Fertilizer Mismanagement Impacts on Pasture Health

Final report for OS20-132

Project Type: On-Farm Research
Funds awarded in 2020: $19,828.00
Projected End Date: 03/31/2022
Grant Recipient: University of Florida
Region: Southern
State: Florida
Principal Investigator:
Cheryl Mackowiak
University of Florida
Expand All

Project Information


Developing a Site-specific Nutrient Management (SSNM) technique for assessing pastures will be a powerful tool for preventing pasture decline due to poor forage fertilization. In 2013, a UF-IFAS taskforce investigated pasture decline problems in Florida. A 2015 internal report by Dr. Lynn Sollenberger (forage agronomist) summarized that of the thirteen tested locations (all in south-central Florida), none of the pasture problems appeared related to poor grazing management. However, low soil fertility (P or K) was identified in about 70% of samples but low P (based upon IFAS recommendations) was not identified in any of the forage samples. Some observed instances of leaf fungal disease were noted, but it was reported not to be a concern. Although the sample size was limited, evidence suggested that plant stress due to low potassium fertility was often the cause to poor pasture health where they sampled. However, sampling was greatly limited and secondary causes, such as weather and root diseases were not well documented.

In the SSNM method, a single nutrient factor is omitted from a treatment area otherwise receiving complete fertilization. Different single plant essential nutrients can be tested at once at the plot-scale. In comparison of dose-response method, the SSNM method can more quickly determine which nutrients are most likely affecting yield and stand health within a producer’s field, under their unique conditions. We have adapted the SSNM method for on-farm pasture testing by increasing the number or treatments “omission plots” and adding replication. The SSNM technique was originally developed to test and demonstrate fertilization effects on rice and it can be a valuable tool for many on-farm commodities (Dobermann and Cassman, 2002). Applying this technique at ranches across the state (south Florida to the Panhandle) helps us to determine leading nutrient concerns using a greater representation of soil types and environmental conditions under a restricted budget. As nutrient depletion develops, the plant may also display symptoms from other stressors, such as low soil moisture and fungal diseases.

An on-farm pilot test was established in a bahiagrass pasture in 2016. By 2018, low K plots at this location had declining yields, stand density, and increased disease occurrences (Bipolaris and Take-All root rot). In comparison, omitted P plots maintained bahiagrass productivity while also negatively impacting the predominant weed species. This is an especially exciting finding, as bahiagrass was reported to have a critical tissue P value below 0.15% (Silveira et al. 2011), which is lower than for many other plant species. Fertilization management also impacts root/rhizome biomass, which is a leading method for building soil organic matter. In the pilot study, K depleted plots were losing root biomass compared to plots depleted in P or from plots receiving complete fertilization through either class AA biosolids or by mineral fertilization.

Project Objectives:

The proposed treatment plots were established in pastures at three working ranches across Florida in 2019 and another in 2016. The modified SSNM method will be used, relying on replicated “omission plots” to test and demonstrate temporal responses of bahiagrass pasture stands to individually depleted nutrients (N, P, K). We included a Class AA biosolids treatment to represent an organic, slow-release macronutrient source and viable micronutrient source, which is often overlooked in pasture and hay production operations in the southern US. Additionally, we will provide limited analytic support within the tri-state area (Florida, South Georgia, South Alabama), in terms of soil and tissue fertility and disease diagnostics of bahiagrass fields suspected of pasture decline. We will rely on extension agents to identify suspect fields, which will result in an additional 6 to 12 samples from suspect pastures to help increase what we learn at the four ranch test sites, alone. Participating extension agents and cooperators will help with bahiagrass stand assessments, review management practices, and contribute to outreach publications that will be aimed at southern ranchers. The benefit of using previously established plots is that the nutrients of greatest concern (P and K), typically take more than one growing season to become depleted and affect the forage stand. There are currently no funds supporting this effort, just as P and K deficiencies are beginning to be observed. The three pastures (and 1 hay field) were excluded from livestock via temporary fencing since the time of plot establishment, in order to minimize confounding effects of animals redistributing nutrients across treatment plots during the study.

