MLRA 121: Kentucky Bluegrass E. Perfect. T. Karathanasis, and G. Haszler University of Kentucky
This major land resource area (MLRA 121) occupies 2.95 million ha (7.28 million acres) in north-central Kentucky, south-eastern Indiana, and southern Ohio.
Climate The climate is temperate, humid, continental type. Winters are characterized by short cold spells, frequent sharp changes in temperature, and fairly high humidity. Summers are hot and humid. Relative humidity is generally higher at night, and the average value at dawn is about 80% as compared to 60% in midafternoon. Precipitation is usually well distributed throughout the year (Fig. 1), with a mean annual total of 1,132 mm. Brief periods of drought occur in summer, while periods of excess moisture often occur in winter and spring. Thunderstorms number about 47 each year, 25 of which occur in summer. Average seasonal snowfall is 450 mm. The average annual temperature is 13°C with an average frost-free period of 180 days. The prevailing wind is from the south
Geology and Topography Elevation ranges from about 200 m on the Ohio River flood plain to about 300 m on the higher ridges around Lexington. The topography ranges from highly dissected hills that have local relief of about 50-100 m to undulating broad upland plains that have local relief of about 25 m. The Bluegrass physiographic region of Kentucky lies between the Ohio River valley in the north and the Knobs physiographic region in the south (Fig. 2). It can be divided into three distinct areas - the Inner Bluegrass, the Hills of the Bluegrass, and the Outer Bluegrass.
The Inner Bluegrass occupies a circle in the middle of the Bluegrass with its center near Lexington. It contains the oldest rocks exposed in Kentucky: dominantly thick-bedded limestones of Middle Ordovician age that were raised to their present position by uplift along the Cincinnati Arch. It is characterized by gently rolling terrain and a thick, fertile, residual soil. Some of the limestone strata are phosphatic, and weathering of these rocks has enhanced the fertility of the soil. The gently rolling surface is modified by some karst development such as sinkholes, sinking streams, and springs. Although karst features are not as numerous or well developed as in the Mississippian Plateaus of the Mammoth Cave area, some springs were the sites of early settlements such as Georgetown, Harrodsburg, and Lexington.
The Inner Bluegrass physiographic region is underlain by limestone of the Cynthiana, Lexington, and High Bridge Formations. The Cynthiana Formation is mainly limestone but is interbedded with thin layers of calcareous shale. The High Bridge formation is along the Kentucky River gorge. It is massive limestone, the oldest exposed rock in the state. The Lexington Formation underlies most of the Inner Bluegrass area. It is thinbedded, shaley limestone that is mostly phosphatic.
There are also several fault systems in the region. Most of the area is an old eroded peneplain. In steep areas, the exposed rocks are less resistant to weathering and have cut deep narrow valleys that have fairly long, steep, and sharp-crested ridges. Limestone bluffs occur where short tributary streams flow through gorges to the Kentucky River. Most surface water in the Inner Bluegrass eventually drains into the Kentucky River.
Rock formations become progressively younger towards the boundaries of the Inner Bluegrass region. Gentle regional inclination of the strata carries the older rocks beneath the surface and younger ones appear. The rocks exposed at the surface at Lexington are several thousand feet deep at Madisonville and Pikeville.
The Hills of the Bluegrass lie in a circle separating the Inner and Outer Bluegrass and constitute a well-dissected plateau area characterized by narrow winding ridges and valleys. Hillsides generally slope 20 to 30%, and surface rock is frequently present. Rock formations are Late Ordovician in age and differ markedly in lithology from Inner Bluegrass rocks. Calcareous shale, siltstone, and limestone of the Eden and Garrard Formations underlie the Hills of the Bluegrass physiographic area. Many of the formations contain interbedded shales and limestones, the youngest exposed rock in the area. Consequently, they are softer and less resistant to erosion. Stream erosion has cut a multitude of valleys. Hills and steep slopes dominate the landscape, and little flat land is present. This part of the Bluegrass Region has sometimes been called the Eden Hills Shale Belt.
