Interpreting Nitrate Concentration in Tile Drainage Water

Purdue Extension publication AY-318-W

Download a printable PDF of this publication

Authors

Sylvie Brouder, Brenda Hofmann, Eileen Kladivko, Ron Turco, Andrea Bongen, Purdue University Department of Agronomy; Jane Frankenberger, Purdue University Department of Agricultural and Biological Engineering

Figure 1. Average monthly distribution of rainfall, drainflow, and N load at the (a) Water Quality Field Station (WQFS) and (b) Southeast Purdue Agricultural Center (SEPAC).

Figure 1. Average monthly distribution of rainfall, drainflow, and N load at the (a) Water Quality Field Station (WQFS) and (b) Southeast Purdue Agricultural Center (SEPAC).

 

Introduction

The combined objectives of maximum corn yields, efficient use of inputs (including fertilizer), and improved water quality in agricultural watersheds has led to both agricultural and regulatory interest in measuring nitrate nitrogen (NO3-N) concentrations in field drainage tile outlets. Crop, soil, and fertilizer management practices can have a strong influence on the amount of nitrogen (N) that is leached through soil and susceptible to loss in tile drainflow. However, other factors beyond a farmer’s control can also strongly influence NO3-N leaching and these need to be considered when attempting to evaluate the success of N management practices based on NO3– N concentrations measured in tile drainflow.

The purpose of this publication is to (1) provide some general guidelines for interpreting measured NO3-N concentrations in drainflow samples collected through monitoring programs, and (2) describe the most important factors influencing drainflow NO3-N concentrations. It is important to understand that some N loss is an inevitable outcome of improving the drainage and thus the agricultural productivity of naturally poorly drained soils. Nitrate is mobile in soils and is carried by water moving through soils. Subsurface drainage tiles create a direct pathway to carry NO3-N from the bottom of the root zone to the adjacent surface water. Whenever tile lines flow, NO3-N is lost from the field.

In Indiana, tile lines typically flow intermittently from late November to June. During this time, temperatures are cool and plants are either not present or are small and, as a result, combined water losses through evaporation and transpiration are minimal. In northern Indiana, tiles may cease flowing over long periods during the winter, causing the majority of the flow to occur between April and June. A six-year study at the Purdue University Water Quality Field Station (WQFS) in Tippecanoe County, for example, found that greater than 50 percent of the SOILS (TILLAGE) Agronomy Guide annual tile flow occurred in May and June (see Figure 1a). In contrast, in much of central and southern Indiana, tile lines flow all winter, and up to three-fourths of the annual drainflow and nitrate losses occur during the November to April period before field work and fertilization for the next crop begins. Results from a 15-year study at the Southeast Purdue Agricultural Center (SEPAC) in Jennings County illustrate this point (see Figure 1b) and underscore the potential importance of growing winter cover crops as “trap crops” for the N in the soil. The timing of drainflow during the year has important consequences for potential management of nitrate leaching to tiles.

General Guidelines for Interpretation of NO3-N Concentrations in Tile Drains

Table 1. General guidelines for interpreting NO3-N concentrations in tile drainage water. The interpretation is derived from numerous studies conducted throughout the cornbelt and highlights land management strategies commonly found in association with a concentration measured in tile as the tile leaves the edge of the field. † Economically optimum N rate is the rate that maximizes the return on investment in N fertilizer and therefore may be slightly lower than the N rate that maximizes crop yield.

Table 1. General guidelines for interpreting NO3-N concentrations in tile drainage water. The interpretation is derived from numerous studies conducted throughout the cornbelt and highlights land management strategies commonly found in association with a concentration measured in tile as the tile leaves the edge of the field. † Economically optimum N rate is the rate that maximizes the return on investment in N fertilizer and therefore may be slightly lower than the N rate that maximizes crop yield.

