| |
Web URL(s): | https://scisoc.confex.com/scisoc/2021am/meetingapp.cgi/Paper/135396 Last checked: 03/24/2022 Requires: JavaScript |
Publication Type:
| Report |
Content Type: | Abstract or Summary only |
Author(s): | Zhou, Qiyu |
Author Affiliation: | University of Wisconsin-Madison, Madison, WI |
Title: | Precision nitrogen (N) management with machine learning approaches |
Section: | Golf turf management oral: Cultural practices, physiology, and water (includes student competition) Other records with the "Golf turf management oral: Cultural practices, physiology, and water (includes student competition)" Section
C05 turfgrass science Other records with the "C05 turfgrass science" Section
|
Meeting Info.: | Salt Lake City, Utah: November 7-10, 2021 |
Source: | ASA, CSSA and SSSA International Annual Meetings. 2021, p. 135396. |
Publishing Information: | [Madison, Wisconsin]: [American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America] |
# of Pages: | 1 |
Abstract/Contents: | "Nitrogen (N) is the most limiting nutrient for turfgrass growth. Instead of pursuing the maximum yield, most turfgrass managers use 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 into account temperature, limiting its accuracy. In this study, we proposed a machine-learning-based turf growth model using the random forest (RF) model for estimating short-term turfgrass clipping yield which guided N application decisions. This study aimed to compare various N application strategies in terms of N applied and turfgrass performance characteristics. The field experiment was conducted on two sand-based research greens at University of Wisconsin-Madison turfgrass research facility, Madison WI in 2020 and 2021. RF model was built based on a set of variables including 7-day weather, traffic intensity, soil moisture content, N fertilization rate, and the spectral reflectance - normalized difference red edge (NDRE) vegetation index. Four different N strategies were planned based on 1) estimated turfgrass yield using RF model; 2) PACE TURF growth potential (GP) model; 3) reflectance data-guided (NDRE) method; and 4) experience-depended method that constantly applying 10kg ha-1 every other week. The accumulative applied N fertilization ranked (from high to low) of four N strategies was PACE TURF GP mode, experience-depended method, RF model and NDRE-based method. N application strategy made according to RF model saved around 50% N fertilizer compared to Pace Turf GP model, while the average turfgrass visual quality was similar. NDRE-based methods produced the least turfgrass visual performance and an unsatisfied playing surface. This study demonstrated the feasibility of using a machine-learning-based yield prediction model to guide reasonable N fertilization decisions on golf course putting greens" |
Language: | English |
References: | 0 |
Note: | This item is an abstract only! "46-4" |
| ASA/CSSA/SSSA Citation (Crop Science-Like - may be incomplete): Zhou, Q. 2021. Precision nitrogen (N) management with machine learning approaches. Agron. Abr. p. 135396. |
| Fastlink to access this record outside TGIF: https://tic.msu.edu/tgif/flink?recno=317026 |
| If there are problems with this record, send us feedback about record 317026. |
| Choices for finding the above item: |
| Web URL(s): https://scisoc.confex.com/scisoc/2021am/meetingapp.cgi/Paper/135396 Last checked: 03/24/2022 Requires: JavaScript |
| Request through your local library's inter-library loan service (bring or send a copy of this TGIF record) |