Scheduling Fertilizer Application and Rate to Optimize Plant Growth and Minimize Nitrogen Leaching

Return to: MREC Home Page

Return to: MREC Research Index


Charles A. Conover, Ph.D.and Loretta N. Satterthwaite*

University of Florida
Central Florida Research and Education Center
Institute of Food and Agricultural Sciences
CFREC-Apopka Research Report RH-96-7

By scheduling fertilization of plants during the growing cycle to more optimally fit plant growth needs rather than applying fertilizer at a steady rate, growers may be able to minimize nitrogen leaching and the potential for surficial aquifer and ground water contamination. Growth of plants can be related to a modified sigmoid curve. If fertilizer were applied in a similarly modified sigmoid fashion, plant growth would likely be optimized and nitrogen leaching from containers possibly reduced. To test that hypothesis, the following experiments were performed.

Materials and Methods

Experiment 1: Croton 'Petra' [Codiaeum variegatum (L.) Blume 'Petra'] liners were potted, one plant per pot, into 15-cm (6-inch) pots using Fafard Mix T#4 on May 28, 1993. Plants were placed in a glass greenhouse where maximum light level attained was 380 mol•sl•m-2 (2000 ft-c) and set points for heating and cooling were 18C (65F) and 35C (95F), respectively. Plants were fertilized using 24N-3.5P-13.3K (Peters 24-8-16) at a total rate of 9.1 g per 15-cm pot over a 26-week period (equal to 273 g N • m-2 • yr-1 or 2400 lb N/A/yr). Each pot was an experimental unit and there were 10 fertilizer treatments with 5 replications per treatment. On day 10 (week 1) after potting, all plants received their first fertilizer. Initial charge of fertilizer was either 3.8% (0.35 g; control), 5%(0.46 g), 10% (0.91 g) or 15% (1.36 g) of the total fertilizer to be applied during the experiment duration. (Schedule of fertilizer application and percent applied to-date on file at CFREC-Apopka.) Thereafter, the control treatment pots received a weekly fertilizer application equal to the initial charge (0.35 g). Other treatments consisted of delaying further fertilizer applications until week 3, week 5 or week 7. On June 21 (week 3), one-third of the pots initially fertilized at 5 % received 0.36 g fertilizer and continued receiving that amount weekly thereafter; one-third of the pots initially fertilized at 10% received 0.37 g fertilizer and then received 0.34 g weekly thereafter; and one-third of the pots initially fertilized at 15% received 0.38 g fertilizer and then received 0.32 g weekly thereafter. On July 5 (week 5), one-half of the remaining pots that had received an initial charge of 5% received 0.45 g fertilizer, followed by 0.39 g weekly thereafter; one-half of the remaining pots that had received an initial charge of 10% received 0.42 g fertilizer, followed by 0.37 g weekly thereafter; and one-half of the remaining pots that had received an initial charge of 15 % received 0.39 g fertilizer, followed by 0.35 g weekly thereafter. On July 19 (week 7), the remaining pots from each initial charge of 5%, 10% and 15%, received 0.47 g followed by 0.43 g weekly thereafter, 0.4 g followed by 0.41 g weekly thereafter or 0.33 g followed by 0.39 g weekly thereafter, respectively. The first fertilization after the initial charge was chosen to receive any excess amount of fertilizer that could not be divided for equal application during the remainder of the experiment simply because the plants had not received fertilizer for a time and we wanted all the plants to have received an equal amount of fertilizer at experiment end.

At transplanting, all "watering in" leachate was collected and poured back through the medium until each pot was wet; remaining leachate from each pot was acidified, refrigerated, then combined with all leachate from irrigation before fertilization (sample #1) and processed in the manner described below. All leachate was collected from each pot at each irrigation during a two-week period. All leachate except that from the last irrigation of the two-week period was acidified to a pH of about 2.0, to prevent microbial action that could change the characteristics of the sample, and then refrigerated at 4C. Leachate collected from the last irrigation of a two-week period was measured for pH and electrical conductivity, then combined with refrigerated leachate from the same pot; total volume was recorded and a 20-ml sample was prepared, acidified to 2.0 pH, and sent (14 sample times in all) to the lab (Univ. of Florida, IFAS Analytical Research Lab, Gainesville, FL 32611-0740) for analysis for phosphorus, NH4-N and NOx-N.

In addition to leachate characteristics, initial height was recorded on May 28 and a plant grade, based on a scale of 1 = dead; 2 = poor quality, unsalable; 3 = fair quality, salable; 4 = good quality; and 5 = excellent quality, and final height were recorded on December 3 and December 7, respectively.

