ASABE Logo

Article Request Page ASABE Journal Article

Soil and Crop Effects of a Subsurface Fluid Lime Applicator

Joaquin Casanova1,*, Garett Heineck2, Melissa K. LeTourneau3, Jeremy C. Hansen1, Jenny L. Carlson4, David Huggins2

Published in Applied Engineering in Agriculture 40(3): 351-362 (doi: 10.13031/aea.15939). 2024 American Society of Agricultural and Biological Engineers.

1    NW Sustainable Agroecosystems, USDA Agricultural Research Service, Pullman, Washington, USA.

2    USDA Agricultural Research Service, Pullman, Washington, USA.

3    Ginkgo BioWorks Inc, Sacramento, California, USA.

4    Crops and Soil Science, Washington State University, Pullman, Washington, USA.

*    Correspondence: joaquin.casanova@usda.gov

The authors have paid for open access for this article. This work is licensed under a Creative Commons Attribution NonCommercial NoDerivatives 4.0 International License https://creative commons.org/licenses/by nc nd/4.0/

Submitted for review on 12 January 2024 as manuscript number MS 15939; approved for publication as a Research Article by Associate Editor Dr. Kevin McDonnell and Community Editor Dr. Heping Zhu of the Machinery Systems Community of ASABE on 12 April 2024.

Mention of company or trade names is for description only and does not imply endorsement by the USDA. The USDA is an equal opportunity provider and employer.

Citation: Casanova, J., Heineck, G., LeTourneau, M. K., Hansen, J. C., Carlson, J. L., & Huggins, D. (2024). Soil and crop effects of a subsurface fluid lime applicator. Appl. Eng. Agric., 40(3), 351-362. https://doi.org/10.13031/aea.15939

Highlights

Abstract. Deep-banding ammonia and urea-based nitrogen fertilizers under continuous no-tillage can result in stratified acidification at the fertilizer-injection depth. Currently, producers address this acidification through primary tillage operations that mix the soil and dilute the acidification, but increase the hazard of soil erosion. Typically, lime is required to ameliorate soil acidity; however, surface-applied lime does not immediately address subsurface acidity, and often tillage is used to incorporate lime and correct acidified layers created through deep-banded fertilizer placement under no-tillage. Here, we designed, developed, and tested a subsurface applicator for fluid lime that can target specific soil depths and rates. The liming system was tested in Eastern Washington on a field that had been under continuous no-tillage for 25 years, had stratified soil acidity, and was to be planted to winter pea, a crop known to be sensitive to acid soil conditions. Soil pH and carbon distribution, microbial changes, biomass, and leaf area index (LAI). Results indicated that the lime targeted the seed zone and raised pH significantly; effects on biomass were marginally significant; and effects on LAI and vegetation cover were insignificant. Topography had a significant effect on all variables due to changes in applicator effectiveness with changes in soil mechanics, meaning as-applied rates varied from desired rates.

Keywords.Liming, No-tillage, Precision agriculture, Soil acidity, Soil microbiology.

Agricultural soils globally (Zhang et al., 2023) and in the inland Pacific Northwest (iPNW) are experiencing increasing acidification, largely driven by the continuous use of ammoniacal-based fertilizers (McFarland and Huggins, 2015). Applied ammonia- and urea-nitrogen fertilizers release H+ during nitrification; furthermore, accelerated base cation (Ca2+, Mg2+) leaching can occur as NO3- is readily moved to the subsoil during high-precipitation winter fallow periods. Deep-banded fertilizer in no-till systems leads to stratification of acidity, lowering the pH in the fertilizer placement zone (Brown et al., 2005). Soil acidification can increase active aluminum, which is toxic to crops and inhibits root growth and nutrient acquisition (fig. 1; Koenig et al., 2013). The long-term effects of lime application include increasing yield and phosphorous availability (Holland et al., 2019).

Acidic soils can impact the soil microbiome as measured by phospholipid fatty acid (PLFA) analysis, which studies the microbial composition in the soil through the fatty acid composition (Frostegård and Bååth, 1996). Low pH has the effect of decreasing arbuscular mycorrhizal (AM) fungi and increasing some bacteria (Bååth and Anderson, 2003; Rousk et al., 2010). Total microbial biomass is reduced in acidic soils (Pietri and Brookes, 2009). Symbiotic bacteria that form nodules in leguminous crops are important for nitrogen fixation, and nodulation is reduced under low pH (Lie, 1969).

Topographical factors can lead to spatial variation in soil pH, as south slopes can experience less base cation leaching, toeslopes typically have more soil organic matter with greater buffering capacity, while summits can have more leaching, and north slopes often have deeper soils and fewer base cations near the surface (Egli et al., 2007). This spatial variability in soil pH drives the need for precision agricultural methods to address field-scale soil acidity.

