectively. We also compared the performance of BCV-Predictor with eleven baselines techniques and found the proposed approach outperform on most of the meta-datasets.

Reliability modeling of multi-version software system incorporating the impact of infected patching

PurposeThe purpose of the present work is to mathematically model the reliability growth of a multi-version software system that is affected by infected patches.Design/methodology/approachThe work presents a mathematical model that studies the reliability change due to the insertion of an infected patch in multi-version software. Various distribution functions have been considered to highlight the varied aspects of the model. Furthermore, weighted criteria approach has been discussed to facilitate the choice of the model.FindingsThe model presented here is able to quantify the effect of an infected patch on multi-version software. The model captures the hike in bug content due to an infected patch.Originality/valueMulti-version systems have been studied widely, but the role of an infected patch has not been yet explored. The effect of an infected patch has been quantified by modeling the extra bugs generated in the system. This bug count would prove helpful in further studies for optimal resource allocation and testing effort allocation.

Green Lacewings and Water Sprays for Azalea Lace Bug Control1

Abstract The azalea lace bug, Stephanitis pyrioides, is a serious pest of azaleas and rhododendrons which is often controlled by systemic insecticides. However, the efficacy of softer approaches such as biological control and water sprays against this pest on rhododendrons is unknown. Therefore, we tested the commercially available green lacewing predator, Chrysoperla rufilabris, and water sprays on lace bug infestation in one laboratory and four field trials. First, 2nd instar predator larvae were confirmed to consume lace bug nymphs and sometimes adults. Second, tapping predator larvae from hexcel units over dry leaves of potted rhododendrons and shaking loose eggs over wet leaves were reliable application methods. Third, predator larvae released onto potted rhododendrons lowered lace bug counts for two weeks. Fourth, after four bi-weekly applications, plants receiving egg cards or water-sprays had reduced lace bug counts and fewer damaged leaves than control plants. Fifth, landscape plants receiving the sequential combination of water spray followed by predator egg releases had 44 to 90% lower lace bug abundance and fewer damaged leaves than the control. After lace bugs were initially dislodged, hatching predators might have consumed hatching lace bugs. Index words: biological control, Chrysoperla rufilabris (Burmeister), mechanical control, rhododendron, Stephanitis pyrioides (Scott) Species used in this study: Azalea lace bug (Stephanitis pyrioides Scott), green lacewing (Chrysoperla rufilabris Burmeister), Rhododendron (Rhododendron spp.)

Testing a Threshold-Based Bed Bug Management Approach in Apartment Buildings.

We tested a threshold-based bed bug (Cimex lectularius L.) management approach with the goal of achieving elimination with minimal or no insecticide application. Thirty-two bed bug infested apartments were identified. These apartments were divided into four treatment groups based on apartment size and initial bed bug count, obtained through a combination of visual inspection and bed bug monitors: I- Non-chemical only in apartments with 1–12 bed bug count, II- Chemical control only in apartments with 1–12 bed bug count, III- Non-chemical and chemical control in apartments with >12 bed bug count, and IV- Chemical control only in apartments with ≥11 bed bug count. All apartments were monitored or treated once every two weeks for a maximum of 28 wk. Treatment I eliminated bed bugs in a similar amount of time to treatment II. Time to eliminate bed bugs was similar between treatment III and IV but required significantly less insecticide spray in treatment III than that in treatment IV. A threshold-based management approach (non-chemical only or non-chemical and chemical) can eliminate bed bugs in a similar amount of time, using little to no pesticide compared to a chemical only approach.

Behavioral Response of the Brown Marmorated Stink Bug (Hemiptera: Pentatomidae) to Semiochemicals Deployed Inside and Outside Anthropogenic Structures During the Overwintering Period.

