Assessing the integrated pest management practices of Pennsylvania, U.S., landscape companies

Data Analysis

Cluster analysis (SPSS Inc., Chicago, IL) was used to determine whether meaningful IPM practitioner segmentations could be created, based on participants’ answers to several questions. Cluster analysis has been used by researchers to define consumer segments related to their preference for horticultural products such as edible flowers (Kelley et al. 2001) and geraniums (Behe et al. 1999), perceived consumer plant knowledge (Hardy et al. 1999), and perceived consumer knowledge about Plant Health Care and IPM practices (Sellmer et al. 2003). Variables used for clustering were based on survey respondents’ knowledge about IPM and cultural practices. Cluster group responses were then analyzed to determine the participants’ level of knowledge of and use of IPM techniques and practices.

It is our goal that this information be available to researchers and extension personnel to help focus their educational efforts on topics of which landscapers lack adequate knowledge. By using K-means, clusters of sizes 2, 3, and 4 were examined using eight cluster algorithms. After examining each cluster size, the three-cluster solution was selected to develop practitioner segments.

General

Of the 1,800 landscape firms who received a survey, 185 completed them, resulting in a response rate of 10%, slightly below the benchmark response rate of 11% for direct-mail surveys (Reed 1999). The most likely reason for the low response rate was due in part to mailings of the first survey and subsequent survey communications reaching companies during their busy spring and early summer season.

Among respondents, 74% reported attaining a degree beyond high school. Over half (52%) were responsible for pest monitoring activities for their business. Forty-six percent reported having between 11 and 20 years of industry experience. Seventy-seven percent of participants were owners or partners in the business. A majority of respondents (87%) were involved in pest management decisions for the business. Twenty-five percent reported gross sales for 1998 of over $500,000, and 63% of the operations have been in business over 10 years.

The top ten most recognized IPM practices among respondents were selecting resistant plant varieties (87%); monitoring (83%); removing heavily infested landscape plants (82%); removing plant debris (81%); matching plants to site conditions (80%); using insecticidal soaps, horticultural oils, or repellents (77%); releasing beneficial insects, mites, or other organisms (74%); pruning (74%); keeping records (72%); and mulching (68%). Less than half of respondents routinely employed IPM (45%), while 40% employed IPM strategies some of the time. In comparison, only 28% of landscape maintenance and lawn care firms surveyed in Atlanta, Georgia, U.S. (Hubbell et al. 2001) were employing IPM practices. Although a higher percentage of Pennsylvania respondents reported that they routinely employ IPM strategies than in Atlanta, there still appear to be opportunities for extension education to increase the use of monitoring, pest identification, site analysis, cultural practices, and record keeping. Our results suggest that the targets of education in these areas are those companies who responded that they “sometimes” (40%), “rarely” (11%), or “never” (4%) employ IPM strategies.

Site analyses, monitoring, and record keeping are three key concepts of landscape IPM practices (Raupp 1985; Rhoads 1985; Nielsen 1990). Of these three concepts, respondents were aware of the importance of monitoring; over 96% monitor for insects and mites, diseases, and weeds combined on a weekly basis during the growing season.

Inconsistencies in recognition and implementation of IPM practices were found among respondents. For example, 80% recognized the importance of matching plants to the intended site; however, only 34% reported conducting site analysis on these sites. Similarly, 72% recognized the importance of keeping pest records, although only 20% did so. If site analyses and record keeping are not implemented, significant limitations confront the landscape manager in their ability to (1) properly select and site plants to ensure survival and success (Hartman et al. 2000; Lilly 2001); (2) identify key pests, key plants, and potential problem areas for developing management plans (Nielsen 1990); and (3) retain and interpret valuable monitoring information when making management decisions. Not taking these steps can result in the unnecessary expenditure of valuable financial and human resources each time a problem reappears at a site where previous documentation of the problem could have expedited diagnosis and treatment. These survey results suggest that greater educational programming may be necessary to demonstrate the practical, cultural, and financial importance of site analyses and record keeping.

Of respondents who monitor for specific pests, a majority monitor for insects and mites during the growing season (82%), followed by weeds (78%) and diseases (73%). Weekly monitoring was most common for each pest (37% insect and mites, 33% weeds, and 28% diseases) followed by daily monitoring (28%, 23%, and 23%, respectively). Weekly monitoring of insects and mites, weeds, and diseases meets the recommendations of Raupp and Cornell (1988) for monitoring frequency in the U.S. mid-Atlantic region. Few respondents (7%) employ growing degree days (GDD) as a tool for initiating insect and mite monitoring activities, even though resources are available for horticulturists interested in employing GDD (Orton 1989). In southeastern Pennsylvania, a cooperative group of growers, landscapers, arboreta and botanic garden personnel, and extension educators have initiated the Penn-Del IPM Research Group, which cooperatively compiles and publishes GDD weekly during the growing season (Hoover 2002). The survey results, however, suggest that efforts to educate landscapers about the use of GDD and plant phenology can and should be expanded and reinforced within Pennsylvania.