A common practice used by farmers to treat acidic soil is to apply a pH-neutralizing material that can come in several different forms, depending on the source and material formulation, which, in turn, affect the application method (Mahler and McDole, 1994). The main sources include limestone, burned lime, slaked lime, marl, oyster shells, and byproducts like ash or sugar beet "lime." Lime can be in the chemical form of calcium carbonate, or calcium carbonate with magnesium carbonate (dolomitic lime); oxides; hydroxides; and by-product materials. Lime can be applied in solid form, as powder, prilled, or as a fluid suspension. The effectiveness of lime in neutralizing acidity can be quantified by the effective calcium carbonate equivalent (ECCE), which is a function of mesh size (for powders) and purity (Thompson et al., 2016). Lime is usually surface-applied and incorporated through tillage, or during seeding or fertilizer application.

Lime materials are not very soluble and do not readily move in soil. Correcting soil pH is typically achieved by physically mixing the material throughout the soil depth, requiring a pH adjustment, or by placing lime in close proximity to the acidified area. Studies in the iPNW generally show that the impact of surface-applied lime on soil pH is limited to near the soil’s surface. These include increases in soil pH in the top 5 cm, which may be above the stratified acidity zone in no-till systems, and here, increases in yield have been small and mixed (Brown et al., 2008). Consequently, surface-applied lime in conservation- and no-till systems results in further stratification in pH as, without tillage, the lime is not well incorporated. This difficulty of surface-applied lime has been noted elsewhere (Azam and Gazey, 2021; Hume et al., 2023). Even without lime, no-tillage leads to pH stratification (fig. 1; McFarland, 2016). There are limited options for effectively addressing acidification while maintaining soil conservation practices, particularly as deep-banded fertilizer applications continue exacerbating the problem. Thus, in this article, we explore an option for directly applying lime to the subsurface to target the highly stratified acidity layer.

Figure 1. Stratification in soil pH at Rockford and Pullman, WA which were under continuous no-tillage (left) and KCl extractable aluminum response to decreasing soil pH (right) (McFarland, 2016).

To target the lime application to the acidified fertilizer band-depth in no-tillage, we developed a system to directly inject fluid lime into the subsurface. This is accomplished by way of cultivator sweeps, which limit soil inversion and are a lower-disturbance tillage implement (Bista et al., 2017). Soil and winter pea plant characteristics were monitored throughout the growing season in 2023. First, we describe the test site and sampling protocols, the calibration of the implement, and the details of its application in the field. We then investigated the effects of the lime on soil chemistry, including pH, total carbon and nitrogen, and 13C, which are found in higher concentrations in lime compared to background soil. We also examined the soil microbiome through PLFA, crop growth, and yield. Finally, we discuss changes for an improved system based on this season’s results.

Materials and Methods

Here, we describe the research site, design and test procedures for the lime application, in-season sampling and analysis protocols, and statistical methods.

Site Description: R. J. Cook Agronomy Farm, Pullman, Washington

Study Site

The R.J. Cook Agronomy Farm (CAF, fig. 2), north of Pullman, Washington, is a 60-ha site operated as part of the USDA Long-Term Agroecosystem Research (LTAR) network (Kleinman et al., 2018), with a comprehensive meteorological, harvest, and soil dataset (Huggins, 2015). The field is divided into West (conservation tillage) and East (no-till) portions, 23 and 37 ha, respectively, both under dryland and wheat-based cropping systems. The soil series are primarily Thatuna (a fine-silty, mixed, superactive, mesic Oxyaquic Argixeroll), Palouse (a fine-silty, mixed, superactive, mesic Pachic Ultic Haploxeroll), and Naff (a fine-silty, mixed, superactive, mesic Typic Argixeroll) (USDA-NRCS, 2023). The region has a Mediterranean climate with a mean annual temperature of 9°C and a total a total precipitation of 582 mm (rain and snow equivalent) (NOAA Climate Data Online, 2023).

Figure 2. R.J. Cook Agronomy Farm (46°47'N, 117°5'W, 770 m above mean sea level) Northeast of Pullman, Washington. Cook West and East fields are outlined in yellow.
Figure 3. Lime test plots at CAF. From top to bottom four topographic locations are described: A (northern backslope), B (summit), C (southern backslope), and D (toeslope). Plot coloration describes the actual range of lime applied to each treatment in kilograms per hectare (shown in the legend). Individual rectangles are the 2 passes of the tractor per plot, side-by-side.

For this study, four topographical blocks were selected in the eastern half of the East field (fig. 3) to examine different topographical positions, using the Ruhe (1960) classification: A (northern backslope), B (summit), C (southern backslope), and D (toeslope). These were further subdivided into eight 5 × 10 m plots each, with each plot divided into two passes of the tractor with sweep injection system (2.5 × 10 m), with four lime rate treatments (with one as a control, referred to as T0, and three lime rates, T1, T2, and T3) assigned in a randomized complete block design, containing two replications per treatment per block. The range of liming rates was determined based on soil samples taken in Fall 2022 using methods from McFarland et al. (2020). Lime was applied on 6 October 2022, prior to seeding Austrian Winter Pea on the same day.