The Outer Bluegrass occurs in three discontinuous arcs around the outer perimeter of the Bluegrass region of Kentucky. It differs from the Inner Bluegrass in that it is more rolling, but has fewer hills and more flat land than the Hills of the Bluegrass. In the Louisville and Bardstown areas, the topography developed on the carbonate rocks of Silurian and Devonian ages appears similar to parts of the Bluegrass Region and is frequently included with it. Here, again, the geology of the area affected local topographic development. A conspicuous spring zone occurs at the base of some of the Silurian dolomites, and some of these springs were focal points of early distilleries in this part of Kentucky. The falls of the Ohio River with its unique and famous coral beds is situated at the outer edge of this region.
The outer edges of the Bluegrass region are characteristically lowlands with little relief being developed on shales of Mississippian and Devonian ages. The outcrop area of Kentucky’s oil-shale deposits is in this part of the state. These black shales, which form one of the most easily recognized rock units in Kentucky, are of additional geologic interest because they are the reservoir and source rocks for much of the natural gas produced from the Big Sandy Gas Field in eastern Kentucky.
Landuse Most of MLRA 121 has been cleared of forest and at present about 45% of the land area is in pasture. Bluegrass (which is not native to the area), orchard grass, and fescue in mixtures with clover or bluegrass alone, are the main pasture species. About 30% of the area is in cropland, although acreage varies widely from county to county depending mainly on the topography. Farms are mostly small to medium size. Burley tobacco is the main cash crop. Corn, barley, and wheat are also grown. Red clover grass mixtures, lespedeza, and alfalfa are the principal hay crops. Livestock uses most of the grain, pasture, and hay produced. The Bluegrass Region is famous for high quality livestock, especially racehorses. There are many horse farms in the region. Straw from small-grain crops is used on horse farms for bedding. The production of beef cattle is also significant. Dairy cattle, sheep, and hogs are produced in fewer numbers. There are still several small wooded areas scattered throughout the Bluegrass. The larger areas are along river bluffs and on steeper slopes near major creeks. The forest vegetation is mixed hardwoods. Several varieties of oak, hackberry, black walnut, black cherry, black locust, white ash, American elm and Kentucky coffee tree are important species. Eastern red cedar is dominant on the drier slopes and on abandoned farmland. Urbanization is minor except around Louisville, Cincinnati, and Lexington.
Water Resources Water is present as both surface and ground water. Surface water occurs as rivers, streams, ponds, reservoirs, and wetlands. Ground water occurs in the pore spaces within rocks and alluvium, in fractures, and in solution openings or conduits in areas underlain by carbonate rocks (e.g. limestone). The median depth to ground water in the Bluegrass Region is about 6 m. Surface water often enters or returns to the ground water system through sinkholes and cave openings. Surface and ground water supplies are susceptible to pollution from natural, agricultural, and industrial sources. Naturally occurring substances such as iron, manganese, barium, fluoride, hydrogen sulfide, and salt may be present at objectionable levels. Bacteria from sewage, septic tanks, and animal wastes are a common problem. High levels of nitrate-nitrogen, pesticides, and organic chemicals threaten water supplies in some areas.
Soils Soils in the Bluegrass Region are mainly Hapludalfs and Paleudalfs. They are fine to moderately-fine textured, and have a mesic temperature regime, a udic moisture regime and mixed mineralogy. The major soil associations are Eden, Lowell-Fairmount, and Maury-McAfee (Table 1). Soil physical properties for the most commonly occurring soil series are given in Table 2. STATSGO soils are shown in Fig. 3.
Fig. 3. STATSGO soils of the Kentucky Bluegrass in MLRA 121.
The Eden association occurs on highly dissected hills throughout the Bluegrass Region. The soils are moderately deep and well drained. Eden and Culleoka soils (Udalfs) commonly occur over the Eden and Garrard Formations. They were formed in clayey residuum from thinly bedded limestone, shale, and siltstone. Soils of the Lowell-Fairmount association occur on rolling ridge tops and side slopes with many dissections by small streams in the Outer Bluegrass Region. They are deep to shallow, well drained, and formed in clayey residuum from limestone or limestone interbedded with shale. Lowell and Faywood soils (Udalfs) overlie the Cynthiana Formation. The Maury-McAfee association consists of deep to moderately deep, well drained, soils formed in clayey limestone residuum. They occur on upland plains with some karst areas and dissections in the Inner Bluegrass Region. Maury and McAfee soils (Udalfs), generally medium to high in phosphate, commonly occur over the Lexington Formation. Other minor soils in the area include Loradale and Fairmount (Udolls), Mercer, Salvisa, and Donerail (Udalfs) in upland landscapes, and Huntington and Woolper (Udolls), Melvin (Aquents), Nolin (Ochrepts), Elk, and Ashton (Udalfs) on lowland and floodplain positions.