Less than 20% of respondents base their insect and mite, disease, and weed monitoring on reports or notification from the PDA plant inspectors (13%, 11%, and 6%, respectively), Penn State Extension educators (17%, 12%, and 7%, respectively), or colleagues (16%, 15%, and 8%, respectively) on pest problems in their area. Fewer utilize the scouting reports that are available by telephone for insect (5%), disease (5%), or weeds (1%) for initiating monitoring activities. Although few scouting reports are available for weeds, scouting reports on arthropods and some diseases are available through cooperative IPM sources in Pennsylvania (for example, sepaipm.cas.psu.edu and paipm.cas.psu. edu/ipminfo.html).

Confidence in accurately identifying insects and mites, diseases, and weeds was relatively low, with respondents reporting greatest confidence in “always” accurately identifying weeds (39%), followed by insects and mites (29%) and diseases (15%). Greater than half (51% to 60%) “usually” accurately identified the insect or mite, weed, or disease problem. Confidence in “always” accurately identifying growing-condition problems on the landscape site was also only 25%, with 60% “usually” identifying the growing-condition problem. These results suggest that additional educational programming in insect and mite, disease, weed, and cultural or environmental problems should be developed and implemented to increase IPM practitioner confidence and accuracy.

Visual inspection, traps (i.e., sticky cards, pheromones, and black-light traps), foliage samples, and indicator plants are some of the techniques available to monitor insect and mite pests in the landscape (Raupp 1985). Among these techniques, a majority of the respondents (96%) employ visual inspection followed by foliage samples (42%) and indicator plants (41%). Few (17%) use traps in their monitoring regime. The survey did not ask why traps are not being used, but this lack of use may be partly due to a lack of understanding or information about the benefits or methods of using traps.

Less than one-third of participants reported sending disease and insect samples to a diagnostic lab for identification (18% insect and mite, 28% disease, respectively). Less than half of the respondents related insect and mite (38%) or disease (42%) presence to current and forecasted environmental conditions. Greater than half of respondents related arthropod (57%) and disease (75%) presence to the plant’s condition. In addition, 70% related plant variety to disease presence. Thirty-two percent of respondents reported identifying the presence of beneficial insect or mites during their monitoring activities.

Incidence, severity, and time of year were predominantly used in deciding whether to treat for arthropods and diseases, with severity being most common at 89% and 87%, respectively. Less than 30% considered phenological indicators prior to treating for arthropods and diseases. Preventive blanket sprays for insects (25%), cost of treatment for diseases (28%), and customer expectation for disease control (28%) were not important factors in deciding to treat.

Respondents most often considered weed severity in their decision to treat (78%). A majority of the respondents targeted persistent weed problems in their management activities (78%), while 61% reported eradicating perennial weeds before planting. Timing of herbicide applications to ensure effective control is a constant issue in landscape management. One method is to target the vulnerable point in the life cycle of a specific weed; however, only 45% reported using this technique for weed control. This finding may be partly due to lack of information on the effective time to target specific weeds. A common approach to reducing weed pressure is to prevent germination of weed seeds; 44% of the respondents reported using geotextile fabrics or plastic weed barriers, while 63% employed pre-emergence herbicides to achieve this goal.

Among practices “often” used by participants to reduce plant related health issues, matching plants to site conditions was most common (91%). Weed maintenance was a common activity (91%), followed by fertilization programs (81%). Fewer than one-quarter (21%) of the landscape managers employed drip irrigation at their sites; however, this percentage may increase after the recent multi-year drought in Pennsylvania and the northeastern U.S. Only 28% reported “often” employing soil tests. When asked how often they conduct soil tests, only 14% tested once a year, while 38% reported testing less than every other year. Similarly, a majority of the respondents reported “never” (81%) using foliar nutrient tests to verify plant nutrient levels. This high rate of landscapers who do not employ foliar nutrient analysis may be due to the lack of knowledge or experience in using this diagnostic tool, or it may be that plant nutrition has not been considered as an issue in overall plant health.

Among current pest and disease management practices, few (9%) claimed to use beneficial organisms, while over half of respondents claimed to conserve beneficials through their selection of pesticides (51%) and use of cultural practices (54%). For insect and mite control, more respondents reported using horticultural oils (46%) compared to insecticidal soaps (27%). Few employed natural products (19%), while 63% reported to use chemical pesticides.

IPM Practitioner Segments

Cluster analysis was used to determine whether meaningful IPM practitioner segments could be created, based on participants’ answers to several key questions regarding implementation of IPM strategies and practices including monitoring, record keeping, site analysis, and regularity of pesticide applications. Three distinct segments were identified: “IPM savvy,” “part-time IPMer,” and “reluctant IPMer.” Each segment clearly defined what practices are currently used to monitor and control insects/mites, diseases, and weeds, along with cultural practices and site factors.

Trends emong Segments

The IPM segments revealed important differences in understanding and implementation among the three groups. Segment comparisons also identified similarities among the IPM groups.