Sampling and Analysis

Several samplings took place over the 2023 harvest year. Each sampling event collected data on soil, vegetation, and aerial imagery, described below. The sampling protocol is intended to be yearly in Spring for several seasons to monitor year-to-year changes.

Soil Samplings

Soil cores were taken on 10 June 2022, prior to lime application to determine baseline soil chemistry. Samples were taken to 30 cm and subdivided into 0–5 cm, 5–10 cm, 10– 20 cm, and 20–30 cm depths. Three cores, composited, were taken with a utility task vehicle-mounted hydraulic Giddings soil sampler (4 cm diameter) on the plots in each block designated as “control.” The divided samples were sent to Best Test Labs (Best Test Analytical Services, 2023) for the following analyses, among others: aluminum (KCl extractable), organic matter (OM), base cations, cation exchange capacity (CEC), and Adams-Evans buffer pH. The baseline measurements of pH and buffer pH were used to establish lime treatment rates, assuming a calcium carbonate equivalent (CCE) of 74% from the manufacturer of the lime (Microna Agriculture, 2023) and 15 cm depth, following the lime requirement estimation from McFarland et al. (2020).

The following spring, after the fall lime application (27 April 2023), hand cores were taken to 15 cm (4 cm diameter) in two of the eight plots in each block, at the T1 and T2 treatments. Samples were taken along two transects in one half of each sampled plot, with each core spaced at 5, 10, 15, 20, 25, and 30 cm from the plot edge. Cores were subdivided into 0–3 cm, 3-6 cm, 6–9 cm, 9–12 cm, and 12–15 cm, dried, ground, and sieved to 2 mm, then tested for 1:1 pH. Referred to as “detailed sampling,”  this gave a two-dimensional look at the lime distribution in the soil and an assessment of the uniformity of the applicator along a transect.

On 21 May 2023, another soil sampling was conducted using the same hand soil corer. In this sampling, referred to as “bulk sampling,” three cores were taken in each of the plots in each block. Coring was constrained to one half of the plot; the other half was reserved for biomass sampling later. The three cores were subdivided into 0-5 cm, 5-10 cm, and 10-15 cm depths and composited. Samples were dried, ground, and sieved to 2 mm, then tested for 1:1 pH with an Accumet AB200 pH meter and bulk carbon, nitrogen, and d13C using a Costech ECS4010 Elemental Analyzer coupled to a Thermo Finnigan Delta XP isotope ratio mass spectrometer. The lime itself was also tested for d13C, which is much higher in the lime than in soil (1.097 for the lime used in this study, compared to -26.2327 in the baseline soil samples), allowing us to monitor carbon derived directly from the applied lime vs organic matter and other soil carbon. In this case, an increase in soil d13C can be attributed to the lime.

The final sampling was conducted on 24 May 2023, which was conducted for biological analyses of the bulk, rhizosphere, and rhizoplane soil. We follow the procedures of Hansen et al. (2018), where rhizosphere refers to soil loosely adhering to the roots, and rhizoplane is tightly adhering to the roots. The bulk soil was collected as described above at 0-5 cm, 5-10 cm, and 10-15 cm depths. In the same plots as the bulk soil sampling, enough pea plant roots were sampled with a shovel to get sufficient soil from the rhizosphere and rhizoplane portions. The sampled plants were placed in coolers and immediately taken to the lab. The collected soil and root mass were then gently broken apart to eliminate bulk soil not associated with roots. Rhizosphere soil was obtained by shaking off the loosely adhering soil from the roots. After the rhizosphere soil was removed and collected, the remaining plant roots with the more tightly adhering soil were placed in a sterile 50-mL tube. 10 mL of sterile distilled H2O was added to each tube to submerge roots, vortexed for 1?min, and sonicated for ten?min to remove rhizoplane soil (Schlatter et al., 2020). The rinsed roots were removed from the tube and the resulting soil and solution was freeze dried and saved for analysis. After freeze drying, PLFA and neutral lipid fatty acid (NLFA) analysis was conducted following the high-throughput method described by Buyer and Sasser (2012). Peak responses of fatty acids having 20 carbons or less were summed into biomarker groups as described by Hansen et al. (2018). Because the rhizosphere and rhizoplane analysis indicate a symbiotic microbiome around the roots, and the root samples were destructively sampled in this analysis, we did not track nodulation.

Plant Samplings

During the season, leaf area index (LAI) was measured in the same plots’ halves as the soil sampling, using an ACCUPAR LP-80 from METER Group (METER, 2023), with one above canopy measurement and five below along a 1 m transect. LAI was measured on 22 May 2023, 27 June 2023, and 17 July 2023. Biomass samples were taken on 20 July 2023 in the halves of plots not used for soil sampling. All plants in a 2 × 1 m quadrat were cut at the soil surface and bagged. Samples were allowed to air dry, then weighed, threshed to separate grain and residue, weighed, and oven-dried at 60°C for 48 h. The oven-dried samples were finely ground and tested in a Costech ECS4010 elemental analyzer (Valencia, Calif.) for carbon and nitrogen content.