Because of the importance of both tile drainage and nitrogen fertilizer to corn production in the Eastern Cornbelt, NO3-N concentrations in tile drainage water have been studied extensively by university researchers. Collectively, these studies suggest some common guidelines for interpreting the meaning of an individual concentration result from a water testing program. However, caution should always be exercised in drawing conclusions from an individual sample, because concentrations may fluctuate over short time periods without a specific known cause.

When corn productivity is not restricted by environmental conditions (e.g., saturated soils during stand establishment or drought during pollination), the optimal N fertilizer rate will usually result in NO3-N concentrations of <20 ppm in tile drainage water the following fall/spring (see Table 1). It should be noted that growing a row crop such as soybeans that fixes its own N from the atmosphere, does not drastically reduce NO3-N concentrations in drainage water. Crop residue decomposes after harvest, releasing a portion of the fixed N and making it available for use by a subsequent crop or for loss through leaching. Studies have consistently shown NO3-N values in spring drainage water following a soybean production year to be comparable to, or only slightly less than, values observed following an N-efficient corn crop.

When a lack of N fertilizer limits yields in an otherwise optimal growing season, the NO3-N concentrations in tile drainflow the following spring may be less than 10 ppm. Cover crops may also reduce tile NO3-N concentrations if the crops grow sufficiently to trap significant quantities of residual N, then decompose at an appropriate time to release the trapped N to a growing crop. In the SEPAC study on a low organic matter soil, for example, nitrate-N concentrations averaged 8 ppm in a no-till corn soybean rotation with a winter wheat cover crop grown after the corn. However, truly low drainflow NO3-N concentrations only consistently occur under grasslands, pastures receiving minimal inorganic N fertilizer, or land that has been taken out of production and put in the Conservation Reserve Program (CRP).

When spring NO3-N concentrations exceed 20 ppm, improper N management may be the cause. However, over-application of N is not the only factor that can result in high NO3-N concentration in drainage water. Therefore, it is important to understand the underlying cause of the water quality problem before implementing changes in N management strategies.

Factors Resulting in High (>20 ppm) NO3-N Concentrations in Drainage Water

Management Factors

N rate

Table 2. Impact of N rate and rotation on NO3-N concentrations in tile drainage water measured the following spring. Yields were not responsive to N at rates higher than 150 lbs. per acre.

Table 2. Impact of N rate and rotation on NO3-N concentrations in tile drainage water measured the following spring. Yields were not responsive to N at rates higher than 150 lbs. per acre.

When the amount of N fertilizer applied exceeds the needs of the cropping system, the residual N remains in the soil after harvest. This N is already in the form of NO3-N and is readily lost when late fall or early spring rainfalls cause tile drains to flow. Studies conducted in Wisconsin have shown that N fertilizer rates that were 45 lbs. per acre more than the economic optimum consistently resulted in soil water NO3-N concentrations >20 ppm. Results from Purdue University’s WQFS also illustrate this concept. In 1997, yields for continuous corn were maximized by fertilizer N rates of 150 lbs. per acre (see Table 2). N rates of 170 and 190 lbs. per acre did not significantly increase yields. However, the following spring, NO3-N concentrations in tile drainage water from plots receiving 150 lbs. N per acre averaged 14.5 ppm while those from 170 and 190 lbs. N per acre averaged 30.5 and 50.5 ppm, respectively.

Improper Crediting of N from Organic Sources

Figure 3. Nitrate concentrations in 2000 in drainage water from a continuous corn crop fertilized on May 5, 2000, with 160 lbs. preplant N per acre. The different color dots represent different replicates of the same treatments at WQFS.

Figure 3. Nitrate concentrations in 2000 in drainage water from a continuous corn crop fertilized on May 5, 2000, with 160 lbs. preplant N per acre. The different color dots represent different replicates of the same treatments at WQFS.

