There are ~25 million golfers in the United States who play on 16,752 golf courses (Royal Ancient (R&A), 2019; National Golf Federation (NGF), 2021). According to the US Bureau of Labor Statistics, golf is the second most popular participation sport in the US behind basketball (Woods, 2017). However, managing golf course turfgrass often requires repeated use of fungicides o provide acceptable control of damaging plant diseases. Without pesticide applications, golf course greens are often not able to survive in US climates (Rossi and Grant, 2009). Dollar spot (Clarireedia spp.) is the most common golf course turfgrass disease in temperate climates around the world, and on cool-season turfgrass in North America dollar spot is caused by C. jacksonii (Smiley et al., 2005; Salgado-Salazar et al., 2018; Sapkota et al., 2022). Optimal conditions for dollar spot symptom development include temperatures between 15 and 30°C and relative humidity in excess of 85% (Smiley et al., 2005). Left uncontrolled, tan-colored foci 2–5 cm in diameter can interrupt the playability and the aesthetics of the turfgrass playing surfaces (Smiley et al., 2005). The primary fungicides used to control dollar spot in the US include chlorothalonil, fluazinam, and multiple fungicides from the demethylation inhibitor (DMI) and the succinate dehydrogenase inhibitor (SDHI) classes of fungicides (Latin, 2021). In most temperate US climates fungicide applications targeting dollar spot begin in May and end in September or October (Smiley et al., 2005). Controlling dollar spot on golf course greens and fairways is generally in the best economic interest of those in the golf business to retain clientele. Indeed, more money is spent on fungicides to suppress dollar spot than any other turfgrass pest, and a single golf course can easily exceed $10,000 USD annually targeting dollar spot (Vargas, 2005; Golf Course Industry, 2015).

The high cost of fungicides to suppress dollar spot has led to the development of multiple dollar spot forecasting models to help optimize fungicide application timing (Sapkota et al., 2022). Models developed by Mills and Rothwell (1982); Hall (1984); Ryan et al. (2012) have been developed to predict dollar spot symptom development, however they have not been widely implemented by the golf course management industry due to poor performance. More recently, the Smith-Kerns Dollar Spot Model (SKM) was developed that uses a 5-day moving average of average daily air temperature and average daily relative humidity to produce a probability that dollar spot will develop on any given day (Smith et al., 2018). The authors were involved in the development of the SKM and found that a model probability of 20% acted as an acceptable threshold to time preventative fungicide applications for dollar spot control (Smith et al., 2018). The SKM helps reduce the number of overall fungicide applications targeting dollar spot by eliminating applications when there isn't a threat of symptom development, and it has become widely utilized by the golf course industry (Smith et al., 2018; Sapkota et al., 2022).

The repeated use of fungicides to control dollar spot has led to widespread documentation of C. jacksonii populations that are resistant to a variety of fungicide classes including the demethylation inhibitor (DMI), dicarboximide, succinate dehydrogenase inhibitor (SDHI), and benzimidazole fungicide classes (Detweiler et al., 1983; Golembiewski et al., 1995; Burpee, 1997; Sang et al., 2015; Sapkota et al., 2022). In addition, the frequent application of pesticides to prevent turfgrass diseases can also present a variety of possible health risks to golf course workers, the golfing public, and the environment. Previous research has indicated that golf course pesticide use had quantifiable human health effects on golf course superintendents (Kross et al., 1996; Knopper and Lean, 2010), but that the chronic inhalation and dermal risk of pesticides to golfers was relatively low (Murphy and Haith, 2007; Wong and Haith, 2014). Ecologically, Haith and Rossi (2003) found that runoff of chlorothalonil, iprodione, and pentachloronitrobenzene (PCNB) from fairways may routinely affect aquatic health in surrounding water bodies. Metcalfe et al. (2008) found concentrations of PCNB in golf course streams, a byproduct of the fungicide quintozene, at concentrations toxic to developing fish. Multiple studies have detected high levels of arsenic in soils and groundwater of golf courses from the use of arsenic containing pesticides (Cai et al., 2002; Feng et al., 2005; Pichler et al., 2008). In addition, Gunstone et al. (2021) reviewed 400 studies of soil invertebrates and 2,800 tested parameters (the effect of a specific pesticide on a specific soil invertebrate) and found that 70.5% of tested parameters showed negative effects, 1.4% showed positive effects, and 28.1% showed no effect. Soil invertebrates, critical to nutrient cycling, carbon sequestration, and soil morphology are often overlooked in pesticide risk assessment.