System Design and Test

The lime applicator (fig. 4) consisted of a John Deere 5425 (Moline, Ill.) and three 0.84 m wide cultivator sweeps, which had a total width of 2.5 m. On the underside of each sweep, 10 steel tubes were welded, spaced 5 cm, with rear-facing openings to inject into the soil. The tubes connected to a manifold were fed with fluid lime from a tank. Lime was moved into the manifold with a hydraulic pump, metered by a solenoid, and flow rates were measured using a FLOMEC flow meter (Great Plains Industries, Wichita, Kans.). In practice, the lime applicator injects the soil at a rate moderated by adjusting the tractor speed.

Figure 4. (Top) Fluid lime injection implement mounted on the three-point hook-up on a John Deere 5425 tractor for calibration. Lime was loaded into a tank, which was connected to a hydraulically driven pump that pressurized a PVC wet manifold. The manifold fed lines connected to each sweep. (Bottom left) Closeup of lime injection manifold and lines connected to sweeps. (Bottom right) Hydraulic pump and flow meter.

This required a two-fold calibration. First, the relationship between hydraulic pressure, set with a pressure relief valve, and lime flow rate was established with a stationary tractor and a full tank of water in place of lime. This is a common way to calibrate due to the high cost of fluid lime. The fluid was pumped into a graduated bucket during a 60-s period, using two nozzles per sweep. The measured volume divided by 60 seconds gave the flow rate, and measuring multiple nozzles indicated uniformity across the sweeps. Second, to dose lime in the field, the sweeps were tested in a field near CAF with similar soil. While the lime was being injected, the tractor drove over the equivalent area of the experimental plot. The tractor speed was adjusted in each pass, and afterward, the total lime applied as measured by the meter was noted. This gave a way to relate tractor gearing to the lime rate during lime application on the test plots.

Lime was applied to the test plots on 6 October 2022 at a nominal depth of 10 cm to target the stratified soil acidity zone. The lime was HydroCal fluid lime from Microna (Microna Agriculture, 2023) with 74% CCE and 16 lb/gal, which was used to calculate the lime rate in kg/ha. The target lime rates were 0 kg/ha (T0), 2000 kg/ha (T1), 4000 kg/ha (T2), and 8000 kg/ha (T3), although as-applied rates differed from these by up to 50%.

Statistics

To analyze the data, variation in as-applied rates prevented using the lime rates as categorical treatment levels. Instead, we used a mixed model approach using the R package lmerTest (Kusnetsova et al., 2017) in R 4.3.1 (R Core Team, 2016), with lime rate as a continuous variable. In order to ensure numerical stability, lime rate data was scaled by the maximum as-applied rate. Soil variables, grouped by depth, along with biophysical and PLFA variables, were examined. For pH and ??13C, we looked at the difference between baseline and post-lime measurements. PLFA variables were log-transformed to ensure normal distributions. The effect of lime rate was determined with a mixed model, with lime rate and topography as fixed effects, with a random intercept from a rep × block error term. Models with topography only, topography and lime, and topography × lime interaction were considered. The model with topography only was compared to the null model via likelihood ratio [lrtest in package lmtest (Zeilis and Hothorn, 2002)] to determine if there were topographical effects, then a model with topography and lime without interaction compared to topography only, and the model with interaction was similarly compared to the model without. In some cases (e.g., grain yield), we examined relationships with pH at 5 cm using similar mixed model structures.

To check sweeps’ accuracy in application, we used box and whisker plots of as-applied rates grouped by block and nominal treatment level. Plots of soil pH by transect position and depth from the detailed sampling help to assess lime uniformity.

Results and Discussion

Applicator Performance

.
Figure 5. Each panel represents a topographic position in the field, with the block name in the panel title:A (northern backslope), B (summit), C (southern backslope), and D (toeslope). Colored box-and-whisker plots represent each lime treatment within each position. Target lime treatments were T0 = 0 kg/ha, T1 = 2000 kg/ha, T2 = 4000 kg/ha, and T3 = 8000 kg/ha. The observed range of lime (CaCO3) applied is shown in kilograms per hectare. The y-axis is arbitrary for clarity in presenting box-and-whiskers.