Treatments are as follows: 1) unfertilized control (no N, P, or K fertilizers), 2) complete fertilizer (N + P + K), 3) minus N (complete fertilizer except for N), 4) minus P (complete fertilizer except for P), 5) minus K (complete fertilizer except for K), and 6) complete fertilizer supplied as Class AA biosolids, based upon N requirement. This treatment supplies macro- and micronutrients, except for K, which will be applied as mineral K. Additionally, it will be used only with the spring application and mineral sources relied upon for the second application (to lessen high soil P accumulation rates). Treatments targeting N, P, and K will be applied at an 80 lbs acre-1 (90 kg ha-1) rate using soluble, mineral fertilizer, when soils test low or medium for nutrients other than N (UF-IFAS does not rely on a soil N analysis for determining N fertilization application recommendations).

Plots will be staged (cut to 7.5 cm stubble height) at spring “green-up”. Forages will be harvested twice per growing season (approximately June and September, depending on location), which is often the frequency used with bahiagrass hay producers and this will allow for better nutrient budget tracking, compared to more frequent harvests but limited tissue analyses. Treatments will be applied in spring and again after the first cutting. Harvested forage will be collected, wet weights recorded, then oven-dried, reweighed, and the dried forage analyzed for plant essential nutrients and crude protein (N x 6.25) by a commercial lab (Waters Agricultural Lab, Camilla GA). Soils will be sampled (composite 10 cores per plot to 15 cm depth using soil probe) at the end of each growing season and analyzed for pH, nutrient fertility (Mehlich-3 extractant), and estimated CEC, by a commercial lab (Waters Agricultural Lab). At the end of the second year, plots will also be compared for soil organic matter (Walkley Black method) by UF Plant and Soil analytic lab (Gainesville, FL). Additionally, roots + rhizomes will be sampled (one 900 cm2 x 15 cm depth sample per plot) at the end of the 2021 growing season, in order to compare treatment root mass differences, which contribute to building soil organic matter. Loose soil will be shaken from root mat and roots kept in a cooler for return to the lab, where they will be separated from soil, rinsed and then dried, in order to record dry mass. Data will be compiled and analyzed, following the final cutting of each season. The results and interpretations will be shared with participating county extension faculty and cooperators, soon thereafter. This will include a written report and personal visit or phone call to verify reports were receive and to address concerns/questions. Additionally, an advisory meeting will be scheduled in the Jan/Feb timeframe (participating agents and cooperators), in order to discuss past season results, trouble-shooting for Year 2, and potential implications for future pasture and hay field nutrient management practices, including state fertilization recommendations.


Click linked name(s) to expand/collapse or show everyone's info
  • Roy Carter - Producer (Educator)
  • Libbie Schmidt - Producer
  • Paulette Tomlinson - Producer
  • Dr. Marcelo Wallau (Researcher)
  • Joseph Walter
  • John White


Materials and methods:

There is concern and conflict over what appropriate fertilization recommendations for (Paspalum notatum) bahiagrass pastures should be. Many of the UF-IFAS recommendations were based on decades-old data and using often unspecified varieties. Additionally, pasture decline reports are addressed individually and typically as independent events due to various causes. It was our overall goal to determine the degree fertilization practices and plant nutritional health contribute to pastures that appear to be in decline. The Omission Plot method of demonstrating and testing bahiagrass responses to gradually depleted fertility (focus on N, P and K), is a somewhat novel way to observe and track what happens to a forage stand under slowly increasing fertility stress, under real-world conditions (on-farm) and different soils. Additionally, in order to shorten the time it takes for a nutrient deficiency to occur, we fenced off the on-farm test plots from livestock and repeatedly removed the forage with each sampling, effectively exporting those nutrients off the treatment site.