Many of the recommendations for N fertilizer management developed by land-grant universities, including the Tri-State Fertilizer Recommendations developed for Michigan, Indiana, and Ohio, assume continuous corn production as the baseline cropping system. They also assume that the only additional N added to the system will be in the form of inorganic N fertilizer. However, many farmers use production systems that add significant amounts of N to the system as organic N. Common organic N sources that have significant replacement value for inorganic fertilizer N include manure and soybean residue. Table 2 shows the impact of failing to take credit for the N in soybean residue on water quality at the WQFS. When the amount of N required to optimize yields of continuous corn (150 lbs. per acre) is applied to corn grown in rotation with soybean, average NO3-N concentrations in tile drainage water increase from 14.5 ppm to 38.5 ppm.

Timing of N Application

The longer the period between the time of N fertilizer application and the time of crop N demand, the greater the opportunity for N to be leached through the soil and lost in tile drains. An example of what can happen during a rainy period after application of nitrate-containing fertilizers is shown in Figure 3. Preplant N fertilizer (28 percent ureaammonium nitrate) was applied at the recommended rate on May 5, 2000, 19 days before planting. In this year, 5.25 and 8.1 inches of rain fell in May and June, respectively, 1 and 4.2 inches above the 30-year local average for these months. Some of the applied N immediately appeared in the drainage water, increasing NO3-N concentrations from below 10 ppm to values that ranged between 10 and 35 ppm. Had the N application been delayed until sidedressing, less N would have been leached and concentrations would have remained lower during May and into June. Had the N application occurred earlier (e.g., the first week of April), NO3-N concentrations in tile drainage water would have been elevated throughout April as well.

Form of N Used

In Figure 3, the N fertilizer used was 28 percent urea-ammonium nitrate and therefore contained a portion of the N already in the leachable NO3-N form. Had a different N source such as anhydrous been applied, the impact of N fertilization on water quality would have been delayed and therefore reduced. This is because the application of anhydrous kills microbes in the zone of the band application. The conversion of NH4-N to NO3-N is delayed until the band is recolonized and biological activity can proceed. Leaching rainfalls occurring during this period of no or reduced biological activity will not result in much fertilizer N loss. In the SEPAC study, for example, where anhydrous was applied in late April throughout the study, there was no consistent increase in concentrations in drainflow after the fertilizer application (see Figure 4). Selecting the most suitable fertilizer product for soil conditions is a key component of developing an efficient N management strategy.

Nonmanagement/Environmental Factors

Bad Weather, Crop Productivity and N Use

Figure 6. Average continuous corn yields and corresponding rainfall pattern at the WQFS. The amount of preplant rainfall is the sum of rainfall occurring in the month before and the month during planting.

Figure 6. Average continuous corn yields and corresponding rainfall pattern at the WQFS. The amount of preplant rainfall is the sum of rainfall occurring in the month before and the month during planting.

Results from a six-year study at the WQFS illustrate the extent to which high NO3-N concentrations in tile drainage water may be beyond a farmer’s control. When yields fall significantly below the expected yield or goal due to reasons other than N deficiency, the crop may not use all the available N. The residual can be leached into drainage water the following spring. At the WQFS, a linear relationship was observed between yield and drainage water NO3-N concentration (see Figure 5). For every 14 bu. per acre decrease in yield from a yield goal of 150 bu. per acre, NO3-N concentration in drainage water increased 10 ppm. Throughout the Cornbelt other researchers have observed similar patterns. Unfavorable weather patterns, especially hot dry conditions during crop growth and development, can significantly raise drainage water NO3-N concentrations the following spring.

Good Weather and Mineralization of Soil Organic Matter:

Surprisingly, especially favorable weather resulting in good yields may also negatively affect water quality the following spring. In Figure 5, results from the 1997 growing season appear as an outlier. In this year, yields of approximately 160 bu. per acre were followed by spring NO3-N concentrations above 20 ppm. The 1997 growing season was marked by less rain and saturated soil conditions during planting and early crop growth but relatively more rain during August when compared to other years (see Figure 6). This rainfall pattern was optimal for crop growth but also for soil biological processes. The amount of N released from soil organic matter was likely greater than in other years and some of this additional N was not used by the crop.