Testing and calibration before field application showed that the flows across the different nozzles (table 1). The coefficient of variation for the output rate was equal to 10.2%, indicating adequate uniformity. Tractor speed, governed by gearing and RPM, was used to moderate application rates. A test prior to use on the test plots gave a calibration for application rate vs. lime rate (table 2). During application on the test plots, cumulative flow over each pass was noted, and there was considerable difficulty in achieving desired application rates precisely—some were off by over 50%. Figure 5 shows the as-applied rates in the different blocks and treatments. This was due to three major factors: topography, soil properties, and the operation of the tractor and its implement. Lower-lying areas with higher organic matter tended to clog the nozzles, which had to be periodically cleaned out. The organic matter from 0–5 cm was 6.40% in D, 4.35% in C, 2.97% in B, and 5.91% in A. Previous soil descriptions at the site indicated a clay content of 8% on all blocks but B (summit), where it was 10%. Uphill vs. downhill passes added variability in tractor speed, which also influenced lime rate. In higher gear settings, the tractor needed to gain sufficient speed before entering the plot area while the operator simultaneously dropped the three-point, opened the solenoid, and engaged the hydraulic pump. Finally, changes in the lime level in the tank affected the flow rate and the mass keeping the sweeps in the soil at a uniform depth, since draft force and vertical force increase with depth (Kiss and Bellow, 1981). Figure 6 shows the pH profile with depth from the detailed sampling. In most plots, the high pH due to lime is not uniform and is skewed highest near the surface (3 cm), but plot B showed more variability. This was on a summit, and the soil was harder and drier. These soil conditions made maneuvering the tractor more difficult and made control of depth and rate difficult relative to the others. Additionally, the tank was refilled prior to use on the summit, giving a greater downward force. These factors may have contributed to the peak in pH seen at 6 cm in some of the B-block transects. Overall, the detailed sampling demonstrated uniformity across a transect, with coefficients of variation near or less than 10% in the top 6 cm. With depth, the distribution is obviously not uniform. This is to be expected, as lime takes time to redistribute within the profile.

Figure 6. Each panel describes the soil test results from the initial post-application hand sampling in 2000 kg/ha (T1) and 4000 kg/ha (T2) treatments. Under the treatment name is listed the horizontal distance from the plot edge in centimeters. Colored lines within each panel represent the topographical positions ±1 standard error. pH values are reported for each soil sampling depth in centimeters. A = northern backslope, B = summit, C = southern backslope, and D = toeslope.
Table 1. Sweeps uniformity test.
Hydraulic Pressure
(psi)		Nozzle[a]Volume
(mL)		Time
(s)		Rate
(mL/s)
60LFR14006023.33
60LRR14006023.33
60MFR14006023.33
60MRR15506025.83
60RFR15006025.00
60RRR14006023.33
41LFR11006018.33
41LRR12006020.00
41MFR12006020.00
41MRR14006023.33
41RFR13006021.66
41RRR12006020.00

    [a]    LFR = Left sweep, Front Right nozzle, MRR = Middle sweep,     Rear Right nozzle, etc.

Soil Chemistry

Table 2. Speed calibration test.
TrialTotal
(gal)		Gear[a]Treatment Width
(m)		Length
(m)		PassesRate
(L/ha)		Rate
(kg/ha)
14.40A2 - 1600NA51026662.569452.46
22.40A4 - 1600T251023634.135155.89
31.60B3 - 1600T151022422.753437.26
44.98A2 - 1600NA51027540.8110698.47
53.74A3 - 1600T351025663.188034.60
62.85A4 - 1600T251024315.526122.62
71.24B3 - 1600T151021877.632663.88

    [a]    Tractor gear settings, the alpha numeric value indicates the main and sub gear setting, the second numeric value indicates the tractor engine RPM setting.

To examine the effect of the soil chemistry, pH from 0– 5 cm and 5–10 cm were examined, as well as the change in pH from the baseline measurements. There was a significant treatment effect on soil pH and on the increase in soil pH [table 3, fig. 7(1-2)]. The D block (toeslope) had the smallest pH increase. This may be due to the high organic matter there, which contributed to the higher reserve acidity that buffered the impact of the lime.

The ??13C in the top 0–5 cm and 5–10 cm showed similar relationships [table 3, fig. 7(3-4)], although there was also a significant effect of block. This implies that the different topographical positions retained different amounts of carbon from lime. It seemed to be lowest in areas with higher organic matter (D, organic matter 0–5 cm 6.40%), which possibly indicates more carbon from lime was released as CO2 as it neutralized reserve acidity in the organic matter or that the higher proportion of total carbon due to the organic matter meant carbon from the lime was a smaller fraction of the carbon.

Table 3. Results for lime rate effects mixed models.
pH
Increase
(0-5 cm)		pH
Increase
(5-10 cm)		??13C
Increase
(0-5 cm)		??13C
Increase
(5-10 cm)
Topography only*
Without interaction********.
With interaction

    [a]    *** = p<0.001, ** = p<0.01, * = p<0.05, . = p<0.1.

Figure 7. pH and ??13C increases from baseline by block vs. applied lime rate. Shaded areas show ±1 standard error. A = northern backslope, B = summit, C = southern backslope, and D = toeslope.

Crop Performance

Table 4 shows the ANOVA results for vegetation parameters. Peak LAI, biomass, and grain yield had a weak or no treatment effect. Grain yield and grain nitrogen were examined again using lime rate rather than treatment level, with significant treatment, block, and treatment x block effects. In the C block (south slope), there was an increasing trend in grain yield with lime rate and pH, contrary to the others [fig. 8(1-2)]. This block was the only one in which the 5–10 cm pH was above the recommendation for peas, 5.5 (Mahler and McDole, 1987), though all blocks exceeded this threshold in 0–5 cm. This block also received the most sun and had the least amount of time under snow cover. Interestingly, grain nitrogen concentration did show a treatment effect. Grain nitrogen increased with lime rate and pH in all blocks, though pH was not statistically significant [fig. 8(3-4)], which may be due to changes in the microbiome or the reduction in active aluminum due to neutralization. However, one season is insufficient to see the full effects of liming, which needs to be distributed into the soil and requires time to react. Brown et al. (2008) did not find significant effects of lime on yield in direct-seeded cereal crops; however, long-term studies that included legume effects of liming have been found (Holland et al., 2019).