Since bahiagrass is our dominant perennial grass across all of Florida, we chose working ranch locations across the state that also are in close proximity to other large cattle operations, so those most affected by our research have opportunities to visit a site near them. Florida being a large state, has some variation in weather and even climate (more subtropical as one moves down the peninsula), which translates into a greater need for locating on-farm sites across the state (Fig. 1). SARE 1-year Fig 1 The Osceola research site was established in a pasture in 2016, while the other trial sites were established in mid-2019. The six treatments were as follows: 1) Complete fertilizer (N + P + K), 2) complete fertilizer supplied as Class AA Biosolids + KCl and applied, based upon N requirement, 3) -P (complete fertilizer except for P), 4) -K (complete fertilizer except for K), 5) -N (complete fertilizer except for N), and 6) unfertilized Check. The biosolids application rate was based on total N (not available N) through 2021. Over-fertilizing with P is a concern; however, P availability from Class AA biosolids should be less than from mineral fertilizer sources. This study will provide on-farm evidence of how available this P source was in Florida soils. Additionally, many ranchers are concerned that the current P recommendations are too low for bahiagrass and these sites were to provide evidence if this was the case. Treatments requiring N or K2O, received 80 lbs acre-1 (90 kg ha-1) and treatments requiring P received 40 lbs P2O per acre to ensure no target nutrient limitation and also it helped to test the Mehlich-3 calibration for soil test P. Soil pH (adjusted with Ag lime) and other macronutrients were supplemented as needed, based upon IFAS recommendations, while no micronutrients (Fe, Mn, B, Zn, Cu) were routinely reapplied (biosolids contain micronutrients). Tissue testing of all plant essential nutrients (excluding Mo) was used to track plant nutritional status within the different treatments. Two harvests (manual mower to cut to 7.5 cm stubble height) were conducted per year and location. Harvested clippings were returned to the lab and placed in forced air ovens at 60 C for at least one week or until dry. Samples were ground to pass through 2-mm sieve and stored at room temperature prior to shipping to a commercial lab for nutrient analysis. Soils were collected during the winter in 2020 and 2021, by using a soil probe and compositing 6 subsamples per plot. Soils were returned to the lab and air-dried (30 C) until dry. Soils were passed through a 2-mm sieve and sent to a commercial lab for soil fertility analysis (Mehlich-3 extraction). Data were analyzed using PROC MIX procedure in SAS (SAS version 9.4, Cary, NC). Means were compared using the PDIFF procedure and adjusted by Tukey's test at the 5% significance level. 

Research results and discussion:

A Site-specific Nutrient Management (SSNM) approach was used to help assess potential relationships between soil fertility imbalances on bahiagrass stand productivity, health, and longevity. With changing and often more hostile growing conditions, maintaining pasture health will likely become more challenging in the southeast U.S. To better understand producer awareness of potential problems, thirty-three Florida cattle producers completed a survey during 2020 that addressed potential pasture issues from their perspectives. In terms of impacts over the previous three years, over a third had often been impacted by excessive rainfall, compared to only 12% who had often been impacted by excessive drought conditions. Again, nearly 25% often had trouble establishing new pastures and over half often had weed pressure challenges. Weed species often proliferate in weak pastures. Approximately 25% recognized challenges with maintaining adequate soil fertility, while a third conveyed that soil fertility never or infrequently impacted their pastures or hay fields. More revealing was that among those who had observed pasture decline in the previous five years, just under half used soil fertility reports and less than 10% relied on tissue analysis as tools to address suspected pasture decline. When asked what training topics they wanted more information on, 42% included soil health and/or fertility and 51% ranked soil sampling as being a highly important tool for maintaining pasture health. Although the polling pool was small, a couple of trends seem apparent: 1) only half of sampled producers consider soil maintenance and management highly important to managing pasture health, and 2) and less than half who had pasture decline considered soil fertility to as a primary contributor to decline.

In 2020, the fall harvests demonstrated that a complete mineral fertilizer resulted in similar herbage yields for three of the four locations (Fig. 2).SARE 1-year Fig 2 The Columbia site produced less forage and it is suspected that this is likely due to the inherent lower moisture at this location. Additionally, lower surface soil wettability might play a factor at this location. We will follow-up by testing surface soil infiltration at all locations. It is interesting to note that Columbia and Gulf share the same soil series (Blanton sand), so there may be other factors at play, as well. Plots not receiving phosphorus (P) fertilizer performed as well as the complete fertilizer treatment for at least half of the locations and in fact, there is a trend of improved yields at Osceola, which also has among the lowest soil P fertility values. Forage tissue concentrations and soil fertility data will be shared in the final report. As one would expect, removal of N from the fertilizer depressed production, but this was also the case for potassium (K), where yields were dropped at most locations (Fig. 2).