Other System Factors to Consider

Table 3. Differences among replicates of the fall swine manure treatment at the Water Quality Field Station. Mean tile drainage water NO3-N concentrations, drainflow volumes and estimated mass N loss are for January through early May 2000.

Table 3. Differences among replicates of the fall swine manure treatment at the Water Quality Field Station. Mean tile drainage water NO3-N concentrations, drainflow volumes and estimated mass N loss are for January through early May 2000.

Tile Flow Rate: While the focus of N management and water quality discussions tends to be on the concentration of NO3-N in drainage water, the real concern for both producers and environmentalists is the total amount of N lost. Concentration is certainly a key factor governing the amount lost, but should not be considered independently of the amount of water flowing from a tile drain. The amount of NO3-N lost is the product of the volume of water lost and the concentration of NO3-N in that water. The importance of this concept is illustrated by data from individual replicates of an N management treatment that uses fall swine manure as the only supplemental N source for a continuous corn system. The average NO3-N concentrations measured in the drain from replicate 1 was 14 ppm, while other replicates had concentrations from 22 to 33 ppm (see Table 3). However, the volume of water drained from replicate 1 was several times higher than that drained from the other replicates, resulting in the greatest total amount of N loss in replicate 1. These data demonstrate both the need to understand the relationship between concentration and flow volume and the fact that identical management systems may have different water quality impacts due to differences in hydrology.

Tile Spacing

Table 4. Selected statistics (annual average, minimum and maximum values, and the coefficient of variation) for NO3-N concentrations measured in 24-hour flow proportional composite samples of tile drainflow. Tile spacing (TS) treatments are 33 feet, 66 feet, and 100 feet (based on WQFS data). †Average values within columns followed by the same letter are not significantly different (P = 0.05 level).

Table 4. Selected statistics (annual average, minimum and maximum values, and the coefficient of variation) for NO3-N concentrations measured in 24-hour flow proportional composite samples of tile drainflow. Tile spacing (TS) treatments are 33 feet, 66 feet, and 100 feet (based on WQFS data). †Average values within columns followed by the same letter are not significantly different (P = 0.05 level).

Since inter-tile spacing alters the rate of water movement through the soil, it should also affect water quality. A study at the WQFS compared the impact of the recommended tile spacing (66 feet) with a narrower spacing (33 feet) and a wider spacing (100 feet). The recommended spacing resulted in concentrations that averaged 4 to 5 ppm lower than the wider or narrower spacings (see Table 4). The recommended spacing optimized yields (see Figure 7) and N uptake in grain. The lower NO3-N concentration in tile drainflow reflects this relatively better crop N-use efficiency with 66-foot spacing. However, the total mass N loss was lowest with 100-foot spacing reflecting the lower volume of water removed by the tile drain.

Somewhat different results were obtained on the low organic matter soil at the SEPAC study site. Tile spacings were 16, 33, and 66 feet, and there were no consistent differences in nitrate concentrations among spacings. However, the narrower spacings (more intensive drainage) had greater drainage than the wider spacings. Over the 15-year study, the amount of annual rainfall removed by the tile drains was, on average, 20.6 percent, 14.8 percent, and 12.0 percent for the 16-, 33-, and 66-foot spacings, respectively. Generally the nitrate-N losses were greater from the narrower spacings compared with the wider spacings, because the drainflow per acre was greater from the narrower spacings. This was especially true in years with high N fertilizer rates, continuous corn, and no cover crop (1986-88 period, see Figure 8) compared to the years with a corn-soybean rotation, lower fertilizer N rates, and a winter cover crop after corn (1997-99). The results suggest that wider drain spacings are preferable for reducing nitrate loads to surface waters, and that future drainage design should optimize drain spacing to reduce nitrate loads while providing adequate drainage for crop growth.