Table 4. Results for lime rate effects mixed models.[a]
BiomassGrain YieldGrain
Nitrogen		Residue
Nitrogen		Peak
LAI
Topography only*********
Without interaction****.*.
With interaction******.

    [a]    *** = p<0.001, ** = p<0.01, * = p<0.05, . = p<0.1

Figure 8. Dry grain and grain nitrogen by block vs applied lime rate and pH 0-5 cm. Shaded areas show ±1 standard error. A = northern backslope, B = summit, C = southern backslope, and D = toeslope.

Microbiome

PLFA analysis provides many results, and table 5 shows the ANOVA results for some variables where significant effects were found. Arbuscular mycorrhizal (AM) fungi PLFA and neutral lipid fatty acids (NLFA) in the rhizoplane (RP), rhizosphere (RS), and bulk top 5 cm showed significant treatment effects. AM fungal populations tended to increase in higher pH environments [fig. 9(1-2)]. Similarly, this effect of pH on AM fungi was seen on grain yield and grain nitrogen [fig. 9(3-4)]. The change in the microbiome may have increased nutrient availability (Chen et al., 2018). Though the effect of AM fungi on grain yield was positive and significant (p<0.001), the effect on grain nitrogen was mixed and marginally significant, so there was likely a combination of factors influencing the peas’ development. Yin et al. (2021b) did not find a significant effect of lime on fungal composition but did on bacteria (Yin et al., 2021a). Other studies of the effect of pH have shown significant results (Bååth and Anderson, 2003; Rousk et al., 2010).

Table 5. Results for lime rate effects mixed models.[a]
AMFungi
0-5 cm		AMFungi NLFA
0-5 cm		AM Fungi
RP		AM Fungi NLFA
RP		AM Fungi
RS		AM Fungi NLFA
RS
Topography only*.
Without interaction**..*
With interaction****

    [a]     *** = p<0.001, ** = p<0.01, * = p<0.05,. = p<0.1. AM = arbuscular mycorrhizae, RP = rhizoplane, RS = rhizosphere, NLFA = neutral lipid fatty acid.

Table 6. Results for pH effects mixed models.
Grain
Yield		Grain
Nitrogen		AM Fungi
0-5 cm		AM Fungi
NLFA
0-5 cm		AM Fungi
RP		AM Fungi
NLFA
RP		AM
Fungi
RS		AM Fungi
NLFA
RS
Topography only****.
Without interaction***.
With interaction***..

    [a]    *** = p<0.001, ** = p<0.01, * = p<0.05,. = p<0.1. AM = arbuscular mycorrhizae, RP = rhizoplane, RS = rhizosphere, NLFA = neutral lipid fatty acid.

Figure 9. AM fungi in the bulk soil (0-5 cm) by block as a function of pH (0-5 cm) and lime rate, and grain yield and grain nitrogen as a function of AM fungi. Shaded areas show ±1 standard error. A = northern backslope, B = summit, C = southern backslope, and D = toeslope.

Design Revisions

Several design revisions are needed to better control the application rate and depth. First, the sweeps should be modified so that they maintain a more consistent depth. This could be accomplished through ultrasonic sensors monitoring the height of the frame above the soil and applying downward pressure to the sweeps with hydraulic actuators to achieve a target depth. Second, clogging of the nozzles could be avoided through the use of flared metal aprons surrounding them, to create a small pocket in the soil around the orifice. An alternative approach may also be to switch to subsurface placement of pelletized lime, which would avoid clogs. Finally, relying solely on tractor gearing to govern application rates was fallible, so feedback between the flow meter and a solenoid controlling flow could lessen variability in as-applied rates from target rates. To reduce variability in the future, the tractor should have a modified rear selective control valve to allow continuous flow that can be engaged by the same switch that opened the solenoid coming out of the pump.

Conclusions

A system using cultivator sweeps for subsurface application of fluid lime in agricultural soils was tested over a growing season of winter peas in the inland Pacific Northwest. Using sweeps to incorporate lime directly is a conservation tillage acidity mitigation approach as opposed to tilling surface-applied lime. However, the accuracy of the implement was poor, with wide variation (nearly 50%) from target rates. Several design improvements are necessary to provide better control and more reliable operation, including feedback between tractor speed and lime rate and automatic depth control. While the performance must be improved, it had the desired effect of mitigating soil acidity. The test plots will continue to be monitored over the crop rotation, as lime can take time to diffuse into the soil and affect soil chemistry and microbiome. After improvements, the system will be tested under multiple soil conditions.