Expanding our SSNM demonstration from Osceola County to an additional three counties across Florida, we provided additional opportunities for getting the word out that long-term fertility trials were taking place on-farm. Pasture decline is typically, a chronic and slowly progressing situation that can often go unnoticed until something(s) occurs that triggers a system crash. When it comes to poor soil fertility (or nutrient imbalances), expression in forages may not be obvious, particularly when there is no positive control (an area without any soil nutrient limitations) to compare against. These past two years (2020 and 2021) provided initial visual and compositional information of three major essential nutrients, N, P, and K regarding their deletion from normal soil fertility management.

Of the four trial sites reported here, three trials (Orange, Columbia, and Gulf counties) were initiated in 2019, while the fourth (Osceola County) was initiated in 2016. To better compare across locations, the results are presented as years 1, 2, …n, rather than dates, to better compare Osceola County results at the same treatment age as the others, even though growing conditions (weather/climate) can change from year to year. Annual bahiagrass production interacted with year, location, and treatment, as might be expected. However, by the end of Year 3, plots receiving complete fertilizer as either mineral or as class AA biosolids performed similarly and plots that did not receive P fertilizer performed equally well. In comparison, plots that did not receive K fertilizer had at least 20% less production, while plots that had not received N fertilizer had at least 40% less forage. In comparison, plots that were not fertilized (check plots), lost at least 50% production.  Table 1 provides production values by location and year after initiation. Identifying similar temporal responses under different soil types and conditions helps with educating producers on managing their soil fertility and the potential impacts they might encounter. It should be noted that Gulf County, with a somewhat heavier soil type was also associated with greater annual bahiagrass production than the sandier soil sites (all other sites), while the remaining three were similar, even though the Orange County site had relatively high initial soil fertility (Table 2) due to a previous, multi-year receivership of Class B biosolids that ended several years ago. The Osceola site can be used as a model regarding the potential timing of impacts through lack of balanced fertilization longer term. For example, eliminating P fertilizer had no impact on yield over three years at any of the on-farm sites, but even after six years, forage production remained comparable to the complete mineral fertilization management at Osceola County.

Soil fertility is often the preferred method for determining bahiagrass nutritional requirements and fertilizer needs. Initial soil fertility varied among on-farm sites, where P ranged from 19 to 817 mg kg-1 and K ranged from 13 to 22 mg kg-1 (Table 3). As with many southeastern U.S. states, Florida does not recommend testing for available soil N. Based upon initial soil fertility, Osceola County had an initially low soil test P and all counties had low initial soil test K. Regarding soil test P, there was a time x location interaction, when after three years, soil P fertility remained statistically unchanged at all but the Osceola site, which declined from 19 to 12 mg P kg-1. However, there was no treatment impact on soil P fertility within the first three years. For soil K, there was no location by treatment interactions. Across locations, the minus K treatment had similarly low soil K to the Check plots, while minus N and minus P treatments had greater soil K over time (Table 4). It was expected that K fertility would be greater under the minus N treatment, since there was less forage to pull K out of the soil, thereby allowing it to accumulate. The complete, biosolids and minus P treatments had similarly high soil K fertility compared to the depleted treatment but relative to IFAS recommendations soil K fertility remained low (< 35 mg kg-1) in all but the minus P and minus N treatments. In Florida pastures, it may be economically challenging to meet the IFAS medium soil K rating.

Soils provide the nutrient sources to the plant. However, soil fertility values may not always reflect the plant nutritional status. Since plant tissue nutrient concentrations can shift during the season, mid-summer (during active growth) tissue sampling may better reflect plant nutritional status. For this discussion, only data following the first of the seasonal harvests were further compared over three years of sampling for tissue content. Whole (above-ground) plants were sampled, rather than the most fully expanded leaf tissue to simplify the process and provide data for a nutrient removal budget. Across locations and time, the minus N treatment had among the lowest tissue N values but remained near or slightly above 1% N (~6.25% crude protein). In this case, crude supplementation would be needed to support beef cattle grazing. Bahiagrass has been noted to have developed associative N2 fixation and this may have helped maintain the plant N status even while production declined. Interestingly, the minus K treatment maintained greater tissue N than any of the other treatments (Table 5). Even so, tissue N values never exceeded 2% (~ 12.5% crude protein).

