Spatial Variation of Soil Physical Properties: A Precursor to Precision Tillage
J.P. Fulton, L.G. Wells, S.A. Shearer, and R.I. Barnhisel
Fulton, J.P., L.G. Wells, S.A. Shearer and R.I. Barnhisel. 1996. Spatial Variation of Soil Physical Properties: A Precursor to Precision Tillage. ASAE Paper No. 961002. International Meeting, Phoenix, Arizona. July 14-18.
Abstract
The spatial variation of bulk density and cone index in a Maury silt loam soil was assessed. Field data from a university research farm was used to evaluate the variability of bulk density and cone index to determine whether they were correlated. Bulk density measurements were collected using a Campbell-Pacific nuclear soil moisture/density gauge while cone index was measured using a hydraulically operated cone penetrometer. The results of the study showed little correlation between cone index values and dry density at a near field capacity soil water content. Furthermore, the analysis of site-specific tillage reveals the potential use of this management technique to save farmers money by only tilling those areas where dry density and cone index problems exist. No bulk density measurements equaled or exceeded 1.60 Mg/m3, which is presumed to limit root growth in silt loam soil. Conversely, cone index readings greater than 2 MPa, the nominal value associated with excessive soil compaction, were recorded at various points in the field. Potential fuel savings of site-specific deep tillage to remedy only compacted areas in a field grid versus subsoiling the entire field is presented.
Introduction
Site-specific farming has introduced a management practice by which farmers can begin analyzing and dealing with cropland variability. Site-specific farming is based upon the notion that fields used for agricultural production are not uniform. Variations of soil physical properties, nutrient levels and water content occur from field to field and within fields. These spatial variations result from many factors such as previous farming practices, topography of the land, and nutrient application inaccuracy. With site-specific technology farmers are adjusting application rates of pesticides and fertilizers along with plant populations in hopes of optimizing crop production across fields. This type of application practice eliminates overapplication of nutrients and overpopulation that can be detrimental to plant growth.
Minimum tillage crop production is receiving widespread application across the nation, owing to its potential for reducing energy costs, soil erosion and across cropland. However, machinery traffic can compact soil so as to reduce crop productivity and yield. Due to the nature of soil compaction and its variability within fields, farmers need to assess such variability by making soil physical property measurements such as soil bulk density or soil strength. Such measurements would allow farmers to determine the spatial variation of compaction and develop Global Information System (GIS) databases utilizing Global Positioning System (GPS) in their fields. If soil compaction can be measured and mapped, decisions may be made to specify tillage only in areas of fields where detrimental compaction exists. In addition, it might be desirable to prescribe the depth of tillage within certain areas of a field. Thus, within a field, tillage depth would vary according to the depth of soil compaction. Draft load and required energy could be reduced by tilling only to the depth of soil compaction and only in those areas containing compaction problems, thereby minimizing input cost.
The main focus of this study was assessing the variability of soil strength and bulk density within a field and determining if deep tillage might be required in a no-till situation to alleviate the existence of soil compaction detrimental to crop growth. The specific objectives were: 1)to measure soil bulk density and soil cone index at field grid points; 2) to determine if bulk density and cone index were correlated; and 3) to assess the need for deep tillage and the potential savings using precision technology.
Background
Current cropping practices produce a cycle between soil compaction produced by off-road equipment and the alleviation of this condition by means of tillage or natural processes such as freezing and thawing. The adverse effects of soil compaction on crop growth have been recognized for years. Bulk density and soil strength are two physical properties which quantify soil compaction. Buader et al. (1981) investigated the effect of four continuous tillage systems on mechanical impedance of clay loam soil. They reported that increases in bulk density are correlated with increases in penetration resistance. Veihmeyer and Hendrickson (1948) reported that soil compaction reduces root growth. Such soil conditions can decrease crop yields, a result which is certainly undesirable to farmers.
Farmers want to be productive by enhancing plant growth and maximizing yields. Philips and Kirkland (1962) and Morris (1975) reported corn yield reductions of 10 to 22 percent due to compaction. For each 1kg/m3 increase in bulk density, a decrease in maize grain yields of 18% relative to the yield on a noncompacted plot (Canarache et al., 1984). Increased soil compaction can reduce yields in potatoes of up to 22 percent (Saini and Lantagne, 1974) and decrease wheat growth (Feldman and Domier, 1970). These results illustrate the potential for compaction to depress crop yields. Extremely dense soil impedes root growth and thereby limits water consumption of plants. Thus, soil compaction must be managed in order to keep its detrimental effects to a minimum.