Tillage

Since tillage mixes the soil and can accelerate the breakdown of crop residue and mineralization of soil organic matter, it might be expected that conventional tillage would increase NO3-N loss in tile drainage water. In general, the studies that have been conducted have shown only small influences on drainflow NO3-N concentrations (with no-till being lower than conventional) and volumes (with no-till being higher than conventional) and these have not translated into differences in system total N loss.

Sampling Protocol Factors to Consider

Effect of Sampling Frequency

Finally, both the timing and the number of samples collected can affect the sample value itself. The data from the WQFS illustrate the critical point that concentrations may vary widely on a day-to-day basis. For example, in 2000, 66-foot tile spacing resulted in average NO3-N concentrations of 18.8 ppm, but individual values ranged from 5 ppm to 37 ppm (see Table 4). There is a certain amount of error associated with any given value that can only be reduced by collecting numerous samples in time (for a given tile line) or space (multiple tile lines for the same system).

Table 5. Percent probability that the NO3-N concentration measured in one sample collected weekly, monthly, or quarterly will be within 5 percent, 15 percent or 25 percent of the average annual value for that time period.

Table 5. Percent probability that the NO3-N concentration measured in one sample collected weekly, monthly, or quarterly will be within 5 percent, 15 percent or 25 percent of the average annual value for that time period.

For a given tile line, Table 5 shows the probable accuracy of sampling on a weekly, monthly, or quarterly (once every 3 months) basis, rather than daily. These results show the likelihood that the average concentration of daily composite samples taken periodically during a year will fall within a given percentage of the true average concentration (which was assumed to be the average of the daily concentrations). Results are based on a three-year analysis of three drains at SEPAC, using composited daily samples (flow-proportional samples over a 24-hour period). For example, if the true average concentration of water in a certain tile was 20 ppm, there is a 52 percent chance that the average of samples taken on one day every week would be between 19 ppm and 21 ppm (i.e., within 5 percent), and a 99 percent chance that the weekly average would be between 15 ppm and 25 ppm (i.e., within 25 percent). Monthly sampling only has a 93 percent probability of accuracy within 25 percent, and quarterly sampling only a 79 percent probability. Sampling less than quarterly is less likely to result in a reliable estimate of average nitrate concentration.

Figure 9. Relationship between rate of drain flow and NO3-N measured in one drain at SEPAC in 2000.

Figure 9. Relationship between rate of drain flow and NO3-N measured in one drain at SEPAC in 2000.

Using flow-proportional sampling of drainage water to collect a sample every day would be the most accurate way to assess the average concentration; however, this is usually not feasible except in research studies. The accuracy of typical “grab samples” will be less than the values given in Table 5, but we don’t know how much less. Best approaches to sample collection to derive meaningful information at a minimum cost are the subject of several ongoing research initiatives.

Effect of Drainflow Volume

People often ask if they should consider the rate of flow in the tile drain when sampling drainage water to monitor NO3-N concentrations. Results from both the WQFS and SEPAC research sites have shown that NO3-N concentrations in tile drains are not typically strongly influenced by drain flow. Over the 15-year study period at SEPAC, there was a weak negative correlation of concentration with flow rate, meaning there was only a slight tendency for higher flow rates to have lower concentrations. An example of the correlation for one year and one drain at SEPAC is shown in Figure 9. It is probably advisable to avoid sampling on days of extreme high or low flows, but “typical” high flows during spring should not pose a problem.

 

It is the policy of the Purdue University Cooperative Extension Service that all persons have equal opportunity and access to its educational programs, services, activities, and facilities without regard to race, religion, color, sex, age, national origin or ancestry, marital status, parental status, sexual orientation, disability or status as a veteran.

Purdue University is an Affirmative Action institution. This material may be available in alternative formats.

Published: January 2005

Purdue University, 610 Purdue Mall, West Lafayette, IN 47907, (765) 494-4600

© 2019 Purdue University | An equal access/equal opportunity university | Copyright Complaints | Maintained by Soil and Water

If you have trouble accessing this page because of a disability, please contact Soil and Water at agweb@purdue.edu.