Across locations and time, the minus P plots had significantly lower tissue P than all other treatments, even the check plots that received no fertilizer (Table 6). The recommended critical low tissue P for bahiagrass was set at 0.15% for Florida. By Year 6, bahiagrass at the Osceola County location would sometimes test below 0.15% but not consistently and yields did not seem negatively impacted, even at this low content. As with N, the minus K plots had among the highest tissue P values and Class AA biosolids had equally high tissue P. Biosolids are a good source of slow-release, plant-available P and these data reflect that. Additionally, under biosolids management, soil P fertility was comparable to other treatments, as mentioned previously. The soil P test did not correlate strongly with tissue P and further calibration between tissue and soil test P may benefit Florida producers.

Tissue K values were excessively low (0.56%) in the minus K plots, even when averaged across locations and years (Table 7). None of the treatments resulted in high tissue K values. Combined with the relatively low soil K fertility, applying K as one-time application per season should be reconsidered. In this study, K fertilization was split-applied (half in spring and half after first harvest). In sandy Florida soils, split K applications should be considered (apply with N fertilizer), whenever soil test results show low soil K.

The budgeting of fertilizer nutrients or amount removed by the production system via the forage biomass, provides additional insight on potential demand and losses to the environment (Table 8). For example, the complete mineral fertilizer, biosolids and minus P treatments had similar amounts of N removed through their biomass. In comparison, the minus K treatment had significantly less N removed from the soil compared to the complete mineral fertilizer of the minus P treatment, due to lower production rates. It was most pronounced at the Columbia County farm, where nearly one third less N was taken up. With high rainfall events noted by many surveyed ranchers, there is an increased risk of nitrate leaching loss from the unused fertilizer. It is interesting to note that N removal from the Orange County site was similar for all fertilized treatments. This is likely due to the high inherent fertility (past Class B biosolids applications) prior to the start of this study. The check treatment (no fertilizer applied) response appeared to be reflective of initial soil properties, where the finer textured soils (Gulf County) accumulated 61 kg ha-1), compared to Columbia County at 25 kg ha-1).

As mentioned previously, the three-year comparison of eliminating P from the fertilizer had little to no impact on production but it had impacted the amount of P that was removed from the land. The greatest impact was at Gulf County, with a 36% decline in removal by the bahiagrass. In comparison, there was no loss in P uptake at the Orange County farm since inherent soil P concentrations were high from the start of the study. Based on an average removal rate of 20 kg P ha-1 (or 46 kg P2O5 equivalents) per annum, it may take decades of hay removal before soil P concentrations would become deficient. If continued to be managed as pasture, it might never become deficient. Meanwhile, the sandy Florida soils are prone to nutrient leaching, including P.

Eliminating K fertilizer had a large impact on forage production within a couple of years of deleting it from the fertilization management. Within three years, K removal dropped 60 to almost 80%, depending upon the location. Removing P from the fertilizer did not limit K uptake or removal, while N depleted systems removed only about one half the amount they could have if they received ample N fertilization (Table 8). Low N limited forage production but there was little sign of stand thinning. Soil P deficiency may often take many years to reach under most Florida pastures, due to P from excreta. In comparison, eliminating K fertilization for a short time (one or two years) while continuing to apply moderate to high rates of N fertilizer, impacts can lower forage production directly. However, in the case of Argentine bahiagrass, Take-All root rot and various foliar fungal diseases, such as Bipolaris, can hasten a stand’s demise (Figure 3) (Figure 4). In this case, fertilizer imbalance (fertilizer N:K too high) can lead to a train wreck and it might be one contributor to pasture decline. In situations where fertilizer prices are limiting a farmer’s fertilization options and production can be put on hold, consider prioritizing the most productive fields for a balanced fertilizer application (based upon soil fertility report) and greatly limit or avoid N applications to other fields, particularly if other major soil nutrients are deficient and are not affordable. As demonstrated by the data, bahiagrass that does not receive N fertilizer will tend to maintain adequate nutrient balance over multiple years. Additionally, tissue analysis through the collection of above-ground clippings during the active growing season can provide greater insight into potential soil fertility deficiencies. This tool should be used in combination with a soil report rather than as a substitute.

