[et_pb_section fb_built=”1″ _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”][et_pb_row _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”][et_pb_column type=”4_4″ _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”][et_pb_cta title=”GreenKeeper Growth Series Part 4″ button_text=”Click Here” _builder_version=”4.21.0″ _module_preset=”default” background_color=”#8C8C8C” box_shadow_style=”preset1″ global_colors_info=”{}”]<\/p>\n
In the last article of our Growth Series<\/a>, we explored how putting green growth responds to changes in weather. Heavy rainfall increased clipping volume somewhat, while the first significant cold front of autumn produced a much larger reduction in growth. Those findings were interesting, but they also reinforced something most superintendents already recognize: putting greens don’t all grow the same way.<\/em><\/p>\n [\/et_pb_cta][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n Two golf courses can experience similar weather, apply similar amounts of nitrogen, and use comparable PGR programs, yet collect very different quantities of clippings. Even greens within the same facility can have noticeably different growth rates. That makes predicting growth more complicated than simply looking at temperature or calculating growing degree days.<\/p>\n Over the past several months, we’ve been using GreenKeeper’s clipping-volume database to explore how accurately we can predict growth and, more importantly, how those predictions can become more useful for individual golf courses.<\/p>\n [\/et_pb_text][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n Temperature is one of the strongest drivers of turfgrass growth, which is why growth potential models have been useful for decades. However, temperature alone cannot explain everything that influences clipping production.<\/p>\n Research by Qiyu Zhou and Doug Soldat (2021)<\/a> at the University of Wisconsin demonstrated this challenge. They used machine learning to predict creeping bentgrass clipping yield from weather and management information. Their model performed reasonably well on Wisconsin research greens but poorly when applied to a golf course in Minnesota. Predictions improved considerably when they developed a model using that golf course’s own clipping data.<\/p>\n Our GreenKeeper research has reached a similar conclusion using clipping records from a much larger collection of golf courses. A general growth model can provide a useful starting point for understanding how growth changes with weather, but it doesn’t necessarily know how much grass an individual facility normally produces.<\/p>\n Fortunately, we found that developing a completely separate growth model for every golf course may not be necessary. By using clipping measurements to adjust the general model to match a course’s normal growth level, we can improve its predictions without requiring an entirely new model.<\/p>\n [\/et_pb_text][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n GreenKeeper’s general growth model for cool-season putting greens is deliberately simple. It uses one weather measurement: the average temperature over the past three days, calculated from each day’s high and low. From that temperature, the model estimates two things.<\/p>\n The first is how many putting greens are still in their mowing season. We learned this from when golf courses in the GreenKeeper network begin and stop recording clippings each year. When the three-day average temperature falls to about 45 \u00b0F, roughly half of cool-season courses have stopped collecting clippings for the year. At about 54 \u00b0F, roughly 8 in 10 are still mowing, and above 70 \u00b0F nearly all of them are.<\/p>\n The second is how fast a green grows while it’s still being mowed. Growth is close to zero near freezing, about 0.65 quarts of clippings per 1,000 square feet per day at 50 \u00b0F, about 1 quart at 68 \u00b0F, and about 1.3 quarts between 77 and 86 \u00b0F. It eases slightly in the hottest weather, to about 1.15 quarts at 95 \u00b0F.<\/p>\n The model multiplies those two estimates together. On a day when the three-day average is 50 \u00b0F, for example, it predicts about 0.44 quarts per 1,000 square feet. At 68 \u00b0F, it predicts about 1 quart. We built the model from one-day clipping measurements at about 200 cool-season golf courses using the GreenKeeper App between 2024 and August 2026, and we tested it on golf courses it had never seen.<\/p>\n What the model leaves out is just as important. We tested adding humidity and the time of year, and neither improved predictions at golf courses the model hadn’t seen, so we left them out. The model also doesn’t yet know about rainfall, irrigation, nitrogen, PGR applications, or mowing height. Because it doesn’t know the calendar, it treats a 55 \u00b0F day in April the same as a 55 \u00b0F day in October, even though greens usually grow more in spring than in fall at the same temperature. And because it only looks back three days, it doesn’t know how long a cold spell has lasted, and it can’t tell a frost from a cold night. In our latest analysis, the model captured about half of the slowdown in growth that followed the first cold front of autumn.<\/p>\n [\/et_pb_text][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n One of the questions we’ve been investigating is how quickly GreenKeeper can learn what’s normal for a particular golf course.