The brown marmorated stink bug, Halyomorpha halys (Stål), is an invasive species from Asia capable of causing severe agricultural damage. It can also be a nuisance pest when it enters and exits anthropogenic overwintering sites. In recent years, pheromone lures and traps for H. halys have been developed and used to monitor populations in field studies. To date, no study has investigated the applicability of these monitoring tools for use indoors by building residents during the overwintering period. Herein, we 1) assessed when in late winter (diapause) and spring (postdiapause) H. halys begins to respond to its pheromone (10,11-epoxy-1-bisabolen-3-ol), 2) evaluated whether pheromone-based tools can be used reliably for monitoring H. halys adults in unheated and heated buildings, and 3) elucidated the potential for indoor management using pheromone-baited traps. A 2-yr trapping study suggested that H. halys began to respond reliably to pheromone-baited traps after a critical photoperiod of 13.5 h in the spring. Captures before that point were not correlated with visual counts of bugs in buildings despite robust populations, suggesting currently available pheromone-baited traps were ineffective for surveillance of diapausing H. halys. Finally, because baited traps captured only 8-20% of the adult H. halys known to be present per location, they were not an effective indoor management tool for overwintering H. halys. Our study contributes important knowledge about the capacity of H. halys to perceive its pheromone during overwintering, and the ramifications thereof for building residents with nuisance problems.

Field Study of the Comparative Efficacy of Three Pyrethroid/Neonicotinoid Mixture Products for the Control of the Common Bed Bug, Cimex lectularius.

Three insecticide mixtures that contain two classes of insecticides (pyrethroid and neonicotinoid) were recently developed to control bed bugs. We evaluated three integrated bed bug management strategies in apartments, each using the same non-chemical control methods and one of the three insecticide mixture products: Tandem (lambda-cyhalothrin + thiamethoxam), Temprid SC (beta-cyfluthrin + imidacloprid), and Transport Mikron (bifenthrin + acetamiprid). No insecticides were applied in the Control apartments. In all apartments, we installed vinyl mattress encasements (if not already present) and applied steam to beds and other infested upholstered furniture. Insecticide sprays were applied in the three treatments. Each treatment and the Control included 8–10 occupied apartments. Re-treatment was conducted during biweekly inspections if necessary. After eight weeks, the mean (± SEM) bed bug count reduction in the Tandem, Temprid SC, Transport Mikron, and Control was 89 ± 9, 87 ± 6, 98 ± 1, and 23 ± 54%, respectively. Only Tandem and Transport Mikron treatments resulted in significantly higher population reduction than the Control at eight weeks. There were no significant differences in mean percent reduction among the three treatments (Tandem, Temprid SC, Transport Mikron) at eight weeks. Tandem spray caused significantly faster bed bug reduction than Temprid SC spray and Transport Mikron spray.

Evaluation of a model community-wide bed bug management program in affordable housing.

Low-income apartment communities in the United States are suffering from disproportionally high bed bug, Cimex lectularius L., infestations owing to lack of effective monitoring and treatment. Studies examining the effectiveness of integrated pest management (IPM) for the control of bed bugs in affordable housing have been limited to small subsets of bed-bug-infested apartments, rather than at the apartment community level. We developed, implemented and evaluated a complex-wide IPM program for bed bugs in an affordable housing community. Proactive inspections and biweekly treatments using a combination of non-chemical and chemical methods until bed bugs were not detected for three biweekly monitoring visits were key elements of the IPM program. A total of 55 bed-bug-infested apartments were identified during the initial inspection. Property management was unaware of 71% of these infestations. Over the next 12 months, 14 additional infested apartments were identified. The IPM program resulted in a 98% reduction in bed bug counts among treated apartments and reduced infestation rates from 15 to 2.2% after 12 months. Adopting a complex-wide bed bug IPM program, incorporating proactive monitoring, and biweekly treatments of infested apartments utilizing non-chemical and chemical methods can successfully reduce infestation rates to very low levels.

Efficacy of an Essential Oil-Based Pesticide for Controlling Bed Bug (Cimex lectularius) Infestations in Apartment Buildings.