The level of compaction which requires tillage for a given soil type is not well understood. No generally accepted rule of thumb exists which states that a certain bulk density or penetrometer strength limits plant productivity. However, some studies have been conducted which address these two parameters in predicting detrimental effects to plant growth. Bowen (1981) suggested a general rule (with many exceptions)that bulk density measurements of 1.55, 1.65, 1.80 and 1.85 Mg/m3 can impede root growth and thus will reduce crop yields on clay loams, silt loams, fine sandy loams, and loamy fine sands, respectively. Bulk density greater than 1.2 Mg/m3 for clay soil, 1.6 Mg/m3 for loam soil, and 1.8 Mg/m3 for sandy loam adversely affected the root growth of rice (Kar et al., 1976). Singh et al. (1992) proposed a bulk density less than or equal to 1.3 Mg/m3 in any soil as nonlimiting to crop growth. However, Singh et al. (1992) stated that due to the lack of research literature, the maximum value of bulk density which may be considered unusable by plants is 2.1 Mg/m3 in any type of soil.
Soil strength is an indicator of how easily roots can penetrate soil. Cone index is a measure of soil strength and is measured using a penetrometer. The magnitude of mechanical impedance to root penetration which decreases plant growth is also unknown. Ehlers et al (1983) stated that the penetrometer resistance limiting to oats was 3.6 MPa in tilled Ap horizon, but 4.6 to 5.1 MPa in the untilled Ap horizon and in the subsoil. The limiting penetrometer resistance depends upon the soil conditions and characteristics and the crop of interest. Ayers and Perumpral (1982) pointed out that dry density had a considerable influence on cone index at low moisture contents for soils containing a certain percentage of clay. Cone index became less dependent on dry density at higher moisture contents. Sojka et al (1990) studied the effect of penetrometer resistance on sunflowers. A soil strength corresponding to a penetrometer resistance of 2MPa produces some root restriction and a resistance of 3 MPa creates a total barrier to root elongation. A maximum root growth pressure for citrus is 1.5 MPa. Murdock et al (1995) suggested a penetrometer reading of 2.07 MPa (300 psi) as indicative of severe compaction for Kentucky soils. A penetrometer measurement of 2.0 MPa generally regarded as sufficient to hinder the growth and development of crops. However, precise cone index levels which limit plant growth for specific soil types have been rarely documented.
The development of precision farming has risen from the recent interest in increasing productivity of land (Evans and Han, 1994). Site-specific farming utilizes GPS and GIS systems connected to automatic controllers which regulate field inputs. Future research and development may make site-specific control applications economically viable in agricultural production. As technology advances, the cost of GPS/GIS systems will decrease and the feasibly of such systems to farmers will increase. These systems may also perform multiple tasks such as controlling the inputs of planters and sprayers, allowing the sharing of equipment costs between different applications. Thus, farmers can use one controller for various applications. Inputs can usually be decreased by varying the rate of application to meet requirements which vary spatially over fields and therefore save farmers money. Lal et al (1993) successfully demonstrated that GIS can be combined with site-specific models for regional planning and productivity analysis.
Thus, precision farming presents the possibility of prescribing only certain regions and depth of tillage within agricultural fields by mapping the depth and magnitude of soil compaction. A GIS database can be developed in which a farmer can determine the depth of tillage to reduce compaction and save money by reducing fuel consumption.
Experimental Methods
This study was conducted between April 9 and 17, 1996 on a Maury silt-loam soil located at the Woodford County Research Farm of the University of Kentucky near Versailles, Kentucky. A 7.06 hectare field was selected as the site for performing the investigation. Before converting it into crop land in 1992, the field existed as a pasture. Since then, only no-till crop production has occurred with a yearly corn/wheat/soybean rotation. The field was subdivided into a 30.5m x 30.5 m (100 ft x 100 ft) grid system which created seventy six individual grid cells for collecting physical properties on a site-specific basis. Bulk density, cone index and moisture content were measured at the center of each grid cell. Nominal depths of 150 mm, 300 mm, 450 mm and 600 mm (6", 12", 18" and 24") were selected for making these measurements.
