Subplots facilitate assessment of corn yield losses from weed competition in a long-term systems experiment | Agronomy for Sustainable Development, Agronomy

Open Access

Table III

Models for defining the relation of weed biomass and weed cover ratings at the Beltsville Farming Systems Project. The overall model of weed biomass, W (g m⁻², as a function of the percent total weed cover, C, and the sum of the percent composition of smooth pigweed plus common lambsquarters (P + L) had significant C × Year and C × (P + L) × Year interactions (n = 69, P < 0.0001, R² = 0.819).

Year	Model ^a	Slope ^b when P + L = 0	Slope ^b when P + L = 94 ^c
1997	W = 1.95C + 0.0183C(P+L)	1.95	3.67
2000	W = 0.30C + 0.0469C(P+L)	0.30	4.71
2001	W = 1.95C + 0.0187C(P+L)	1.95	3.71
2002	W = 4.61C	-	4.61
2004	W = 3.25C	-	3.25
2005	W = 4.90C	-	4.90

Models included a C(P + L) parameter in 1997, 2000, and 2001 because P + L values ranged from 0 to 100 (mean = 62) in those years, but no C(P + L) parameter was included in 2002, 2004, and 2005 because P + L values had a narrow range (90 to 100 with few exceptions, mean = 94) in those years.

Slope of weed biomass (g m⁻²) per unit percent weed cover.

A value of P + L = 94 was chosen as the basis for comparison across all years because this was the mean encountered in 2002, 2004, and 2005.