Bed bugs (Cimex lectularius L. and Cimex hemipterus F.) are among the most difficult urban pests to manage. Many essential oil-based bed bug control products that are considered reduced risk to mammals compared to synthetic insect neurotoxins have become commercially available, but their effectiveness as a stand-alone control method is unknown. This study assessed the field efficacy of an essential oil-based bed bug control product (EcoRaider; a.i. 1% geraniol + 1% cedar oil + 2% sodium lauryl sulfate) compared to a pyrethroid and neonicotinoid mixture spray (0.075% Temprid SC; a.i. beta-cyfluthrin + imidacloprid). After 12 weeks, the three treatments—EcoRaider, Temprid SC, and EcoRaider + Temprid SC caused 92.5 ± 2.7, 92.9 ± 3.0, and 91.7% ± 2.7% bed bug count reduction, respectively. No significant differences existed in the bed bug reduction among the treatments. Bed bugs were eliminated from only 22% of the treated apartments. Among those still with bed bugs, 76% of the residents did not know bed bugs were present. We documented the residents’ self-control practices and discussed the potential of using essential oil-based insecticides in bed bug management programs to minimize the health risks to building occupants and pets and to slow down the development of insecticide resistance.

Spatial distribution of stink bugs (hemiptera: pentatomidae) in wheat.

A two-year study was conducted in South Carolina wheat (Triticum aestivum L. (Poales: Poaceae)) fields to describe spatial and temporal dynamics of stink bugs (Hemiptera: Pentatomidae), which were sampled weekly with sweep nets. In 2010, the main phytophagous stink bugs caught in a grid sampling plan across two fields were the brown stink bug, Euschistus servus (Say), the rice stink bug, Oebalus pugnax (F.), the southern green stink bug, Nezara viridula (L.), and the red shouldered stink bug, Thyanta custator (F.), for both adults and nymphs. In 2011, the main phytophagous stink bugs were E. servus, O. pugnax, N. viridula, and T. custator across two fields. Adult stink bug counts adjacent to fallow fields were 2.1-fold greater for all species combined compared with counts adjacent to woods. Spatial Analysis by Distance IndicEs (SADIE) indicated significant aggregation for 35% of analyses for adults and nymph stink bugs at each sampling date. As a measure of spatial and temporal stability, positive SADIE association indices among sampling dates recorded 11, 36, 43, and 16% of analyses for adult E. servus and 7, 50, 50, and 14% for adult O. pugnax in fields A, B, C, and D, respectively. Adult and nymph stink bugs were spatially associated within wheat fields based on SADIE association indices. Seasonal counts of stink bugs were spatially associated with spike counts at least once for each species across the four fields. Future work may investigate practices to reduce stink bug buildup on wheat in the spring and movement to susceptible crops such as corn, Zea mays L.

Life Cycle Traits ofLethocerus deyrollei(Hemiptera: Belostomatidae), in Central Japan: Possibility of Inoculation in Extinct Areas

The life cycle of Lethocerus deyrollei (Vuillefroy) in Sayo-cho, Hyogo, Japan (35 ° 00' N, 134 ° 21' E), was studied in terms of flight, reproduction, and energy allocation as well as developmental zero (t0) and required heat units (K) for a generation. Counts of bugs attracted to electric lighting showed that L. deyrollei flew most frequently in July, the second half of the mating season. The weight of flight muscle showed no significant yearly change despite a conspicuous difference in the flight activities. Fat content in the abdomen was low during summer, when oviposition activity reached a high level in females. Thereafter, lipid accumulation increased to mid-October when flight to an overwintering site took place. The lipid reservoir in late summer may be critical for surviving winter but no significant decrease was observed during overwintering. The number of eggs in the ovary indicated univoltism because no females had mature eggs late in the year. t0 and K were calculated based on laboratory rearing at different temperatures. Thus, estimated K (635 degree days) and local heat mass suggest a univoltine life cycle in Sayo-cho. The possible areas for successful inoculation of L. deyrollei were examined on the basis of available heat mass and day length in places where the species is extinct or not observed. Based on the estimated K and critical day length for reproductive maturation, Niigata and Tokyo fulfill suitable conditions of the Sayo population to complete a univoltine life cycle.