Reducing the risk of injury associated with pesticide use on golf courses has the potential to benefit golf course superintendents, local environmental health, and the golfing public. Bekken et al. (2021) found that pesticide risk on golf courses in the northern US was primarily caused by fungicide use, and the majority of golf course fungicide use targets dollar spot (Vargas, 2005). Pesticide applications to fairways (average size of 11.4 ha), though more infrequent than applications to greens (average area of 1.3 ha), had significantly higher total pesticide risk because their larger area increased potential non-target pesticide exposure (Bekken et al., 2021). Thus, reducing the risk of fairway dollar spot applications is one of the most effective means to reduce overall golf course pesticide risk.

The US Environmental Protection Agency (EPA) Risk Assessment Program reviews the ecological and human health risks of all new and existing pesticides before they can be registered or re-registered for use in the US (US EPA, 2021). The goal of the program is to ensure that the use of pesticides poses no “unreasonable risks” to humans, plants, wildlife, and the environment. However, EPA risk assessments only consider individual pesticide products and therefore do not quantify the risk of pest management programs that commonly contain many different pesticides. One method to estimate pesticide risk of pesticide application programs is to sum the annual weight of pesticide active ingredient applied (Kerns and Tredway, 2013). However, such metrics consider only pesticide exposure and fail to consider differences in toxicity between pesticides (Barnard et al., 1997). To account for this, many pesticide risk indicators have been developed to account for both exposure and toxicity for use in agricultural situations (Greitens and Day, 2006; Oliver et al., 2016; Kniss, 2017; Schulz et al., 2021). Pesticide risk indicators have the added benefit of being able to quickly assess risk of a high number of sites regardless of geographic location at a lower cost than field-based sampling and ecological risk assessment methods. Pesticide risk indicators that consider both toxicity and exposure and that are utilized in this study include: Environmental Impact Quotient (EIQ) (Kovach et al., 1992), hazard quotient (HQ) (Kniss, 2017), and the Pesticide Risk Tool (PRT) (IPM Institute of North America, 2021). Indicators of pesticide risk are not widely used in the golf course management industry despite the heavy reliance on chemical fungicides to provide acceptable conditions.

The objective of this study was to test the ability of the SKM and reduced risk pesticide products to both reduce golf course pesticide risk and control dollar spot infection on fairways and greens in comparison to conventional pesticide programs. We define reduced risk pesticide products as those either labeled as reduced risk by the Reduced Risk Program (EPA Reduced Risk Program, 2021) or products with similar chemistries and toxicity profiles as fungicides included in the Reduced Risk Program. In addition, the study benchmarks average pesticide risk for 23 golf courses in Wisconsin, USA, and tests application programs of lower risk for their efficacy in controlling dollar spot.

Both the SKM Reduced Risk and State Risk Comparison studies indicate that pesticide risk, as measured by a wide variety of pesticide risk indicators, associated with dollar spot management of cool season turfgrass can be reduced by more than 50% while maintaining high levels of dollar spot control. Reductions in pesticide risk in both studies were achieved through both changes to product selection and by applying pesticides less frequently, but product selection generally had a greater effect on reducing pesticide risk. Bekken et al. (2021) also found product selection to be an important determinant of pesticide risk on golf courses in the northern US. One of the central goals of the SKM Reduced Risk study was to determine if the utilization of the Smith-Kerns Model (SKM) would lead to a reduction in pesticide risk. Using the SKM model in the CP-SKM treatment to time fungicide applications led to two fewer fungicides applications compared to the CP, however overall risk was higher than the CP. The CP-SKM treatment, despite two fewer fungicide applications, had higher risk because the amount of chlorothalonil applied in this treatment compared to the CP was greater. Chlorothalonil is often repeatedly applied during the summer months in Midwestern climates to manage a broad range of fungal diseases and to suppress fungicide-resistant dollar spot populations. The CP-SKM treatment had three applications of chlorothalonil compared to two applications in the CP to help improve overall efficacy and resistance management of the CP-SKM treatment. These results make clear, however, that product selection plays a larger role in risk reduction than the overall number of fungicide applications because of the large differences in risk between products.

