Home » Posts tagged 'soil sample'

Tag Archives: soil sample

ABOUT ME

osunpk

osunpk

Since 2008 I have served as the Precision Nutrient Management Extension Specialist for Oklahoma State University. I work in Wheat, Corn, Sorghum, Cotton, Soybean, Canola, Sweet Sorghum, Sesame, Pasture/Hay. My work focuses on providing information and tools to producers that will lead to improved nutrient management practices and increased profitability of Oklahoma production agriculture

View Full Profile →

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 6,037 other followers

Soil sample handling practices can affect soil nitrate test accuracy

From Guest Authors,
Bryan Rutter, PhD student and Soil Testing Lab Manager, Kansas State University
Dr. Dorivar Ruiz Diaz, Soil Fertility Specialist, Kansas State University

The accuracy of a soil test is limited, in part, by the quality of the tested sample. For this reason, strong emphasis is placed on ensuring representative samples are collected in the field. However, these samples must also be handled properly after they have been collected.

Soils are home to a diverse population of microorganisms, many of which help decompose crop residue and cycle nutrients in soils. This nutrient cycling is crucial for crop production, but can skew soil test results if it continues in soil samples after they have been collected.

Microorganisms drive the soil nitrogen cycle

The nitrogen (N) cycle in soils is particularly complex and is strongly influenced by microbial activity and, therefore, temperature and soil moisture conditions. Bacteria and fungi consume organic material and use carbon as an energy source. During this process, N contained in the organic matter undergoes several transformations, ultimately converting it to ammonia. This conversion from organic-N to inorganic-N (NH4+, ammonium) is called “mineralization.” Plants can then take up the ammonium (NH4+), or converted to nitrate (NO3) by certain bacteria through a process known as “nitrification”.

The microbial activity requires moisture and heat, and the processes described above happen more quickly in warm, wet soils than in cold, dry soils. Microbial activity does not stop just because a sample has been collected and put in a bag. This activity continues as long as the environmental conditions are favorable. As a result, soil tests for plant-available N have the potential to change substantially if samples are not handled properly. This is an important consideration for growers because these soil test results are used to determine the profile-N credit and, ultimately, adjust N fertilizer recommendations.

Research study on soil sample storage

A recent study at the K-State Soil Testing Lab illustrates what can happen if sample submission is delayed.  For this study, soil was collected from the Agronomy North Farm (Manhattan, KS) and thoroughly mixed/sieved to homogenize the material. This soil was then placed into sample bags, which were randomly assigned to different combinations of storage temperature and duration. One set of samples was kept in a refrigerator while the other set was kept in a cargo box in a truck bed. To monitor changes in soil test levels over time, three sample bags were removed from the refrigerator and truck box every two days (48 hours) and tested in the lab.

Figure 1. Change in soil test nitrogen parameters over a 14-day storage period. Samples stored in an unrefrigerated cargo box are indicated by purple points. Samples stored in a refrigerator (38F) are indicated by grey points. Graphs by Bryan Rutter, K-State Research and Extension.

Figure 1. Change in soil test nitrogen parameters over a 14-day storage period. Samples stored in an unrefrigerated cargo box are indicated by purple points. Samples stored in a refrigerator (38F) are indicated by grey points. Graphs by Bryan Rutter, K-State Research and Extension.

Figure 2. Difference in the soil test nitrogen credits between refrigerated and unrefrigerated samples over a 14-day storage period. Profile-N credits assume a 24-inch profile soil sample depth, and are calculated as:  N ppm x 0.3 x 24 inches. Graph by Bryan Rutter, K-State Research and Extension.

Take home points from the K-State Soil Testing Lab study:

  • Mineralization and nitrification led to more than a 3x increase in soil test nitrate in the undried and unrefrigerated “Truck Cargo Box” samples (purple points in Figure 1).
  • Soil test nitrogen did not change substantially in refrigerated samples.
  • Profile-N credits calculated from soil test N results were nearly 100 lbs of N/acre higher for the unrefrigerated samples (Figure 2).
  • Improper handling and storage of soil samples can dramatically reduce soil test accuracy and may lead to under or overfertilizing crops.

K-State Soil Testing Lab Recommendations

  • Submit soil samples to the lab as soon as possible, ideally on the same day they were collected.
  • If same-day submission is not possible, samples should be air-dried or placed in a refrigerator set at 40 degrees F or less.

Please see the accompanying article “The challenge of collecting a representative soil sample” for guidance on field soil sampling practices.

For detailed instructions on submitting soil samples to the K-State Soil Testing Lab, please see the accompanying article “Fall soil sampling: Sample collection and submission to K-State Soil Testing Lab”.

For detailed information on how N credits are calculated please see the MF-2586 fact sheet: “Soil Test Interpretations and Fertilizers Recommendations”.

Bryan Rutter, PhD student and Soil Testing Lab Manager
rutter@ksu.edu

Dorivar Ruiz Diaz, Soil Fertility Specialist
ruizdiaz@ksu.edu

The original article can be found on the KSU Agronomy E-update site
https://eupdate.agronomy.ksu.edu/article_new/soil-sample-handling-practices-can-affect-soil-nitrate-test-accuracy-511-4

2013 Wheat and Canola preplant soil test results

Every few years I request the results of all soil samples submitted to OSU Soil, Water, & Forage Analytical Labs (www.soiltesting.okstate.edu) under the crop codes of winter wheat and winter canola.   Within this data set I can look at trends occurring across the state over time.  In this report I will focus on the 2013 results but make some comparison with the 2011 sample values.