Transport of Water and Contaminants: Case Studies Water One of the unique characteristics of some soils in the Bluegrass Region is an unusually high total porosity for their textural classification, which promotes rapid, preferential flow of water and contaminants. Cracks form between ped faces in the blocky and angular blocky structures common to these soils. Old root channels and worm holes also contribute to the total porosity. Porosity is strongly influenced by landuse; for example, values of porosity ranged from 0.36 to 0.52 in the surface 10 cm of a Maury soil subjected to different tillage management practices (Fig. 4). As total porosity increases, both the frequency and the number of connections between macropores can be expected to increase. As a result, saturated hydraulic conductivity increases as a power law function of porosity (Fig. 4). Even though soils in this MLRA are normally only saturated for relatively short periods of time, saturated hydraulic conductivity is still an important measurement since a large proportion of the total annual water movement through the vadose zone occurs during these events. This is because the ability of the soil to conduct water declines precipitously as the profile drains.
The spatial variability of water fluxes through large undisturbed blocks of Maury and Nolin soils has been documented by Quisenberry et al. (1994) and Phillips et al. (1995). These authors showed that a large fraction of the water moves through a relatively small areal percentage of the pore space, regardless of application rate. The non-random distribution of fluxes was interpreted as evidence of preferential flow through a continuous network of tortuous and interconnected macropores. The extent of preferential flow in such experiments can be evaluated by plotting cumulative effluent volume vs. cumulative pore area (Bowman, et al., 1994). Theoretically, if leaching were homogeneous, the resulting plot would be linear as indicated by the straight line in Fig. 5. However, if preferential flow occurs, the response will be curvilinear as indicated by the observed data (open circles) in Fig. 6. Note that over 90% of the effluent drained through less than 50% of the pore area. Additional information on the hydrology of the Maury soil can be found in Cassel (1985) and Römkens et al. (1985).
Non-reactive Solutes Perfect et al. (1998) measured steady-state transport of water, chloride, and bacteria through intact blocks of Maury soil under partially saturated conditions. Three replicate blocks from each of two landuse treatments, conventional-till (disk) corn production, and grass pasture were investigated. The blocks were mounted on a vacuum chamber maintained at –2.0 kPa. Water was supplied to the soil surface with a rainfall simulator at a rate of 1 cm/hr. Volumetric water contents and bulk electrical conductivities were measured at three depths, 5, 15, and 25 cm using time-domain reflectometry (TDR). When steady-state flow was achieved, the concentrations of chloride and E. coli bacteria in the water supply were increased in a stepwise fashion, and the resulting breakthrough curves measured over time. The chloride breakthrough curves (calculated from the bulk electrical conductivity data) were fitted to the convection dispersion equation (CDE) using CXTFIT (Toride et al., 1995). The parameters of the CDE are the dispersion coefficient (D) and the average pore water velocity (v). These parameters were combined to give the dispersivity defined as D/v. The goodness of fit for these analyses (R2 generally > 0.9) indicates the CDE was applicable to the flow conditions in these experiments.
Figure 6 shows typical chloride breakthrough curves measured with the TDR probes as a function of depth. Being closest to the source of the high-conductivity solution, the 5-cm TDR probe was the first to increase after the step increase in chloride concentration. Subsequent increases were observed at the 15- and 25-cm depths. The curves became more S-shaped as the solute moved further from the source and deeper into the soil.