Another central goal of both studies was to determine the ability of reduced risk pesticide programs to control dollar spot. The lowest risk program in both studies, the reduced risk treatment and the 25%-SA treatment, actually had the lowest mean AUDPC dollar spot values across all years and study locations (Tables 4, 5). In addition, these two programs had turf quality and color values that were statistically indistinguishable from the conventional programs. Thus, a reduced risk program was just as effective or more effective in controlling dollar spot and yielding a high-quality turf surface than conventional programs. Altering product selection in favor of lower risk products, pesticide risk for the control of dollar spot on golf course fairways can be reduced sharply without compromising dollar spot control.

The fungicide chlorothalonil had the single biggest impact on risk in both studies. This was evident when comparing the CP and CP-SKM programs in the SKM Reduced Risk study but was also evident in the State Risk Comparison study. Chlorothalonil has long been used in golf course management because of its favorable characteristics as a low-cost, broad-spectrum fungicide that is not susceptible to issues of fungal resistance (Latin, 2021). Despite golf courses only covering less than a half percent the land area as US agriculture, US golf courses account for 10% of total chlorothalonil use (Van Scoy and Tjeerdema, 2014). All of the 23 golf courses that submitted pesticide records to calculate an average pesticide risk for Wisconsin used chlorothalonil in their pesticide application programs. Chlorothalonil, as measured by HQ, is not a highly toxic pesticide because it has modest mammalian acute toxicity. The HQ model was used to defined risk levels of the State Risk Comparison study, and thus the 50%-SA treatment could include chlorothalonil. However, EIQ and PRT assign chlorothalonil a much higher risk value, which is why the 50%-SA treatment was the highest risk treatment, as quantified by PRT and EIQ, in the State Risk Comparison Study. In the PRT, chlorothalonil has a “high risk” score in the aquatic invertebrates and (human) dermal cancer categories, and a “moderate risk” score in the avian reproductive and aquatic algae categories. The EIQ value of chlorothalonil is 37, which is considered “high.” Indeed, recent toxicology studies indicate that chlorothalonil, while not acutely toxic to mammals, is an endocrine distributor in mice, and secondary metabolites of chlorothalonil can be particularly toxic to fish (Zhang et al., 2016; Hao et al., 2019). Our research indicates that reducing or removing chlorothalonil from a pest management program is the single most effective way to reduce pesticide risk on golf courses.

In the 25% treatment in the State Risk Comparison Study and the RR-SKM treatment in the SKM Reduced Risk study, the fungicides in the program are primarily lower risk pesticides such as boscalid, fluxapyroxad, and penthiopyrad from the SDHI fungicide class. Unfortunately, dollar spot populations can quickly develop resistance to SDHI fungicides and practitioners must use them wisely or risk developing SDHI resistant strains on their golf course (Sang et al., 2015; Popko et al., 2018). As such, pesticides like fluazinam were incorporated into the lower risk programs beginning in 2015 as a broad-spectrum, resistance-management fungicide with a lower risk profile than the traditional broad-spectrum fungicides that have higher risk profiles, such as chlorothalonil. Building a pesticide application program around lower risk fungicides while still incorporating lower risk resistance management chemistries, such as fluazinam, is a way to both reduce risk but decrease the likelihood of resistance development.

Trends in pesticide risk over time on US golf courses are unknown due to a lack of nationwide data, but multiple analyses have been conducted in US agriculture. Kniss (2017) found that herbicide use in US agriculture increased from 1990 to 2015 but that acute and chronic mammalian risk was mostly either declining or stable. However, Kniss (2017) only measured the chronic and acute risk trends to mammals. Schulz et al. (2021) found that despite large reductions in pesticide risk to fish, mammals, and birds, pesticide risk to aquatic invertebrates and pollinators has increased significantly over the past 25 years in US agriculture. Increased risk to aquatic invertebrates and pollinators was primarily a result of pyrethroid and neonicotinoid insecticides and came as insecticide use rates declined. These results highlight the importance both of calculating pesticide risk instead of weight of pesticide applied as a measure of ecological impact and estimating pesticide risk to a wide variety of non-target organisms.