Acknowledgments

This research was a contribution from the Long-Term Agroecosystem Research (LTAR) network. LTAR is supported by the United States Department of Agriculture. This research used resources provided by the SCINet project of the USDA Agricultural Research Service, ARS project number 0500-00093-001-00-D. The authors would also like to thank Mr. Ian Leslie, Mr. Zachary Smith, Mr. Zachary Skirvin, Mr. Robert Meadows, Mr. Armando Quintanilla Perez, Dr. Claire Phillips, and Ms. Leah Amazing Brown Chidziwe for help in sample collection and processing.

References

Azam, G., & Gazey, C. (2021). Slow movement of alkali from surface-applied lime warrants the introduction of strategic tillage for rapid amelioration of subsurface acidity in southwestern Australia. Soil Res., 59(1), 97-106. https://doi.org/10.1071/SR19329

Bååth, E., & Anderson, T. H. (2003). Comparison of soil fungal/bacterial ratios in a pH gradient using physiological and PLFA-based techniques. Soil Biol. Biochem., 35(7), 955-963. https://doi.org/10.1016/S0038-0717(03)00154-8

Best Test Analytical Services. (2023). Retrieved from https://besttestlabs.com/

Bista, P., Machado, S., Ghimire, R., Yorgey, G., & Wysocki, D. (2017). Conservation tillage systems. In G. Yorgey, & C. Kruger (Eds.), Advances in dryland farming in the Inland Pacific Northwest (pp. 99-124). Washington State University.

Brown, T. T., Koenig, R. T., Huggins, D. R., Harsh, J. B., & Rossi, R. E. (2008). Lime effects on soil acidity, crop yield, and aluminum chemistry in direct-seeded cropping systems. Soil Sci. Soc. Am. J., 72(3), 634-640. https://doi.org/10.2136/sssaj2007.0061

Brown, T., Koenig, R. T., Harsh, J. B., Huggins, D. R., & Rossi, R. E. (2005). Stratified soil acidity and the potential for aluminum toxicity in Inland Pacific Northwest direct-seed cropping systems. Proc. ASA-CSSA-SSSA Int. Meetings (pp. 6-10). ASA, CSSA, SSSA.

Buyer, J. S., & Sasser, M. (2012). High throughput phospholipid fatty acid analysis of soils. Appl. Soil Ecol., 61, 127-130. https://doi.org/10.1016/j.apsoil.2012.06.005

Chen, M., Arato, M., Borghi, L., Nouri, E., & Reinhardt, D. (2018). Beneficial services of arbuscular mycorrhizal fungi – from ecology to application. Front. Plant Sci., 9. https://doi.org/10.3389/fpls.2018.01270

Egli, M., Mirabella, A., Sartori, G., Giaccai, D., Zanelli, R., & Plötze, M. (2007). Effect of slope aspect on transformation of clay minerals in Alpine soils. Clay Miner., 42(3), 373-398. https://doi.org/10.1180/claymin.2007.042.3.09

Frostegård, A., & Bååth, E. (1996). The use of phospholipid fatty acid analysis to estimate bacterial and fungal biomass in soil. Biol. Fertil. Soils, 22(1), 59-65. https://doi.org/10.1007/BF00384433

Hansen, J. C., Schillinger, W. F., Sullivan, T. S., & Paulitz, T. C. (2018). Rhizosphere microbial communities of canola and wheat at six paired field sites. Appl. Soil Ecol., 130, 185-193. https://doi.org/10.1016/j.apsoil.2018.06.012

Holland, J. E., White, P. J., Glendining, M. J., Goulding, K. W., & McGrath, S. P. (2019). Yield responses of arable crops to liming – An evaluation of relationships between yields and soil pH from a long-term liming experiment. Eur. J. Agron., 105, 176-188. https://doi.org/10.1016/j.eja.2019.02.016

Huggins, D. R. (2015). The Cook Agronomy Farm LTAR: Knowledge intensive precision agro-ecology. Proc. 2015 American Geophysical Union, Fall Meeting. https://ui.adsabs.harvard.edu/abs/2015AGUFM.H53B1665H

Hume, R., Marschner, P., Mason, S., Schilling, R. K., Hughes, B., & Mosley, L. M. (2023). Measurement of lime movement and dissolution in acidic soils using mid-infrared spectroscopy. Soil Tillage Res., 233, 105807. https://doi.org/10.1016/j.still.2023.105807

Kiss, G. C., & Bellow, D. G. (1981). An analysis of forces on cultivator sweeps and spikes. Can. Agric. Eng., 23(2), 77-83.