A Campbell-Pacific nuclear soil moisture/density gauge was used to measure dry density, wet bulk density and moisture content. Luo and Wells (1992) described a dual-probe density gauge in detail and its use in field investigations. Gamma ray attenuation measures the soil bulk density between the two probes while neutron thermalization is used for measuring water content. Various depths between 5.1 cm and 91.4 cm can be selected in 5.1 cm increments. Vertical access holes (2.54 cm in diameter and 30.5 cm apart) for the dual probe gauge were made using a tractor mounted, hydraulically operated device. One probe contains the gamma and neutron sources while the other probe contains the detector for counting the gamma rays and each must be set at the same depth during a test.
A hydraulically powered cone penetrometer was used to measure the soil strength or cone index on each grid cell. Two tests were performed on each grid cell near the location where the bulk density readings were taken. The penetration resistance or force was recorded as a function of depth and stored on a laptop computer. In return, the resistance force recorded from a load cell divided by the base area of the cone determines the cone index value. The rate of penetration was 1829 mm/min (72 in./min), to a depth of 122 cm. From the recorded data, the cone index value was determined by averaging the penetration pressure from the two tests over an interval of ± 1.27 cm from the desired nominal depth.
Results and Analysis
Three dimensional plots of dry bulk density and penetrometer measurements were created at each depth (150, 300, 450 and 600 mm) to visually assess variability across the 7.1 hectare field. All plots were developed in the program, Surfer, with kriging performed to smooth the data. Figure 1 presents the dry bulk density at 300 mm while Figure 2 displays the cone index value at the same depth. The moisture content for this field was near field capacity (approximately 28-30 percent, dry basis) during data collection. Table 1 summarizes the statistical data for each of the soil physical properties at the given depth. The sharp peak present in both plots, points out the existence of bedrock in that area of the field. This particular area is located on a hillside which has probably been eroded by water over the years giving rise to the shallow bedrock. The density of bedrock was assumed to be 2.65 mg/m3. The plots show that these two soil properties vary over the field at the 300 mm single depth, and such variation was also evident at the other depths. Thus, site-specific tillage applications based upon localized dry bulk density or cone penetrometer measurements may be feasible and economically viable to farmers.
When comparing the dry bulk density and penetrometer plots at each depth, the bedrock typically shows up at the same location. However, at certain depths, the penetrometer and dry bulk density plots are not similar in form. Figure 2 exhibits a gradual increase in cone index from the 200 m mark to the end of the field (500 m) where Figure 1 tends to remain more constant over the field. The two figures display similarities, with peaks and valleys occurring around the center of the field on the two plots. It appears that certain regions are very similar when comparing cone index and dry density, but others show opposing results. Overall, the plots tend to demonstrate that dry bulk density and cone index are possibly associated at the 300 and 450 mm depths but not at 150 and 600 mm.
A linear regression analysis was performed at the four different depths (150, 300, 450 and 600 mm) to compare cone index and dry bulk density. Here, the focus was to determine whether the two soil properties are correlated. Data was entered into Quattro Pro according to site, field position and depth. A linear regression was then executed at each depth with the results exhibited in Table 2. The highest r2 was .59, occurring at 450 mm, while the other depths (150, 300, and 600 mm) indicated virtually no correlation between cone index and dry bulk density. This was somewhat surprising since Ayers and Perumpral (1982) and other investigators have documented good correlation between bulk density and cone index. At the relative high water content measure in this study, increased soil strength associated with internal soil cementation is indicate.
Site-specific tillage requirements for this field were assessed by selecting a critical dry bulk density and cone index. A dry bulk density of 1.60 Mg/m3 was selected as the threshold level which would require subsoiling or deep tillage to the depth at which it was measured. Kar et al (1976) suggested a bulk density greater than 1.6 Mg/m3 inhibited the root growth of rice for a loam soil while Bowen (1981) indicated a bulk density of 1.65 Mg/m3 impeded root growth in silt loam soils. No value of dry bulk density exceeding 1.60 Mg/m3 was measured in this study, thus no deep tillage was indicated based upon the root limiting bulk density.