We will continue with a longer-term assessment of these treatments. At the Osceola farm, the lack of K applications had decimated plots by Year 6. In 2022, those treatment plots were split to determine if returning to normal K fertilization practices will result in stand recovery. Additionally, more time is required to assess when soil P deficiencies may begin to manifest in bahiagrass plots. The impact of chronic, fertilization imbalances and interactions with the environment likely affect other soil properties, including soil carbon stocks. These sites may help us elucidate the potentially larger impacts poor fertilization management can have on the larger, agroecological system.

Participation Summary
4 Farmers participating in research

Educational & Outreach Activities

62 Consultations
5 Curricula, factsheets or educational tools
7 On-farm demonstrations
4 Published press articles, newsletters
4 Tours
6 Webinars / talks / presentations
2 Workshop field days
1 Other educational activities: Created producer survey on pasture decline

Participation Summary:

40 Farmers participated
6 Ag professionals participated
Education/outreach description:

With the advent of COVID-19, all in-person activities were prohibited in 2020. Three pasture health articles were published (1 as an electronic newsletter and two in the Florida Cattlemen Magazine. Jung-Chen Liu (MS student, Mackowiak advisor) presented a poster at ASA-CSSA-SSSA on bahiagrass subsoil fertility when grown as a monoculture or as  a mixture with rhizoma peanut (Arachis glabrata), under low fertilizer management. In this case, the ability of rhizoma peanut (legume) to fix N2 provided N fertility to the bahiagrass system and increased surface soil reactive N content. Her poster won 2nd place in Division C-06, Forage and Grazinglands. Additionally, two video trainings on bahiagrass fertilization and management were published on YouTube and can be accessed from the UF-IFAS, Panhandle Ag Enews website. Together, these two videos have received over 10,000 views.

In 2021, we spent much of our effort collecting data, touring producers and presenting findings at events, such as Moultrie Ag Expo and more locally, and at a field day at Osceola county. Time was spent reviewing results with the host producer at each location. The photodocumentation provided many opportunities to show what can happen when fertilization falls out of balance. This also motivated producers to visit their nearest on-farm site to see for themselves. Since the Osceola County site was initiated prior to this sponsorship, the K limited plots were nearly decimated and provided a stark visual impact to those who visited that location. We expect the same impact for the other sites in the coming year or two.

Learning Outcomes

8 Farmers reported changes in knowledge, attitudes, skills and/or awareness as a result of their participation

Project Outcomes

5 Farmers changed or adopted a practice
2 New working collaborations
Project outcomes:

The clear responses in bahiagrass performance due to improved fertilizer management has gained much attention by extension agents and producers. This is exemplified through extension outreach efforts to promote adequate potassium fertilization in pastures. The project's extension team has had discussions with at least 60 clientele describing the early results of this on-farm research, particularly low K fertility impacts. In contrast, the soil P was depleting slowly, which demonstrates that over-applying P (too frequently or at higher rates than necessary) can be economically and environmentally wasteful. We were observing other unexpected benefits by allowing soil P fertility to decline, namely the impact it had on some weed species. The Osceola County location demonstrated this effect well and by the end of 2021, similar results were noticeable at the Gulf County site. If the data trends continue, we expect to demonstrate that by managing fertilization, ranchers can not only increase yields by over 30% and save on their fertilizer bills, they may also save on herbicides. Since bahiagrass covers much of the southeastern coastal plain, savings on fertilizer and herbicide costs, along with  greater pasture health, will benefit these state economies. States with healthy bahiagrass stands will not likely find a more economical production system for supporting livestock. Demonstrating the successful production goals under low soil P fertility also is attractive for use near sensitive waterbodies and for low income livestock producers across the Southeastern Coastal Plain.


Long-term studies are exceedingly difficult to sponsor. I appreciate the assistance SARE has provided. Perhaps SARE may consider applying a portion of their funding to support a few longer-term efforts. Projects such as this one that do not require large sums of money but do require recurring funds over 5 to 10 years, often uncover interesting findings that are masked during shorter duration efforts. For example, our observations of weed suppression under low P fertility, while still maintaining healthy bahiagrass stands is one such example.  

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.