Evaluation of an Insecticide Dust Band Treatment Method for Controlling Bed Bugs

Current bed bug, Cimex lectularius L., control usually involves insecticide applications that pose a high risk of insecticide exposure to residents and applicators. To minimize these risks and the amount of insecticides used, we designed and evaluated a dust band treatment technique. The laboratory assay showed that 1% cyfluthrin dust treated bands are highly effective in killing bed bugs. We further evaluated this technique in bed bug infested apartments. The "dust band" treatment consisted of installing a 3.8-cm-wide fabric band on furniture legs and brushing Tempo dust (1% cyfluthrin) (Bayer Environmental Science, Research Triangle Park, NC) onto the bands. In addition, interceptors were installed under furniture legs. Alpine (0.5% dinotefuran) aerosol spray was applied directly to live bed bugs found on furniture during biweekly inspections. This treatment was compared with two other treatments: "integrated pest management" (IPM) and "control." The IPM treatment included dust bands plus the following: applying hot steam to infested furniture and surrounding areas, installing mattress encasements, applying 1% cyfluthrin dust around room perimeters, and installing interceptors under furniture legs. Alpine aerosol was applied to live bed bugs found during biweekly inspections. In the control group, the apartments received cursory treatment with insecticide sprays by the existing pest control contractor hired by the property management office. Bed bug numbers before and after treatments were determined based on biweekly interceptor counts or a combination of interceptor counts and visual inspections. From 0 to 12 wk, mean bed bug counts of the dust band, IPM, and the control treatment decreased by 95, 92, and 85%, respectively. Both dust band and IPM resulted in higher bed bug reduction than the control. There was no significant difference in the final counts between dust band and IPM treatments. An additional field experiment showed installing 1% cyfluthrin dust band and interceptors in lightly infested apartments prevented bed bug population rebound. Results indicate applying insecticide dust bands to furniture legs is an effective bed bug control technique.

Comparison of Three Bed Bug Management Strategies in a Low-Income Apartment Building.

Bed bug (Cimex lectularius L.) infestations are currently controlled by a variety of non-chemical and chemical methods. There have been few studies on the comparative effectiveness of these control techniques. We evaluated three bed bug management strategies in an apartment building: (1) non-chemical methods only (n = 9); (2) insecticides only (n = 6); and (3) integrated pest management including both non-chemical methods and insecticides (n = 9). The apartments were one-bedroom units occupied by seniors or people with disabilities. Bed bug numbers in each apartment were determined by visual inspection and/or installing intercepting devices under bed and sofa legs. The median (min, max) bed bug counts in the non-chemical methods only, insecticides only, and integrated pest management (IPM) treatment were: 4 (1, 57), 19 (1, 250), and 14 (1, 219), respectively prior to the treatments. The apartments were retreated if found necessary during biweekly to monthly inspections. After 10 weeks, bed bugs were found to be eliminated from 67, 33, and 44% of the apartments in the three treatment groups, respectively. The final (after 10 weeks) median (min, max) bed bug counts in the non-chemical methods only, insecticides only, and IPM treatment were: 0 (0, 134), 11.5 (0, 58), and 1 (0, 38), respectively. There were no significant differences in the speed of bed bug count reduction or the final bed bug counts. Lack of resident cooperation partially contributed to the failure in eliminating bed bugs from some of the apartments. Results of this study suggest that non-chemical methods can effectively eliminate bed bugs in lightly infested apartments.