Experiment 2. On October 29, 1993, Spathiphyllum Schott. 'Petite' (petite peace lily) rooted liners were transplanted into 15-cm (6-inch) pots containing Fafard #4 growing medium. Plants were placed in a glass greenhouse where maximum light level attained was 380 mol•m-2•s-1 (2000 ft-c) and set points for heating and cooling were 18C (65F) and 35C (95F), respectively. As in the previous experiment, each pot was an experimental unit and there were 10 fertilizer treatments with 5 replications per treatment. However, the recommended fertilizer rate for spathiphyllum is lower than that for croton (Conover and Poole, 1990), so plants were fertilized using 24N-3.5P-13.3K (Peters 24-8-16) at a total rate of 6.5 g per 15-cm pot over a 26-week period (equal to 175 g N • m-2 • yr-1 or 1536 lb N/A/yr). On day 10 (week 1) after potting, all plants received their first fertilizer. Initial charge of fertilizer was either 3.8% (0.25 g; control), 5% (0.33 g), 10% (0.65 g) or 15% (0.98 g) of the total fertilizer to be applied during the experiment duration. (Schedule of fertilizer application and percent applied to-date on file at CFREC-Apopka.) Thereafter, the control treatment pots received a weekly fertilizer application equal to the initial charge (0.25 g). Other treatments consisted of delaying further fertilizer applications until week 3, week 5 or week 7. On November 22 (week 3), one-third of the pots initially fertilized at 5% received 0.42 g fertilizer and then received 0.25 g weekly thereafter; one-third of the pots initially fertilized at 10% received 0.33 g fertilizer and then received 0.24 g weekly thereafter; and one-third of the pots initially fertilized at 15% received 0.23 g fertilizer and continued receiving that amount weekly thereafter. On December 6 (week 5), one-half of the remaining pots that had received an initial charge of 5% received 0.29 g fertilizer, followed by 0.28 g weekly thereafter; one-half of the remaining pots that had received an initial charge of 10% received 0.39 g fertilizer, followed by 0.26 g weekly thereafter; and one-half of the remaining pots that had received an initial charge of 15% received 0.27 g fertilizer, followed by 0.25 g weekly thereafter. On December 20 (week 7), the remaining pots from each initial charge of 5%, 10% and 15%, received 0.47 g followed by 0.30 g weekly thereafter, 0.34 g followed by 0.29 g weekly thereafter or 0.39 g followed by 0.27 g weekly thereafter, respectively. As in experiment 1, the first fertilization after the initial charge was chosen to receive any excess amount of fertilizer that could not be divided for equal application during the remainder of the experiment.

Plants were "watered in" and leachate was collected throughout the experiment as described for the previous experiment. Leachate was analyzed for pH, electrical conductivity, phosphorus, NH4-N and NOx-N. In addition, other data collected included: initial and final plant heights (October 29, 1993 and May 2, 1994, respectively); 3-month and 6-month (final) plant grades (February 8 and May 5, 1994, respectively); number of new, deformed leaves periodically throughout the experiment (February 9 and 22, and May 2, 1994); and number of flowers on each pot at experiment end on May 5, 1994.

Data for both experiments were statistically analyzed with Analysis of Variance (PROC ANOVA) or General Linear Models (PROC GLM) procedures using Statistical Analysis System (SAS Institute, Inc., Cary, NC 27512-8000) software. Conversion of pH to hydrogen ion concentration was performed before statistical analysis, but results are discussed as pH.

Results and Discussion

Experiment 1: All plant grades were excellent and there were no differences in grade due to treatment. Change in height from experiment initiation to experiment termination was not different due to delay of fertilizer application after initial charge; however, there were slight differences in the change in height of the plants due to the rate of the initial fertilizer application, with the 5% initial charge of fertilizer yielding the greatest increase in height Cable 1-1). Though statistically significant, this difference in height would not be commercially meaningful. There was no difference due to treatment for the total mg of phosphorus leached for the duration of the experiment. However, both total mg NH4-N and NOx-N leached did show treatment differences (Table 1-2). Changes in NH4-N leachate were slight and probably would not have a significant effect on ground water. However, NOx-N leachate was 23% less for the 5% precharge as compared to the standard treatment (3.8 % precharge and no delay). Increasing the precharge to 10% increased the NOx-N leachate to a level 34% over the standard treatment and 73 % over the 5 % precharge treatment. These data indicate that the precharge level is critical and that it has a strong influence on the potential for NOx-N leachate to enter the ground water.