Golf courses in Wisconsin spend on average $42,000 on pesticides annually (Bekken et al., 2021). Bekken et al. (2021) found that golf courses with lower maintenance budgets have lower pesticide risk than golf courses with higher maintenance budgets, suggesting that money is a limiting factor in pesticide use for lower maintenance budget golf courses. In the State Risk Comparison Study, the 25%-SA treatment reduced risk by 75% compared to the Wisconsin statewide average but only reduced costs by 10%. Thus, for practical purposes the 25%-SA program is cost neutral. In the SKM Reduced Risk study, the reduced risk treatment (RR-SKM) was nearly double the cost of the conventional program (CP). When calculating treatment costs from the provided distributor price list it was apparent the increased cost of the reduced risk programs was largely due to the increased cost of these products compared to older, more highly toxic compounds. For the average US golf course with 11.4 ha of fairways, the annual cost of the RR-SKM program would be ~$58,000 USD, compared to $36,200 for the CP. Given these significant cost differences, golf course superintendents are often not willing or not able to voluntarily pay higher prices for lower risk pesticides. The incentive for golf course superintendents to adopt reduced risk strategies is nuanced and will differ based on numerous situations. There is some scientific evidence that golf course superintendents have worse health outcomes which may be linked to on-the-job pesticide exposure (Kross et al., 1996; Knopper and Lean, 2010). Despite such evidence, Arcury-Quandt et al. (2011) found that most US golf course superintendents interviewed did not think that pesticides present a personal health risk. The authors are aware of no regulations in the US that require golf courses to track pesticide risk, or that set an upper limit on golf course pesticide risk. Even voluntary sustainability certification programs for golf courses, such as Audubon International and GEO-Certified, do not require golf courses to track pesticide risk. Superintendents in the US often feel pressure from golfers to maintain high quality pest-free playing surfaces, and therefore are incentivized to use more pesticide rather than less.

Denmark is a case study of how pesticide risk can be monitored across the golf industry. In Denmark, all golf courses are required to use an online pesticide application software [Scandinavian Turfgrass Environmental Research Foundation (STERF), 2019]. The program calculates pesticide risk using a pesticide risk indicator similar to EIQ. Under an agreement between the Danish Golf Union (DGU) and Danish Government, pesticide risk for each golf course component (greens, tees, fairways, and roughs) is capped every year [Scandinavian Turfgrass Environmental Research Foundation (STERF), 2019]. The amount of pesticide risk allowed on each component is being slowly reduced each year until a minimum sustainable level is reached, though what this level will be is currently a point of debate between the DGU and Danish Government. The Danish model of pesticide risk reduction has been effective in reducing pesticide risk on golf courses in the country and could serve as an example for reducing pesticide risk on golf courses around the world [Scandinavian Turfgrass Environmental Research Foundation (STERF), 2019]. Short of regulation, concepts of pesticide risk and how pesticide risk is calculated and tracked should be part of both the undergraduate turfgrass science curriculum and continuing education programs offered by superintendents' associations. In addition, voluntary sustainability standards in golf such as Audubon International and GEO-Certified could require their golf courses to track pesticide risk through software such as Playbooks for Golf ^®, which can calculate a golf course's EIQ score. Other software programs used by superintendents to track pesticide applications should also consider integrating Pesticide risk indicators, such as EIQ, HQ, or PRT.

Dollar spot is the most common and costly disease for golf course superintendents in temperate US climates to control, but the repeated pesticide applications to control dollar spot can increase fungal resistance and risk both human and environmental health. This study found that lower risk fungicides can reduce pesticide risk by over 50% without sacrificing disease control. While such high levels of risk reduction may not be possible on all golf courses given the constraints of maintenance budgets and the development of fungal resistance, significant reductions in golf course pesticide risk to control dollar spot are achievable simply by making changes in product selection. Future advancements in disease resistant grasses, precision application technology, and plant-based fungicides may bring further reductions in pesticide risk and increase the sustainability of US golf courses.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

KH and PK contributed to conceptualization, design of the study, field work, and editing paper drafts. MB and PK contributed to the data and statistical analyses. MB organized the database, produced the figures and tables, and wrote the first draft of the manuscript. All authors contributed to the manuscript revision and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.