A cone index of 2.0 MPa was chosen as the threshold value based upon the previous cited literature (Sojka et al., 1990 and Murdock et al., 1995). As explained in previous work by investigators, a cone index of 2MPa could correspond to soil conditions which could hinder root growth and plant development. This value is approximately 1.45 times higher than that stated in the ASAE Standard D230.4 (ASAE, 1985) for hard soil conditions. The cone index data was analyzed to determine the locations and depths where measurements exceeded 2 MPa. Thirty nine (39) measurements exceeded the threshold with 34 occurring at the 450 mm depth. Figure 3 illustrates the resulting site-specific tillage depth distribution or grid. Low areas indicate the depth and location for subsoiling. An analysis was performed to determine the potential fuel savings which could be achieved using site-specific deep tillage versus subsoiling of the entire field. ASAE (1995) Standard D230.4 was followed to estimate subsoiler draft force versus tillage depth. Fuel consumption was then estimated assuming engine thermal efficiency of 55% and tractive efficiency of 70% to be 4.6 L/ha and 6.8 L/ha for the subsoiling depths of 300 and 450 mm, respectively.
The site-specific tillage depths indicated in Figure 3 were then used to compute an alternative estimate of fuel required for deep tillage. Based on the 30.5 x 30.5 m grid size used in the study, fuel use could be reduced to 3.4 L/ha, which is approximately one half that required for subsoiling the entire field to a depth of 450 mm. Results suggest that site-specific tillage should be considered in conjunction with development of a physical property data bases for cropland.
Conclusions
The following conclusions were made from this study of assessing the variability of cone index and density:
References
ASAE Standard No. D230.4. 1985. Agricultural Engineering Yearbook, ASAE, St. Joseph, MI 49085.
Ayers, P. D., and J.V. Perumpral. 1982. Moisture and Density Effect on Cone Index. Transactions of the ASAE.
Bauder, J. W., G. W. Randall and J. B. Swan. 1981. Effect of four continuos tillage systems on mechanical impedance of a clay loam soil. Soil Sci. Soc. Am. J. 45:802-806.
Bowen, H. D. 1981. Alleviating mechanical impedance. In Modifying the Root Environment to Reduce Crop Stress, eds. G. F. Arkin and H. M. Taylor, 21-57. St. Joseph, MI: ASAE.
Canarache, A., I. Colibas, M. Colibas, I. Horobeanu, V. Patru, H. Simota and T. Tradafirescu. 1984. Effect of induced compaction by wheeled traffic on soil physical properties and yield of maize in Romainia. Soil Tillage Res. 4:199-213.
Ehlers, W., V. Fopke, F. Hesse and W. Bohm. 1983. Penetration resistance and root growth of oats in tilled and untilled loam soil. Soil Tillage Res. 3:261-275.
Evans, R. G., and S. Han. 1994. Field-Scale GIS Soil Database Creation. ASAE Paper No. 94- 3078. St. Joseph, MI 49085-9659.
Feldman and Domier. 1970. Wheeled traffic effects on soil compaction and growth of wheat. Can. Agr. Eng. 12(1): 8-11.
Kar, S., S. B. Varade, T. K. Subramanyan and B. P. Ghildyal. 1976. Soil physical conditions affecting rice root growth: Bulk density and submerged soil temperature regime effects. Agron. J. 68:23- 26.
Lal, H., G. Hoogenboom, J. P. Calixte, J. W. Jones, and F. H. Beinroth. 1993. Using crop simulation models and GIS for regional productivity analysis. Transactions of the ASAE 36(1):175-184.
Morris, D. T. 1975. Interrelationship among soil bulk density, soil moisture and the growth and development of corn. M. Sci.. Thesis, University of Guelph, Guelph, Onterio. 77 pp.
Murdock, L., Gray, T., Higgins, F., and Wells, K. 1995. Soil Compaction in Kentucky. Publication of the Cooperative Extension Service. University of Kentucky, College of Agriculture.
Phillips , R. E. And D. Kirkham. 1962. Soil compaction in the field and corn growth. Agron. J. 54;29- 34.
Saini, G. R., and H. M. Lantagne. 1974 Le Tassement du sol. Actuality Agricole 34(2):11-13.
Singh, K.K., T.S. Colvin, D.C. Erbach and A.Q. Mughal. 1992. Tilth Index: An approach to quantifying soil tilth. Transactions of the ASAE 35(6): 1777-1785.
Sojka, R. E. , W. J. Busscher, D. T. Gooden and W. H. Morrison. 1990. Subsoiling for sunflower production in the southeast Coastal Plains. Soil Sci.. Soc. Am. J. 54(4):1107-1112.
Veihmeyer, F. J. And A. H. Hendrickson. 1948. Soil Density and root penetration. Soil Sci.. 65:487- 93.
Wells, L. G. And X. Lou. 1992. Evaluation of gamma ray attenuation for measuring soil bulk density, II: field investigation. Tans. of ASAE, 35(1): 27-32.