Evaluation of Two Least Toxic Integrated Pest Management Programs for Managing Bed Bugs (Heteroptera: Cimicidae) With Discussion of a Bed Bug Intercepting Device

The cost and effectiveness of two bed bug (Cimex lectularius L.) integrated pest management (IPM) programs were evaluated for 10 wk. Sixteen bed bug-infested apartments were chosen from a high-rise low-income apartment building. The apartments were randomly divided into two treatment groups: diatomaceous earth dust-based IPM (D-IPM) and chlorfenapyr spray-based IPM (S-IPM). The initial median (minimum, maximum) bed bug counts (by visual inspection) of the two treatment groups were 73.5 (10, 352) and 77 (18, 3025), respectively. A seminar and an educational brochure were delivered to residents and staff. It was followed by installing encasements on mattresses and box springs and applying hot steam to bed bug-infested areas in all 16 apartments. Diatomaceous earth dust (Mother Earth-D) was applied in the D-IPM group 2 d after steaming. In addition, bed bug-intercepting devices were installed under legs of infested beds or sofas or chairs to intercept bed bugs. The S-IPM group only received 0.5% chlorfenapyr spray (Phantom) after the nonchemical treatments. All apartments were monitored bi-weekly and retreated when necessary. After 10 wk, bed bugs were eradicated from 50% of the apartments in each group. Bed bug count reduction (mean +/- SEM) was 97.6 +/- 1.6 and 89.7 +/- 7.3% in the D-IPM and S-IPM groups, respectively. Mean treatment costs in the 10-wk period were $463 and $482 per apartment in the D-IPM and S-IPM groups, respectively. Bed bug interceptors trapped an average of 219 +/- 135 bed bugs per apartment in 10 wk. The interceptors contributed to the IPM program efficacy and were much more effective than visual inspections in estimating bed bug numbers and determining the existence of bed bug infestations.

Methods of analysis for georeferenced sample counts of tarnished plant bugs in cotton

The problem of analyzing georeferenced cotton pest insect samples when a large percentage of the counts are zero is examined. The use of appropriate statistical methods for their analysis is required. To demonstrate this, georeferenced samples (n = 63) of tarnished plant bugs (TPBs; Lygus lineolaris [Palisot de Beauvois] (Heteroptera: Miridae)) were analyzed by three statistical methods and the results were compared. Correlation analysis of the sample counts with 25 classes of cotton growth derived from an unsupervized classification of multispectral imagery was followed by a complete enumeration analysis comprising three scenarios. The first scenario assumed the insect samples were unstratified. A distribution of sample averages was created by complete enumeration of all combinations of samples taken four at a time. The second scenario used imagery of the cotton fields to allocate the samples among three cotton growth categories (marginal, good or best) derived by a supervized classification of the 25 unsupervized classes. The insect samples associated with these categorical habitats were completely enumerated using allocations of 4, 6, 8 or 10 samples at a time from various sample sizes to determine how different allocations affected the results. The mean was not affected, but the standard deviation decreased with increased allocation sizes in all habitats. The third scenario used the two observers and three habitat categories to create six additional complete enumeration distributions by allocating four samples at a time from groups of varying sample sizes. These enumeration distributions are non-parametric estimators of the sampling distribution of: (1) the sample averages of a given sample size when samples are taken from the entire field, (2) the sample averages of a given size when samples are taken from each cotton habitat or (3) the sample averages of a given sample size from samples taken from each habitat by each observer. To support the enumeration analyses, these insect samples were analyzed further by Poisson regression models. These models showed significant differences between TPB counts by the two observers and among the habitats, whereas the observer by habitat interaction was not significant. For every combination of observer and cotton growth category, a Poisson regression model estimated the mean rate of TPB numbers. These means were similar to the corresponding modes of the complete enumeration distributions. The two non-standard methods showed that TPB numbers differed by habitat categories even though there were samples with a zero count, whereas a correlation analysis failed to identify a relationship between TPB sample counts and unsupervized habitat classes.