Table 3 presents the results of the CXTFIT parameter estimation procedure for the TDR data. Dispersivities ranged from just above 1 to nearly 10 cm and decreased with depth. There was no evidence of any interaction between depth and landuse. Regression analysis revealed a significant effect of volumetric water content on dispersivity, with dispersivity decreasing with increasing water content. Lower water contents lead to more tortuous flow paths and a broadening of the velocity distribution. Thus, any structural effects on solute dispersion in unsaturated soils subjected to initial and boundary conditions similar to those in these experiments are likely to be indirect, and can probably be related to differences in water content at a given flow rate produced by differences in pore-size distribution.
Additional chloride breakthrough curves can be found in McMahon and Thomas (1974) and White et al. (1984) for disturbed (sieved and repacked) and undisturbed samples of Maury and Eden soils. These authors also measured break-through of tritiated water. Breakthrough curves for the disturbed samples were S-shaped and typical of low dispersivity, while those for the undisturbed samples were asymmetrical and indicative of high dispersivity. The apparent velocity of Cl was greater than that for 3HOH, which is consistent with anion exclusion by the soil matrix. Initial breakthrough appeared to be more rapid in the Eden soil than in the Maury soil, suggesting greater structural development for the silty clay loam than for the silt loam. Initial breakthrough was between 1.5 and 2 times faster in the undisturbed samples than in the disturbed samples for both soils, despite higher bulk densities for the undisturbed samples (McMahon and Thomas, 1974). The shape of the breakthrough curve was related to the application rate and initial water content. For the undisturbed Maury soil, the time required to reach C/Co =0.5 increased with decreasing application rate (Table 1 in White et al., 1984). The breakthrough curves became more asymmetric, as indicated by high C/Co values at early times (Fig. 2 in White et al., 1984), as the initial water content decreased. This trend indicates increasing dispersivity with decreasing initial water content, which is consistent with the results of Perfect et al. (1998).
Bacteria A typical breakthrough curve for E. coli bacteria in Maury soil under partially saturated conditions is given in Fig. 7. Additional bacterial breakthrough curves for Maury soil can be found in Smith et al. (1985). Clearly, the response to a step increase in bacteria is very different to that for Cl. Most of the bacteria are trapped within the soil matrix, so the maximum relative concentration in the effluent is very low (approximately 10-3 ). Instead of increasing from zero to 1 over time, the C/Co for bacteria remains relatively constant with increasing effluent volume. However, this constant value is established very early on, suggesting that transport occurs through continuous macropores. Further evidence of preferential transport of bacteria can be seen in Table 3 of Smith et al. (1985), which shows that sieving and repacking the soil, destroying macropore continuity, can reduce C/C o by up to two orders of magnitude.
Herbicides and Heavy Metals Seta and Karathanasis (1996a,b; 1997a,b) have conducted a series of intact soil column experiments with Maury (Udalf) and Loradale (Udoll) soils to assess the role of water dispersible soil colloids, with diverse physicochemical and mineralogical composition, in co-transporting selected herbicides and heavy metals through the upper solum. Another objective of their experiments was to identify the colloid, soil, and solution properties facilitating or hindering contaminant transportability.
Water dispersible colloids were fractionated from Bt horizons of six soil series with montmorillonitic, mixed, illitic, and kaolinitic mineralogy. Colloid suspensions of 300 mg/L were mixed with 2 mg/L of atrazine or metolachlor, or 10 mg/L of copper (Cu), zinc (Zn), or lead (Pb) and introduced into duplicate undisturbed soil columns at constant flux. The eluents were collected and analyzed periodically for colloid, herbicide, and metal concentrations. Batch equilibrium isotherm experiments were also conducted to evaluate the herbicide/metal adsorption capacity of the colloids.
The transport of both herbicides was enhanced in the presence of water-dispersible soil colloids. The observed increase was 2 to 18% for atrazine and 8 to 30% for metolachlor. The greater increase in metolachlor transport is probably due to its 4- to 7-fold greater affinity for colloid surfaces as indicated from equilibrium adsorption isotherms. However, a significant portion of the eluted herbicide load was mainly due to exclusion of soluble herbicide species from matrix surface sites blocked by colloids rather than actual transport of colloid-bound herbicide. Therefore, in this process, the colloids appear to act mostly as facilitators rather than as carriers. The amount of extra herbicide transported in the presence of colloids varied with colloid type and transportability (Fig. 8). Colloids with higher surface area, pH, electrophoretic mobility, organic carbon, and smectite content showed greater potential for co-transporting herbicides. In contrast, high amounts of kaolinite, iron (Fe), and Al hydroxides inhibited co-transportability. The type of soil column also influenced herbicide recoveries in the eluent. Maury soil columns with a better macropore system (high hydraulic conductivity) and lower surface area enhanced herbicide co-transportability.
