Analyzing golf course pesticide risk across the US and Europe-The importance of regulatory environment.

This study quantifies golf course pesticide risk in five regions across the US (Florida, East Texas, Northwest, Midwest, and Northeast) and three countries in Europe (UK, Denmark, and Norway) with the objective of determining how pesticide risk on golf courses varied as a function of climate, regulatory environment, and facility-level economic factors. The hazard quotient model was used to estimate acute pesticide risk to mammals specifically. Data from 68 golf courses are included in the study, with a minimum of at least five golf courses in each region. Though the dataset is small, it is representative of the population at confidence level of 75 % with a 15 % margin of error. Pesticide risk appeared to be similar across US regions with varied climates, and significantly lower in the UK, and lowest in Norway and Denmark. In the Southern US (East Texas and Florida), greens contribute most to total pesticide risk while in nearly all other regions fairways make the greatest contribution to overall pesticide risk. The relationship between facility-level economic factors such as maintenance budget was limited in most regions of the study, except in the Northern US (Midwest, Northwest, and Northeast) where maintenance and pesticide budget correlated to pesticide risk and use intensity. However, there was a strong relationship between regulatory environment and pesticide risk across all regions. Pesticide risk was significantly lower in Norway, Denmark, and the UK, where twenty or fewer active ingredients were available to golf course superintendents, than it was in US where depending on the state between 200 and 250 pesticide active ingredients were registered for use on golf courses.

Evaluating Decision Support Tools for Precision Nitrogen Management on Creeping Bentgrass Putting Greens.

Nitrogen (N) is the most limiting nutrient for turfgrass growth. Few tools or soil tests exist to help managers guide N fertilizer decisions. Turf growth prediction models have the potential to be useful, but the lone turfgrass growth prediction model only takes into account temperature, limiting its accuracy. This study investigated the ability of a machine learning (ML)-based turf growth model using the random forest (RF) algorithm (ML-RF model) to improve creeping bentgrass (Agrostis stolonifera) putting green management by estimating short-term clipping yield. This method was compared against three alternative N application strategies including (1) PACE Turf growth potential (GP) model, (2) an experience-based method for applying N fertilizer (experience-based method), and (3) the experience-based method guided by a vegetative index, normalized difference red edge (NDRE)-based method. The ML-RF model was built based on a set of variables including 7-day weather, evapotranspiration (ET), traffic intensity, soil moisture content, N fertilization rate, NDRE, and root zone type. The field experiment was conducted on two sand-based research greens in 2020 and 2021. The cumulative applied N fertilizer was 281 kg ha-1 for the PACE Turf GP model, 190 kg ha-1 for the experience-based method, 140 kg ha-1 for the ML-RF model, and around 75 kg ha-1 NDRE-based method. ML-RF model and NDRE-based method were able to provide customized N fertilization recommendations on different root zones. The methods resulted in different mean turfgrass qualities and NDRE. From highest to lowest, they were PACE Turf GP model, experience-based, ML-RF model, and NDRE-based method, and the first three methods produced turfgrass quality over 7 (on a scale from 1 to 9) and NDRE value over 0.30. N fertilization guided by the ML-RF model resulted in a moderate amount of fertilizer applied and acceptable turfgrass performance characteristics. This application strategy is based on the N cycle and has the potential to assist turfgrass managers in making N fertilization decisions for creeping bentgrass putting greens.

Creeping Bentgrass Yield Prediction With Machine Learning Models.

Nitrogen is the most limiting nutrient for turfgrass growth. Instead of pursuing the maximum yield, most turfgrass managers use nitrogen (N) to maintain a sub-maximal growth rate. Few tools or soil tests exist to help managers guide N fertilizer decisions. Turf growth prediction models have the potential to be useful, but the currently existing turf growth prediction model only takes temperature into account, limiting its accuracy. This study developed machine-learning-based turf growth models using the random forest (RF) algorithm to estimate short-term turfgrass clipping yield. To build the RF model, a large set of variables were extracted as predictors including the 7-day weather, traffic intensity, soil moisture content, N fertilization rate, and the normalized difference red edge (NDRE) vegetation index. In this study, the data were collected from two putting greens where the turfgrass received 0 to 1,800 round/week traffic rates, various irrigation rates to maintain the soil moisture content between 9 and 29%, and N fertilization rates of 0 to 17.5 kg ha-1 applied biweekly. The RF model agreed with the actual clipping yield collected from the experimental results. The temperature and relative humidity were the most important weather factors. Including NDRE improved the prediction accuracy of the model. The highest coefficient of determination (R2) of the RF model was 0.64 for the training dataset and was 0.47 for the testing data set upon the evaluation of the model. This represented a large improvement over the existing growth prediction model (R 2 = 0.01). However, the machine-learning models created were not able to accurately predict the clipping production at other locations. Individual golf courses can create customized growth prediction models using clipping volume to eliminate the deviation caused by temporal and spatial variability. Overall, this study demonstrated the feasibility of creating machine-learning-based yield prediction models that may be able to guide N fertilization decisions on golf course putting greens and presumably other turfgrass areas.