<\/p>\n Our latest analysis suggests that approximately 20 to 30 clipping measurements provide a reasonably stable starting point<\/strong> for adjusting the general growth model to an individual facility. For this analysis, we focused on measurements collected on consecutive days so we could confidently associate each clipping catch with one day of growth.<\/p>\n A few measurements can already tell us whether a golf course generally produces more or fewer clippings than expected. Additional measurements make that estimate more reliable. Importantly, we don’t need several years of data to begin learning something useful.<\/p>\n [\/et_pb_text][et_pb_image src=”https:\/\/www.greenkeeperapp.com\/marketing\/wp-content\/uploads\/2026\/11\/growth-series-4-normal-growth-level.png” alt=”Line chart of the share of golf courses whose estimate of normal putting green growth is already close to its long-run value, by the number of one-day clipping measurements logged: 22 percent after 3 measurements, 17 percent after 5, 27 percent after 7, 35 percent after 10, 44 percent after 14, 59 percent after 21 and 75 percent after 30.” title_text=”growth-series-4-normal-growth-level” show_in_lightbox=”on” align=”center” max_width=”85%” _builder_version=”4.21.0″ _module_preset=”default” box_shadow_style=”preset1″ global_colors_info=”{}”][\/et_pb_image][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” text_text_color=”#5a5952″ text_font_size=”15px” max_width=”85%” module_alignment=”center” custom_margin=”-12px||||false|false” global_colors_info=”{}”]<\/p>\n Figure 1.<\/strong> How clipping measurements help the model learn a golf course. Each point shows the share of 106 cool-season golf courses whose estimate of normal growth, meaning how much more or less their greens grow than the general model expects, was already within 25 percent of the estimate the same course reached after 60 measurements. Only clippings collected the day after the previous mowing are counted, so each measurement represents one day of growth.<\/em><\/p>\n [\/et_pb_text][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n That doesn’t mean 30 measurements are enough to predict exactly how many clippings will be collected tomorrow. Day-to-day growth is influenced by many factors, including weather, nitrogen, PGRs, moisture, and mowing practices. Even with extensive data, accurately predicting every day’s clipping volume remains difficult.<\/p>\n Instead, these measurements help establish a baseline for comparing actual growth with what we would normally expect under the prevailing conditions.<\/p>\n [\/et_pb_text][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n We’ve also built a free public map that lets turf managers explore our general cool-season putting green growth model. The map at growth.greenkeeperapp.com<\/a> shows the daily growth the model predicts for a cool-season putting green across the country, from January 1 of this year through the days ahead in the weather forecast. You can step through the days, animate the season, or search for a golf course to see its predicted growth over time.<\/p>\n The map shows the growth pattern predicted by the broader model. It does not use clipping measurements to adjust predictions to your individual golf course, and it doesn’t account for your nitrogen or PGR applications. It also shows a cool-season green everywhere, so in warm climates such as Florida and the Gulf Coast, it shows what a cool-season green would do in that weather, not what a bermudagrass green is doing.<\/p>\n [\/et_pb_text][et_pb_image src=”https:\/\/www.greenkeeperapp.com\/marketing\/wp-content\/uploads\/2026\/11\/growth-series-4-growth-map-scaled.jpg” alt=”Screenshot of the GreenKeeper cool-season putting green growth map of the United States for November 4, 2026, with predicted daily clipping growth shaded from tan for slow growth to dark green for fast growth, a date control along the bottom and a golf course search box.” title_text=”growth-series-4-growth-map” show_in_lightbox=”on” align=”center” max_width=”85%” _builder_version=”4.21.0″ _module_preset=”default” box_shadow_style=”preset1″ global_colors_info=”{}”][\/et_pb_image][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” text_text_color=”#5a5952″ text_font_size=”15px” max_width=”85%” module_alignment=”center” custom_margin=”-12px||||false|false” global_colors_info=”{}”]<\/p>\n Figure 2.<\/strong> The GreenKeeper growth map for November 4, 2026. Colors show the daily clipping growth the general model predicts for a cool-season putting green under each location’s weather, from tan for slow growth to dark green for fast growth.<\/em><\/p>\n [\/et_pb_text][et_pb_button button_url=”https:\/\/growth.greenkeeperapp.com\/” url_new_window=”on” button_text=”Explore the Growth Map” button_alignment=”center” _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”][\/et_pb_button][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n That’s an important distinction. The general model helps illustrate how weather and seasonal conditions influence expected growth. Your clipping records tell us how the turf at your facility actually responds.<\/p>\n Our longer-term goal is to bring those two sources of information together within GreenKeeper. Rather than asking superintendents to interpret growth potential alone, we want to compare expected growth with observed clipping production and eventually account for the management practices that influence both.