A majority claimed to use IPM strategies to some degree from routinely to rarely (96%). Few respondents relied on outside information from extension, state governmental agencies, or general scouting reports when deciding when and what to monitor. Visual inspection for insects and mites was the monitoring technique of choice for most (91% to 100%) respondents. Traps were employed by fewer than 20% for insect and mite monitoring. Lack of documented use of traps provides an opportunity to develop educational programs centered on their effectiveness, deployment, maintenance, and interpretation of results based on the target insect.

Record keeping was employed by fewer than 22% across the segments, and fewer than half (24% to 44%) conducted complete site analyses for each site. Severity of insect and mite and disease attacks was the predominant consideration for treating (81% to 92%, and 78% to 93%, respectively), while fewer than 30% employed indicators such as growing degree days and phenology of nearby flowering plants.

Matching plants to site conditions, using resistant varieties, selling hardy plants, and offering fertilization and weeding programs were common current practices among respondents in each group, while offering drip irrigation and conducting soil tests were less common (14% to 30% and 20% to 40%, respectively). Among insect and mite control practices, fewer than 12% employed beneficial organisms, while between 41% and 67% employed insecticidal soaps and/or horticultural oils, with 34% to 59% of the respondents in each practitioner segment employing horticultural oils. This finding suggests that information and recommendations by extension educators on the use of insecticidal soaps and horticultural oils are being received and employed by the practitioners. However, the efficacy, availability, and protocols for employing beneficial insects in the landscape may not be as readily available.

Finally, a majority in all three segments have “never” employed foliage nutrient analysis in their diagnoses of plant problems (73% to 90%). This finding suggests that further education on the use, interpretation, and benefits of employing foliar nutrient analyses may be important for adoption of this diagnostic tool.

IPM Practices Relative to Participant Demographics

One question often asked about survey data is whether demographic differences such as education, business size based on gross sales, or years in business impact the implementation of IPM strategies. To address this question, the survey data were analyzed based on the respondents’ education level, gross sales for 1998, and years in business. Education levels were divided into two groups for analysis. One group had formal education up to and including an associate’s degree or trade school, and the other group had obtained a bachelor’s degree or higher. Businesses with gross sales of up to US$100,000 were grouped separately from those generating greater than US$100,001 during 1998. Finally, companies that have been in business up to 14 years comprised one group, while the other group had been in business for 15 or more years. Respondents were evenly split based on education, with 49% having an associate’s degree or less in formal education, while 51% had obtained a bachelor’s degree or higher. Similarly, the businesses were nearly evenly split based on years in business, with 47% in business for less than or equal to 14 years and 53% in business for 15 or more years. In contrast, a majority of the businesses reported gross sales for 1998 of greater than US$100,001, compared to 35% grossing up to US$100,000.

A majority of the respondents represented in these demographic groups reported using IPM strategies to some level, from routinely to rarely in their business (94% to 98%). Overall, few significant demographic differences were found with respect to monitoring frequency or techniques, accuracy in identifying pests or cultural problems, record keeping, treatment decision-making practices, cultural management practices, or use of pesticides.

Among educational groups, significant differences were found in two areas: (1) criteria used in making treatment decisions for arthropod pests and (2) use of cultural practices to reduce weeds. Forty-seven percent of respondents with a bachelor’s degree or higher considered current and forecasted environmental conditions when deciding to treat for insects and mites, while only 28% of the respondents with an associate’s degree or less employed this practice in their decision-making process. Similarly, 81% of respondents with higher post-secondary education reported implementing cultural practices to reduce weeds, while 66% of those with an associate’s degree or less employed cultural practices.

Businesses grossing greater than US$100,001 in sales were significantly more likely to monitor for insects or mites (78%), diseases (74%), and weeds (73%) during the growing season. Of these firms, 84% claimed to monitor often for pests and diseases. These businesses were significantly more likely to have an employee (11%) or a combination of individuals (47%) monitoring for pests, while significantly fewer (38%) of survey participants monitored for pests at the business. Severity of insect and mite infestations was the most common trigger (93%) for treatment decisions with firms grossing over US$100,001, although severity was also important among firms grossing less than US$100,000. Soil testing was used by about 33% of these firms. Fifty-two percent of these firms employed horticultural oils and 69% employed chemical pesticides compared to less than half of the companies grossing under US$100,000 (32% and 48%, respectively). The companies grossing over US$100,001 reported using pre-emergence herbicides significantly more (68%) often than the companies grossing US$100,000 or less (48%).

Companies that were in business for 15 years or more were significantly more likely (74%) to consider time of year when deciding to treat for insects or mites compared to younger businesses (58%). A hierarchy of importance was evident in both younger and older firms, with severity being the issue of greatest consideration followed by incidence and time of year, phenological indicators, and use of blanket sprays. A similar trend was apparent in the general survey data. Significantly more (39%) of the older companies (≥ 15 years in business) conducted soil tests compared to 17% of the younger companies. Older companies reported employing horticultural oils and/or insecticidal soaps (66%) in the treatment of insects or mites, with 57% using only horticultural oils. Older companies were likely to “often” apply fungicides on a schedule (17%) and “always” apply herbicides (10%) on a schedule compared to younger companies (6% and 3%, respectively).