Metal transport was also enhanced in the presence of water dispersible soil colloids by an average of 10- to 30-fold, but the increase was colloid and metal specific (Figs. 9 and 10). Cu and Zn co-transportability in the presence of colloids was greater than that of the controls, 2 to 150 and 5 to 30 times, respectively. For Pb, there was essentially no (or very little) elution of metals in the absence of colloids, suggesting nearly complete sorption by the column matrix. In the presence of colloids, Pb transportability ranged from 8 to 3,000 times that of the control. In most cases, nearly 90% of the increase in metal transport in the presence of colloids was due to colloid-bound metal phases and the remaining due to colloid-metal-matrix exclusion processes.
Increasing the metal concentration in the colloid suspension beyond 10 mg/L hindered metal-colloid transport through the columns due to coagulation, flocculation, flow retardation, and pore clogging. Although colloids with high amounts of 2:1 minerals and organic carbon content generally showed increased metal sorption and co-transportability, the quantitative correlations were not always consistent. In some cases, increased soil organic carbon contents (Loradale soil) appeared to overshadow the effects of macroporosity (Maury soil) on the transport of specific metals.
Karathanasis (1999,2000) also investigated the potential of water-dispersible soil colloids to desorb Pb and contaminated soil matrix surfaces and co-transport it to groundwater. The study involved leaching experiments using intact soil columns contaminated with Pb and colloid suspensions of different mineralogical composition, with deionized water used as a control. The soil columns represented upper solum horizons of Maury and Loradale soils with contrasting macroporosity and organic carbon contents. The soil colloids were fractionated from low ionic strength Bt horizons of Alfisols with montmorillonitic, mixed, and illitic mineralogy and variable physicochemical and surface charge properties. The results indicated a sharp decrease, to near zero, of Pb desorbed by deionized water-flushing solutions after 3 pore volumes of leaching, but a continuous desorption and transport of Pb in the presence of colloids. The colloid-induced desorption and remobilization of Pb was in the range of 10 to 60% of the initial eluent Pb concentration, and it was 20 to 30% greater through the Maury than the Loradale soil columns. Colloids with high surface charge (montmorillonitic) and organic carbon content showed a greater Pb desorption and transport potential, but the amount of remobilized Pb was the result of contributions by both ion exchange and physical exclusion processes.
Barton and Karathanasis (1998) also evaluated the role of soil colloids and their potential to co-transport of contaminants through large soil monoliths under field conditions. The monoliths were formed by hydraulically driving steel pipe sections (50-cm diameter x 80-cm length) into Maury and Loradale soils with contrasting physicochemical characteristics. Water dispersible colloids were fractionated from the same soils, spiked with atrazine and Zn, and introduced into the monoliths at six-hour intervals using 500-ml pulse applications. Atrazine and Zn solutions with no colloid were applied to monoliths from each series to represent the control treatment. Eluents from the monoliths were collected and analyzed periodically for colloid, atrazine, and Zn concentration in the soluble and sorbed phase.
Colloid, atrazine, and Zn recoveries varied greatly with respect to soil type. Colloid recovery in the Loradale soil consistently exceeded 20% and approached 60%, while the Maury soil rarely exhibited recoveries greater than 5%. The presence of colloid in the Loradale eluent enhanced atrazine transport by 40% and Zn transport by 50% over that of the control treatment. In the Maury soil, however, colloid-mediated transport of atrazine was observed only when colloid recoveries exceeded 5%, and the transport of Zn was no different than that observed for the control. Apparently, retardation of atrazine and Zn within the soil matrix increased as the transportability of the Maury colloid was deterred. A settling-rate experiment, performed at varying pH and conductivity (EC) levels, revealed that Maury colloids may flocculate under the pH and EC conditions developed during the leaching process due to resolubilization of Mn, while Loradale colloids remain stable.