<\/p>\n [\/et_pb_text][et_pb_divider _builder_version=”4.21.0″ _module_preset=”default” custom_margin=”||1px|||” global_colors_info=”{}”][\/et_pb_divider][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”]<\/p>\n Consider a situation where clipping volume increases unexpectedly. Did warmer weather increase growth? Is a PGR application beginning to wear off? Did the turf respond to a recent nitrogen application, or is the increase simply normal variation?<\/p>\n Today, superintendents answer those questions using experience, observations, and management records. Our goal is to make that process more objective by combining GreenKeeper’s weather models, clipping measurements, and detailed management histories.<\/p>\n We haven’t yet demonstrated that the new growth model can reliably distinguish the effects of nitrogen and PGR applications from other sources of variation. That’s part of the research we’re working toward, and it will be an important topic in future articles.<\/p>\n For now, the practical recommendation is simple: keep measuring clipping volumes.<\/strong> Record them consistently, preferably from the same representative greens, and continue building a history of how your turf grows through changing weather and management conditions.<\/p>\n Those records are already useful for evaluating growth. Over time, they can also help GreenKeeper develop more accurate, locally informed predictions that make those measurements even more valuable.<\/p>\n The objective isn’t to predict every clipping catch perfectly. It’s to better understand what’s normal for your greens, recognize meaningful changes, and use that information to make better management decisions.<\/p>\n [\/et_pb_text][et_pb_button button_url=”https:\/\/www.greenkeeperapp.com\/register” button_text=”Join GreenKeeper” button_alignment=”center” _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”][\/et_pb_button][et_pb_text _builder_version=”4.21.0″ _module_preset=”default” background_color=”#F2F2F2″ max_width=”94%” module_alignment=”center” custom_margin=”0px||||false|false” custom_padding=”10px|10px|10px|10px|false|false” text_font_size=”15px” box_shadow_style=”preset1″ global_colors_info=”{}”]<\/p>\n Zhou, Q., and D.J. Soldat. 2021. Creeping bentgrass yield prediction with machine learning models<\/a>. Front. Plant Sci. 12:749854. https:\/\/doi.org\/10.3389\/fpls.2021.749854<\/a><\/p>\n [\/et_pb_text][\/et_pb_column][\/et_pb_row][et_pb_row column_structure=”1_2,1_2″ _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”][et_pb_column type=”1_2″ _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”][et_pb_button button_url=”https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/what-happens-to-putting-green-growth-when-the-weather-changes-fast\/” button_text=”\u2190 Part 3: When Weather Changes” button_alignment=”left” button_alignment_tablet=”” button_alignment_phone=”center” button_alignment_last_edited=”on|phone” _builder_version=”4.21.0″ _module_preset=”default” custom_button=”on” button_text_size=”14px” button_letter_spacing=”1px” button_use_icon=”off” global_colors_info=”{}”][\/et_pb_button][\/et_pb_column][et_pb_column type=”1_2″ _builder_version=”4.21.0″ _module_preset=”default” global_colors_info=”{}”][\/et_pb_column][\/et_pb_row][\/et_pb_section]<\/p>\n","protected":false},"excerpt":{"rendered":" In the last article of our Growth Series, we explored how putting green growth responds to changes in weather. Heavy rainfall increased clipping volume somewhat, while the first significant cold front of autumn produced a much larger reduction in growth. Those findings were interesting, but they also reinforced something most superintendents already recognize: putting greens […]<\/p>\n","protected":false},"author":2,"featured_media":102618,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"on","_et_pb_old_content":"","_et_gb_content_width":"","_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[5,64],"tags":[31,55,52],"class_list":["post-102616","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-blog","category-greenkeeper-growth-series","tag-clipvol","tag-modeling","tag-ptm"],"jetpack_featured_media_url":"https:\/\/www.greenkeeperapp.com\/marketing\/wp-content\/uploads\/2026\/11\/growth-series-4-growth-map-scaled.jpg","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/posts\/102616"}],"collection":[{"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/comments?post=102616"}],"version-history":[{"count":5,"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/posts\/102616\/revisions"}],"predecessor-version":[{"id":102622,"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/posts\/102616\/revisions\/102622"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/media\/102618"}],"wp:attachment":[{"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/media?parent=102616"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/categories?post=102616"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.greenkeeperapp.com\/marketing\/index.php\/wp-json\/wp\/v2\/tags?post=102616"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}Why a General Growth Model Isn’t Enough<\/h5>\n
What Goes Into the General Growth Model<\/h5>\n
How Many Clipping Measurements Do We Need?<\/h5>\n
A First Look at GreenKeeper’s Growth Model<\/h5>\n
What Can We Do With That Information?<\/h5>\n
References<\/h5>\n