A novel framework for estimating and analyzing pesticide risk on golf courses.

This study develops a framework that quantifies golf course pesticide risk, explores environmental and economic factors that may be responsible for the observed risk, develops a method to compare golf course pesticide risk to other agricultural crops and investigates how pesticide risk on golf courses can be most effectively reduced. To quantify pesticide risk, we adapt the Environmental Impact Quotient (EIQ) and hazard quotient models for use on golf courses. The EIQ model provides an estimate of overall environmental risk, while the hazard quotient model, as applied here, provides an estimate of pesticide risk to mammals. This novel framework was applied to twenty-two courses in Wisconsin and New York, USA. Using both pesticide risk models, all twenty-two golf courses showed a high coefficient of variation of pesticide risk (<0.76). Within a golf course, mean absolute pesticide risk was at least two times higher on fairways than on greens, tees, or roughs. Mean area normalized risk was at least three times higher on greens than the other three golf course components. Pesticide risk of a component-weighted average of greens, tees, fairways and roughs on each course were within the range of pesticide risk calculated for five other agricultural crops. Our data suggest that variation in pesticide risk on golf courses is related to economic factors, such as maintenance budget, and can be effectively lowered by reducing pesticide use on fairways and selecting products of lower risk. To assist golf course superintendents in developing programs that lower pesticide risk, a new metric was developed: the Risk to Intensity Quotient (RIQ). The RIQ is the ratio of pesticide risk to use intensity and quantifies the average risk of product selection by a golf course superintendent.

Corrigendum: Creeping Bentgrass Yield Prediction With Machine Learning Models.

[This corrects the article DOI: 10.3389/fpls.2021.749854.].

Use of Routine Soil Tests to Estimate Pb Bioaccessibility.

Soil lead (Pb) hazard level is contingent on bioavailability, but existing assays that estimate Pb bioavailability for human health risks are too expensive or otherwise inaccessible to many people that are impacted by Pb-contaminated soil. This study investigated the use of routine soil nutrient tests to estimate soil-Pb bioaccessibility as a surrogate measure of Pb bioavailability. A silt loam soil was spiked to a target concentration of 2000 mg Pb kg-1 with Pb(NO3)2 and amended with H3PO4 (varying P-to-Pb molar ratios) and KCl (Cl-to-P molar ratio of 2:5) to generate soils with similar total Pb concentrations but a range of Pb bioavailability (and bioaccessibility). Soils were extracted using Mehlich 3, Mehlich 1, Bray P1, Olsen, and  micronutrient (DTPA) methods, and the results were compared to U.S. Environmental Protection Agency method 1340 data as well as to extended X-ray absorption fine structure (EXAFS) spectroscopy. The Mehlich 3 and method 1340 treatment effect ratios were well-correlated ( r2 = 0.88, p ≤ 0.05), whereas Bray P1, DTPA, and Olsen results were more reflective of EXAFS data. Preliminary animal-feeding trials suggest that the Mehlich 3 is as effective as method 1340 at predicting the impact of P treatment on Pb relative bioavailability; however, both methods over-estimated the Pb hazard to mice in P-amended soil. Other routine soil tests that have heightened sensitivity to P amendment (e.g., Bray P1) may be promising candidates for Pb bioaccessibility assessment.

Synthesis and detection of oxygen-18 labeled phosphate.

Phosphorus (P) has only one stable isotope and therefore tracking P dynamics in ecosystems and inferring sources of P loading to water bodies have been difficult. Researchers have recently employed the natural abundance of the ratio of (18)O/(16)O of phosphate to elucidate P dynamics. In addition, phosphate highly enriched in oxygen-18 also has potential to be an effective tool for tracking specific sources of P in the environment, but has so far been used sparingly, possibly due to unavailability of oxygen-18 labeled phosphate (OLP) and uncertainty in synthesis and detection. One objective of this research was to develop a simple procedure to synthesize highly enriched OLP. Synthesized OLP is made up of a collection of species that contain between zero and four oxygen-18 atoms and, as a result, the second objective of this research was to develop a method to detect and quantify each OLP species. OLP was synthesized by reacting either PCl(5) or POCl(3) with water enriched with 97 atom % oxygen-18 in ambient atmosphere under a fume hood. Unlike previous reports, we observed no loss of oxygen-18 enrichment during synthesis. Electrospray ionization mass spectrometry (ESI-MS) was used to detect and quantify each species present in OLP. OLP synthesized from POCl(3) contained 1.2% P(18)O(16)O(3), 18.2% P(18)O(2) (16)O(2), 67.7% P(18)O(3) (16)O, and 12.9% P(18)O(4), and OLP synthesized from PCl(5) contained 0.7% P(16)O(4), 9.3% P(18)O(3) (16)O, and 90.0% P(18)O(4). We found that OLP can be synthesized using a simple procedure in ambient atmosphere without the loss of oxygen-18 enrichment and ESI-MS is an effective tool to detect and quantify OLP that sheds light on the dynamics of synthesis in ways that standard detection methods cannot.