EARTHWORM MACROPORES AND PREFERENTIAL TRANSPORT IN A LONG-TERM MANUREAPPLIED TYPIC HAPLUDALF

Deep burrowing earthworm species have been found to be present in soil s with a history of manure application. This study was designed to qua ntity the effects of long-term application of liquid dairy manure and inorganic fertilizer on the distribution of earthworm macropores and i n turn on the preferential transport of water and tracer through a typ ical soil of the karst area of the upper mid-western USA. Large (appro ximately 30 cm diam. by 90 cm long) undisturbed soil columns were take n from plots where liquid dairy manure or inorganic fertilizers had be en applied continuously for 8 yr. The number and size distribution of macropores in soil columns were nearly the same for both inorganic and manure treatments, however, visible surface macropores were continuou s to much deeper depth in soil columns taken from manure than from the inorganic fertilizer plot. Identification of the earthworms a year la ter showed the presence of Apporectodea tuberculata, A. trapezoides, a nd Lumbricus rubellus, subsurface burrowers, as well as L. terristris, a deeper burrowing species in the manure applied plot. Apporectodea t uberculata was the only species present in the inorganic fertilizer pl ot. Number of macropores and macroporosity varied with soil depth. The maximum macroporosity was <2.5% and it occurred at 2-cm depth. The pr edominant macropore sizes were between 1- and 2-mm radii for both trea tments. During breakthrough experiments, Cl- appeared earlier in soil columns taken from the manure plot thereby indicating a greater contin uity of macroports in the manure compared with the inorganic fertilize r treatment. The early appearance of Cl- in the manure treatment, howe ver, was much slower than one would expect based on the number of macr opores and their continuity estimated from the serial sectioning. This suggests that intrusive serial sectioning and image analysis techniqu es probably overestimate the continuity of macropores possibly due to vacuuming of the earthworm casts and other debris that plugs the macro pore channels. Based on macropore size distribution with depth and rel ated breakthrough curves, it is likely that most existing models of wa ter and contaminant transport that simulate macropore flow, will not a ccurately predict the transport of water and contaminant because of th eir assumption that surface visible macropores are continous to deeper soil depths. Data from this study showed that macropore size distribu tion could be described by a normal or log-normal distribution functio n. These functions in combination with information on continuity and t ortuosity of macropores may be sufficient, when used in some current m acropore models, to adequately describe the conducting efficiency of m acropores in soils.