As it pertains to mobile nutrients such as N, S, and B there is little that can be applied from the previous year’s soil samples because their levels in the soil change rapidly.  Samples must be collected every year to determine the current status.  However the soil test levels of immobile nutrients, P, K, Mg, ect are relatively stable over time and the recommendation is to take a close look at these values every three to five years.

In 2013 the number of sample submitted increase.  There were nearly 1000 more wheat soil samples (2733 to 3574) and 200 more canola soil samples (33 to 231).  If the distribution of nutrient levels of the two years are compared the only significant change is that the soil test NO3 level was significantly lower in 2013 (Tables 1 and 2).  This is attributed to the extremely dry 2012 spring and summer which delayed the breakdown of wheat straw and immobilization of residual N.

 

Table 1 and 2. Summary from all samples submitted to SWFAL under the wheat and canola crop codes in 2001 and 2013.

Table 1 and 2. Summary from all samples submitted to SWFAL under the wheat and canola crop codes in 2001 and 2013.

 

Reviewing the 2013 values the most concerning aspect is that 72% of the 3800+ soils samples had a Mehlich 3 P value below optimum soil test phosphorus (STP) of 65 (Figures 1 and 2). That adds up to 109,000 acres needing phosphorus, if you assume each sample represents 40 acres.    There is no way to determine how much P2O5 if any was applied to these particular fields.  However, an estimated impact of not fertilizing can be calculated.   Based on the Oklahoma typical average yield of just below 40 bpa, it would cost the state approximately 575,000 bushels if the land went unfertilized.  At $5.00 a bushel that is $2.8 million in revenue.  To remedy the low STP it would take approximately 2.76 million lbs P2O5 at a cost of $1.5 million ($0.50 per lb).

In the NPKS response study wheat fields across the state were evaluated for a response to additional (in addition to producer’s standard practice) nitrogen, phosphorus, potassium, and sulfur.    Phosphorus was the most limiting nutrient at 7 of the 59 harvest locations.  A response to P occurred more often than any of the other nutrients tested.   It is important to note at all seven fields had been fertilized with P that season, however each time it was below the OSU recommended rate.   The response study was a great reminder that it is important to have a good soil test and to follow the recommendations.

 

 

Figures 1 and 2. 1)Range of soil test P levels (Mehlich 3) for all samples submitted to SWFAL in 2013 under the wheat and canola crop codes. 2) Range of Soil Test P level for all samples with STP<65.

Figures 1 and 2. 1)Range of soil test P levels (Mehlich 3) for all samples submitted to SWFAL in 2013 under the wheat and canola crop codes. 2) Range of Soil Test P level for all samples with STP<65.

 

Soil pH on the other hand showed a slight improvement from 2011. The percent of samples under 5.5 decreased by 4%, 25 to 21. Of the samples <5.5 the majority fall within the 5.0-5.5 category, which for winter wheat is still within the optimum growth window (Figures 3 and 4).  These numbers are a good sign however two points should be made.  There is a significant amount of winter wheat acres that is not sampled; much of this is likely to fall below 5.5 soil pH.

 

Figures 3 and 4. 1)Range of soil pH levels  for all samples submitted to SWFAL in 2013 under the wheat and canola crop codes. 2) Range of soil pH levels for all samples with pH<5.5.

Figures 3 and 4. 1)Range of soil pH levels for all samples submitted to SWFAL in 2013 under the wheat and canola crop codes. 2) Range of soil pH levels for all samples with pH<5.5.

 

Additionally grid soil sampling and variable rate lime should consider on any field which the composite soil sample pH ranges from the high 4’s to the high 5’s.  For example a 75 ac field near Deer Creek had a composite soil sample test pH of 5.3 and buffer index of 6.5.  The OSU lime recommendation, for a wheat crop, was 2.2 ton per acre for a total of 166 tons to lime the entire field.  However the producer grid soil sampled the field himself at a 2.5 acre resolution (31 samples).  Figure 5, shows that the pH of the field ranged from 4.4 to 7.9.  Only 33 tons of lime would be required if the field were limed using a variable rate technologies.   Cutting the total amount applied by 133 tons would save the producer approximately $4000.

 

Figure 5. Soil pH results from a 75 acre field that was grid soil sampled at a 2.5 ac resolution.

Figure 5. Soil pH results from a 75 acre field that was grid soil sampled at a 2.5 ac resolution.

 

Oklahoma wheat and canola producers must take advantage of the weather when it goes their way.  Yet if the crop does not have the proper soil pH and nutrients under it, it will never reach its potential.  Take the time to collect a soil sample and send it in to a lab. The hour it takes to collect the sample a few dollars you spend on analysis will help ensure that crop you are producing has the best chance of hitting maximum yield in the most economically and environmentally sound manner.

 

Related Factsheets

OSU Soil Test Interpretations

http://npk.okstate.edu/documentation/factsheets/PSS-2225web2013.pdf

Fertilization Based on Sufficiency, Build-up and Maintenance Concepts

http://npk.okstate.edu/documentation/factsheets/PSS-2266web.pdf