Kleinman, P. J., Spiegal, S., Rigby, J. R., Goslee, S. C., Baker, J. M., Bestelmeyer, B. T.,... Walthall, C. L. (2018). Advancing the sustainability of US agriculture through long-term research. J. Environ. Qual., 47(6), 1412-1425. https://doi.org/10.2134/jeq2018.05.0171

Koenig, R. T., Schroeder, K. L., Carter, A. H., Pumphrey, M. O., Paulitz, T., Campbell, K., & Huggins, D. R. (2013). Soil acidity and aluminum in the Palouse region of the Pacific Northwest, Report FS050E. Washington State University Extension. https://pubs.extension.wsu.edu/soil-acidity-and-aluminum-toxicity-in-the-palouse-region-of-the-pacific-northwest

Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. (2017). lmerTest Package: Tests in linear mixed effects models. J. Stat. Softw., 82(13), 1-26. https://doi.org/10.18637/jss.v082.i13

Lie, T. A. (1969). The effect of low pH on different phases of nodule formation in pea plants. Plant Soil, 31(3), 391-406. https://doi.org/10.1007/BF01373812

Mahler, R. L., & McDole, R. E. (1987). Effect of soil pH on crop yield in Northern Idaho. Agron. J., 79(4), 751-755. https://doi.org/10.2134/agronj1987.00021962007900040033x

Mahler, R. L., & McDole, R. E. (1994). Liming materials. CIS 787. University of Idaho. Retrieved from https://www.grasshopperfertilizer.com/wp-content/uploads/2011/09/Liming-Materials.pdf

McFarland, C. R. (2016). Liming no-till soils and determining lime requirement in the Palouse region. MS thesis. Washington State University.

McFarland, C., & Huggins, D. R. (2015). Acidification in the inland Pacific Northwest. Crops Soils, 48(2), 4-12. https://doi.org/10.2134/cs2015-48-2-1

McFarland, C., Shiwakoti, S., Carpenter-Boggs, L., Schroeder, K., Brown, T., & Huggins, D. R. (2020). Evaluating buffer methods for determining lime requirement on acidified agricultural soils of the Palouse. Soil Sci. Soc. Am. J., 84(5), 1769-1781. https://doi.org/10.1002/saj2.20111

METER. (2023). ACCUPAR LP-80 - Canopy Interception and Leaf Area Index. Retrieved from https://www.metergroup.com/en/meter-environment/products/accupar-lp-80-canopy-interception-par-leaf-area-index

Microna Agriculture. (2023). Microna top-flow. Retrieved from https://micronaag.com/agricultural-lime-products/dry-products/hydrocal/

NOAA Climate Data Online (2023). https://www.ncei.noaa.gov/cdo-web/

Pietri, J. C., & Brookes, P. C. (2009). Substrate inputs and pH as factors controlling microbial biomass, activity and community structure in an arable soil. Soil Biol. Biochem., 41(7), 1396-1405. https://doi.org/10.1016/j.soilbio.2009.03.017

R Core Team. (2016). R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation. Retrieved from https://www.R-project.org/

Rousk, J., Brookes, P. C., & Bååth, E. (2010). The microbial PLFA composition as affected by pH in an arable soil. Soil Biol. Biochem., 42(3), 516-520. https://doi.org/10.1016/j.soilbio.2009.11.026

Ruhe, R. V. (1960). Elements of the soil landscape. Trans. 7th Int. Congr. Soil Sci., 4, pp. 165–170.

Schlatter, D. C., Yin, C., Hulbert, S., & Paulitz, T. C. (2020). Core rhizosphere microbiomes of dryland wheat are influenced by location and land use history. Appl. Environ. Microbiol., 86(5), e02135-19. https://doi.org/10.1128/AEM.02135-19

Thompson, W. H., McFarland, C., Brown, T. T., & Huggins, D. R. (2016). Agricultural Lime and Liming, Part 3: Aglime product selection and comparison calculator user guide. Washington State University. Retrieved from https://pubs.extension.wsu.edu/agricultural-lime-and-liming-part-3-aglime-product-selection-and-comparison-calculator-user-guide-soil-acidification-series

USDA-NRCS. (2023). Web Soil Survey. Retrieved from https://websoilsurvey.nrcs.usda.gov/app/

Yin, C., Schlatter, D. C., Kroese, D. R., Paulitz, T. C., & Hagerty, C. H. (2021a). Impacts of lime application on soil bacterial microbiome in dryland wheat soil in the Pacific Northwest. Appl. Soil Ecol., 168, 104113. https://doi.org/10.1016/j.apsoil.2021.104113

Yin, C., Schlatter, D. C., Kroese, D. R., Paulitz, T. C., & Hagerty, C. H. (2021b). Responses of soil fungal communities to lime application in wheat fields in the Pacific Northwest. Front. Microbiol., 12. https://doi.org/10.3389/fmicb.2021.576763

Zeileis, A., & Hothorn, T. (2002). Diagnostic checking in regression relationships. R News, 2(3), 7-10. Retrieved from https://CRAN.R-project.org/doc/Rnews/

Zhang, S., Zhu, Q., de Vries, W., Ros, G. H., Chen, X., Muneer, M. A.,... Wu, L. (2023). Effects of soil amendments on soil acidity and crop yields in acidic soils: A world-wide meta-analysis. J. Environ. Manag., 345, 118531. https://doi.org/10.1016/j.jenvman.2023.118531