The percent of applied N that leached from pots was different for those pots receiving a 5 % starter charge of fertilizer and fertilized on a two-week delay schedule than for those receiving a 10% starter charge with further application delayed for four or six weeks and those pots receiving a 15% starter charge and no further fertilizer applications for two weeks (Figure 1). Although percent of applied N that leached was lowest from pots receiving the 5% starter charge, the amount was not statistically different from the constant feed treatment.

Experiment 2. Height change, number of deformed leaves and number of flowers of Spathiphyllum 'Petite' were unaffected by either the percent of fertilizer applied initially or by delaying future applications of fertilizer after the initial charge of fertilizer (Table 2-1). However, plant grade after six months was significantly improved by using a higher initial charge of fertilizer and by delaying future applications of fertilizer after the initial charge.

Total mg of NOx-N in leachate was affected by initial fertilizer application and continued fertilizer delay schedule (Table 2-2), and using "constant feed" as a control (common industry practice), 38% less NOx-N was leached throughout the experiment by using an initial charge of 5%, which also reduced the total amount of NH4-N and phosphorus leached. By delaying further application of fertilizer after the initial charge to the newly planted pots for six weeks, there was also a 31% reduction in the amount of NOx-N leached as compared to the 'industry standard', but a delayed fertilizer application schedule had no effect on total NH4-N or phosphorus leached.

All fertilizer treatments had an effect on pH of the leachate collected. With the exceptions of sample 5, when there was no significant effect due to treatment, and sample 13, when only the to-date percent of applied fertilizer had an effect, pH was always affected by both the to-date percent of applied fertilizer and the delay after initial fertilization before continuing fertilizer application, with no interaction between the two factors (Table 2-3). Additionally, until such time when all pots were being fertilized on a weekly basis, all leachate characteristics showed differences due to percent of fertilizer applied and delay in further application of fertilizer after the initial application.

Conclusions

Research reported in this article indicated that the initial hypothesis-to more optimally fit fertilizer application with plant growth could result in less leaching of NOx-N-was proven to be correct. In both experiments, less NOx-N was leached, with the amounts being 23 or 38%. Although these data indicate that fertilization procedures could be altered to reduce NOx-N leaching, the potential will not be achieved unless fertilization programs and irrigation systems, as typically used in the industry, are redesigned.


*Professor Emeritus and Statistician, University of Florida, IFAS, Central Florida Research and Education Center, 2807 Binion Road, Apopka, FL 32703-8504, respectively.


Literature Cited

  1. Conover, C.A. and R.T. Poole. 1990. Light and fertilizer recommendations for production of acclimatized potted foliage plants. Nursery Digest 24(10):34-36, 58-59.

Figure 1. Percent of N applied (2184 mg) that leached from pots containing Codiaeum variegatum 'Petra' that was fertilized using various starter charges and no delay or 2, 4 or 6 weeks delay before further applications of fertilizer.

Click image for larger view.


  1. Table 1-1. Effects of "sigmoid" application of fertilizer over a 26-week period from May 28, through December 7, 1993 on growth characteristics of Codiaeum variegatum 'Petra'.
Initial fertilizer (% of total applied) Final plant gradez Height changey
3.8 4.9ax 29.2 ab
5.0 5.0a 30.3 a
10.0 5.0a 27.0 b
15.0 5.0a 28.5ab
Weeks delay after initial fertilization before continuing weekly fertilizer application
0 4.9a 29.2a
2 5.0a 29.6a
4 5.0a 28.0a
6 5.0a 28.2a
  1. Zplant grade recorded on December 3, 1993, based on a scale of 1 = dead; 2 = poor quality, unsalable; 3 = fair quality, salable; 4 = good quality; and 5 = excellent quality.
    YHeight change equals final height of plant measured on December 7, 1993 minus initial height measured on May 28, 1993.
    XMean separation within columns by Duncan's multiple range test, 5% level. Means in same columns followed by the same letter are not significantly different at the 5% level.