The results of these studies confirm that the currently widely used two-phase transport models may significantly underestimate herbicide and heavy metal transport through the vadose zone. Therefore, incorporation of the mobile colloid phase is necessary in order to improve predictions.
Literature Cited Barton, C.D. and A.D. Karathanasis. 1998. Colloid-facilitated transport of atrazine and Zn through large soil monoliths Abstracts, American Society of Agronomy Meeting, Baltimore, Maryland.
Bowman, B.T., R.R. Brunke, W.D. Reynolds, and G.J. Wall. 1994. Rainfall simulator-grid lysimeter system for solute transport studies using large, intact soil blocks. J. Environ. Qual. 23:815-822.
Cassel, D.K. (ed). 1985. Physical characteristics of soils of the southern region - summary of in situ unsaturated hydraulic conductivity. p. 143. Southern Cooperative Series Bull. 303. Regional Research Project S-124. North Carolina State Univ., Raleigh.
Karathanasis, A.D. 1999. Subsurface migration of Cu and Zn mediated by soil colloids. Soil Sci. Soc. Am. J. 63:830-838.
Karathanasis, A.D. 2000. Colloid-mediated transport of Pb through soil porous media. Environ. Science (In press).
McMahon, M.A. and G.W. Thomas. 1974. Chloride and tritiated water flow in disturbed and undisturbed soil cores. Soil Sci. Soc. Am. J. 38:727-732.
Perfect, E., M.S. Coyne, M.C. Sukop, G.R. Haszler, V.L. Quisenberry, and L. Bejat. 1998. Solute and bacterial transport through partially saturated intact soil blocks. p. 46. Kentucky Water Resources Research Institute Report, University of Kentucky, Lexington.
Phillips, R.E., V.L. Quisenberry, and J.M. Zeleznik. 1995. Water and solute movement in an undisturbed, macroporous column: extraction pressure effects. Soil Sci. Soc. Am. J. 59:707-712.
Quisenberry, V.L., R.E. Phillips, and J.M. Zeleznik. 1994. Spatial distribution of water and chloride macropore flow in a well-structured soil. Soil Sci. Soc. Am. J. 58:1294-1300.
Römkens, M.J.M., H.M. Selim, R.E. Phillips, and F.D. Whisler. 1985. Physical characteristics of soils in the southern region - Vicksburg, Memphis, Maury series. Southern Cooperative Series Bull. 266, Regional Research Project S-124, Mississippi Agric. and For. Exp. Stn. Mississippi State Univ., Mississippi State.
Seta, A.K. and A.D. Karathanasis. 1996a. Colloid-facilitated transport of Metolachlor through intact soil columns. J. Environ. Sci. Health B31(5):949-968.
Seta, A.K. and A.D. Karathanasis. 1996b. Water dispersible colloids and factors influencing their dispersibility from soil aggregates. Geoderma 74:255-266.
Seta, A.K. and A.D. Karathanasis. 1997a. Atrazine adsorption by soil colloids and co-transport through subsurface envi-ronments. Soil Sci. Soc. Am. J. 61:612-617.
Seta, A.K. and A.D. Karathanasis. 1997b. Stability and transportability of water dispersible soil colloids. Soil Sci. Soc. Am. J. 61:604-611.
Smith, M.S., G.W. Thomas, R.E. White, and D. Ritonga. 1985. Transport of Escherichia coli through intact and disturbed soil columns. J. Environ. Quality. 14:87-91.
Toride, N., F.J. Leij, and M. Th. van Genuchten. 1995. The CXTFIT code for estimating transport parameters from laboratory or field tracer experiments. Version 2. Research Report No. 137. U.S. Salinity Laboratory. Agricultural Research Service. U.S. Department of Agriculture. Riverside, California.
White, R.E., G.W. Thomas, and M.S. Smith. 1984. Modelling water flow through undisturbed soil cores using a transfer function model derived from 3HOH and Cl transport. J. Soil Sci. 35:159-168.
Return to Home Page Home Page SAAESD Electronic document prepared by: D.L. Nofziger, Oklahoma State University Email address: david.nofziger@okstate.edu