Earthworm numbers, distribution, and sampling under conservation tillage

John Berendsen Dickey, Purdue University

Abstract

In order for crops to grow, the soil must be porous enough for root penetration and water and gas flow to occur. Conservation tillage involves less mechanical loosening and mixing of soil than plowing, and so depends more on earthworms to do these tasks. Tillage, crop rotation, and chemical application affect earthworms. The effect of a chemical depends on its toxicity and the proximity of earthworms to the site of application. Our objectives were (1) to develop efficient techniques for measuring earthworm numbers in field cropped habitats, (2) to determine their numbers and distribution under no-till, ridge-till, and fall-chisel (NT, RT, and FC, respectively) systems, (3) to associate their distribution with the distribution of soil wetness and plant residue, and (4) to compare the earthworm communities between two no-till systems, one of which had fewer toxic chemicals applied to it and more years of soybean (Glycine max) than of corn (Zea mays) in the cropping sequence. A variety of sample pit sizes and shapes were evaluated on the basis of the total sample area required to estimate earthworm (surface soil dwellers, SSD: Aporrectodea tuberculata and Lumbricus rubellus) numbers within a 540 m$\sp2$ plot with $\pm$ 10% precision. The most efficient sample unit required 50 to 70% less effort than the least efficient. Stagewise sampling was evaluated for replicated field experiments and production fields, suggesting that one to 15 pits within each of four replicated plots or 10 or 20 widely dispersed pits in a production field were appropriate ranges for sampling design. Sampling pits of one row width (75 cm) were dug in NT, RT, and FC systems of a tillage experiment during one cycle of a corn-soybean rotation. The pits were subdivided into three depth increments to 25 cm, and laterally into five row positions. Distribution of SSD had strong positive correlation with the depth profile of soil wetness and a weaker positive correlation with lateral distribution of plant residue. In general, SSD in NT = RT $>$ FC during the spring, and NT = RT = FC during the fall, before chiseling. No nightcrawlers (NC, L. terrestris) were observed in any system. Sampling in an adjacent NT field which had been planted with soybean for 3 years revealed NC activity and five times as many SSD as the NT in the tillage experiment after corn. Some aspect of corn cultivation, whether fertilization, insecticides, or the nature of the crop residue, reduces SSD and NC numbers.

Degree

Ph.D.

Advisors

Kladivko, Purdue University.

Subject Area

Agronomy|Ecology

COinS