  1. Table 1-2. Effect of "sigmoid" application of fertilizer (over a 26-week period from May 28, through December 7, 1993) to characteristics of leachate collected from the medium being used to grow Codiaeum variegatum 'Petra'.
  Total leached (mg/pot)
Initial fertilizer (% of total applied) NOX-N NH4-N Phosphorus
3.8 170.2abZ 3.0a 508.4a
5.0 131.9a 5.1a 537.9a
10.0 228.8b 5.2a 508.1a
15.0 214.9b 5.0a 528.4a
Weeks delay after initial fertilization before continuing fertilizer application
0 170.2a 3.0a 508.4a
2 185.6a 4.5ab 527.2a
4 185.1a 4.7ab 511.5a
6 204.9a 6.2b 535.7a
  1. ZMean separation within columns by Duncan's multiple range test, 5% level. Means in same columns followed by the same letter are not significantly different at the 5 % level.

  1. Table 2-1. Effects of "sigmoid" application of fertilizer (over a 26-week period from October 29, 1993 through May 2, 1994) on growth characteristics of Spathiphyllum 'Petite'.
Initial fertilizer (% of total applied) Final plant gradeZ Height changeY Deformed leavesX FlowersW
3.8 4.4bv 10.2a 2.8a 2.6a
5.0 4.7a 7.8a 2.1a 1.9a
10.0 4.8a 10.5a 1.9a 1.8a
15.0 4.7a 8.6a 1.8a 1.9a
Weeks delay after initial fertilization before continuing fertilizer application
0 4.4b 10.2a 2.8a 2.6a
2 4.7a 8.3a 1.9a 1.8a
4 4.7a 9.9a 2.2a 1.9a
6 4.7a 8.6a 1.7a 1.9a
  1. ZPlant grade recorded on May 5, 1994, based on a scale of 1 = dead; 2 = poor quality, unsalable; 3 = fair quality, salable; 4 = good quality; and 5 = excellent quality.
    YHeight change equals final height of plant measured on May 2, 1994 minus initial height measured on October 29, 1993.
    XDeformed leaves refers to the total number of new, deformed leaves recorded on February 9, 1994, February 22, 1994 and May 2, 1994.
    WFlowers refers to the number of flowers on each pot on May 2, 1994.
    VMean separation within columns by Duncan's multiple range test, 5% level. Means in same columns followed by the same letter are not significantly different at the 5% level.

  1. Table 2-2. Effect of "sigmoid" application of fertilizer (over a 26-week period from October 29, 1993 through May 2, 1994) to characteristics of leachate collected from the medium being used to grow Spathiphyllum 'Petite'.
  Total leached (mg/pot)
Initial fertilizer (% of total applied) NOX-N NH4-N Phosphorus
3.8 197bZ 6.7a 49ab
5.0 123a 6.4a 41a
10.0 175ab 7.6a 47ab
15.0 208b 10.0b 52b
Weeks delay after initial fertilization before continuing fertilizer application
0 197b 6.7a 49a
2 209b 8.4a 51a
4 161ab 7.9a 45a
6 136a 7.7a 44a
  1. ZMean separation within columns by Duncan's multiple range test, 5% level. Means in same columns followed by the same letter are not significantly different at the 5% level.

  1. Table 2-3. Significance of effect of "sigmoid" application of fertilizer (over a 26-week period from October 29, 1993 through May 2, 1994) on leachate of medium used to grow Spathiphyllum 'Petite'.
  Leachate characteristic analyzed
Sample # Phosphorus NH4-N NOx-N EC pH
1z NSy for all
2x All different due to % and D
3w % % and D % and D % and D % and D
4v % and D % and D % and D % and D % and D
5u % and D NS % and D % and D NS
6t % and D D % and D % and D % and D
7s % % NS NS % and D
8 % NS NS NS % and D
9 NS NS NS NS % and D
10 NS NS NS NS % and D
11 NS NS NS NS % and D
12 NS NS NS NS % and D
13 NS NS NS NS %
14 NS D NS NS % and D
  1. zLeachate collected before any fertilizer applications made.
    yNS, % and D indicate not significant, significant due to percent of fertilizer applied to-date and significant due to delay of fertilizer application after initial fertilization, respectively.
    xAll pots have received initial fertilizer charge of 3.8% ("constant feed"), 5%, 10% or 15% of total fertilizer to be applied during 26-week period.
    wPots receiving "constant feed" treatment are now being fertilized weekly; no other pots have received any fertilizer applications after initial charge.
    vPots on constant feed and 2-week delay treatments are now being fertilized weekly.
    uPots on constant feed, 2-week delay and 4-week delay are now being fertilized weekly.
    tSix-week delay pots have now started receiving weekly fertilizer applications.
    sAll pots are now being fertilized weekly, but percent of fertilizer applied to each pot out of total fertilizer to be applied still varies by treatment.