William M. Edwards

USDA-ARS

Coshocton, Ohio

Watershed studies at Coshocton have documented that summer storms produce less runoff from fields farmed with continuous no-tillage corn than from the same soils when corn is produced with conventional tillage practices. The lack of tillage favors a continuous surface cover of crop residue and the persistence of earthworm burrows, which have been shown to be preferential flow paths for water and chemicals, especially during intense summer storms. We investigated factors that affect preferential water and chemical transport in night crawler burrows in the field using individual burrow samplers and in the laboratory using block of undisturbed soil subjected to simulated rainfall. Rainfall amount and intensity and antecedent soil moisture content affected the amount of water transmitted in earthworm burrows. High intensity storms on relatively dry no-till soils caused the most preferential flow. Transport of surface-applied herbicide was affected by the factors that influenced preferential water flow and by the time of storms relative to the time of herbicide application. Atrazine movement in earthworm burrows was greatest when high intensity rainfall occurred shortly after application.

No tillage has been successfully used as a management practice for controlling erosion on sloping fields in the north-eastern United States for more than 20 years. In some cases, it allows for continuous corn production on slopes that required rotation cropping when clean tillage and cultivation were required for weed control. Without period tillage, the topsoil physical, chemical and biological properties change. In particular, preferential flow paths for infiltrating water can develop and get stronger under longer periods of no-disruptive tillage.

Earthworm burrows, especially those of the night crawler, a large surface-feeding, vertically-burrowing earthworm are potential preferential flow paths that are often more numerous under continuous no-tillage than under conventional tillage. The residue on the no-till surface enhances the night crawler’s environment and the lack of disruptive tillage allows old burrows to persist. Both factors add to the number of burrows per unit area, often reported as >500,000 per acre.

Hydrologically, several factors are important. Clean tilled surfaces tend to crust, especially during and after high intensity storms. When the crusted surfaces occur on slopes, infiltration is less and runoff is greater than from residue-covered slopes. At the Coshocton watersheds, we measured 28 inches of runoff from a plowed corn watershed on a 6% slope during a 4-year period (Table 1). In those same 4 years, a near-by no-till corn watershed on a 9% slope produced <0.5 inch of runoff. Because of the differences in runoff, there may be much more water flowing through the soils in no-till fields than through those that are conventionally tilled.

With more water infiltrating into the no-till fields each year and with the increased number of night crawler burrows in those soils, the opportunity for preferential flow, which bypasses the fine pore matrix, increases. Additionally, annual chemical inputs in a continuous no-till corn field may be much greater than those used when the management involves a crop rotation that includes several years of meadow. Therefore, we investigated the movement of water and chemicals in night crawler burrows and the conditions under which such movement may be important.

Buried plastic bottles and tubing were used to intercept the flow in individual night crawler burrows in no-till corn fields without disturbing the burrow above each bottle (Fig. 1). The sampling arrangement allowed for measuring and sampling water that moved down in each burrow as a result of individual storms.

In 1987, samplers were used in 50 individual burrows in a no-till corn field (Table 2). During the 5 month growing season, 12 storms caused flow in these burrows which was sampled at the 1.5 to 2 foot depth. Rainfall in the 12 storms totaled >11 inches. Within the sampling period, an additional 3.3 inches of rain fell in storms too small to cause flow in the sampled burrows.

An estimated average of 4% of the rainfall in the 12 storms that produced burrow flow infiltrated in the night crawler burrows. This calculation was based on measured numbers of burrows at several locations within the field and on the catch in the 50 monitored burrows. The largest of the 12 storms produced flow in half of the monitored burrows; whereas a much smaller storm of short duration and high intensity caused the highest percent of the rainfall in any storm (10%) to infiltrate in the burrows. Total nitrate transport in the burrow flow of all 12 storms was estimated to be <1 pound per acre.

Macropore flow was also investigated in the laboratory using a rainfall simulator to apply water to the surface of several 1 cubic foot blocks of undisturbed soil taken from no-till corn fields. Percolation (perc) through the bottom of each block was collected in 64 separate cells, which allowed for observing the time and amounts of water and chemical movement through individual segments of the blocks (Fig. 2). In the soils investigated, the dominant flow paths were night crawler burrows.

We sprayed atrazine, as a representative herbicide, on the surface of several blocks and applied simulated rainfall at different intensities and amounts. If the first storm was a small storm or if it fell at low intensity, there was little or no perc and therefore, little or not atrazine movement through the blocks. If the first storm was big, there was more chemical transport. In the second storm on a block, water flow in the wormholes increased but atrazine transport decreased. This relation held for all subsequent storms and did not depend on the size or intensity of the earlier storms. The drier the soil at the start of the simulated storm, the quicker water entered the wormholes and percolated through the blocks. Most of the percolating water and atrazine was transported through just a few burrows in each block.

In another study using the rainfall simulator, we evaluated the length of time between chemical application and the first storm. Three blocks each received a simulated storm either 1 hour, 1 day, 1 week, 2 weeks, or 6 weeks after the atrazine application. In addition to the simulated storms, the 6 week blocks received four natural rainstorms in the field at amounts and times shown in.

The first four data points in (1 hours to 2 weeks) show the effect of elapsed time between chemical application and the first storm on transport of atrazine in earthworm burrows. No rain fell on these blocks before they were brought into the laboratory. The last point of, at 6 weeks, shows the combined effects of time since application and preceding rainfall events. It confirms conclusions from the earlier studies that both time and intervening rainfall reduce the likelihood of significant macropore transport of surface-applied chemicals in subsequent storms.

In summary, these data show that only in the first storm after application, and only if the firs storm is big, of high intensity, an closely follows the herbicide application is there much chance for measurable transport of atrazine in night crawler burrows. In all other storms throughout the year, the concentration of atrazine in the preferential flow shown be very low, often lower than that of the water in the soil matrix. Realizing that the night crawler burrows rarely extend to depths greater than 1 meter in our soils and that water infiltrating in the burrows must subsequently move into the subsoil matrix, there is little evidence that flow of atrazine in earthworm burrows is a substantial threat to groundwater quality in this area.