Seed loss of bean and maize varieties as a function of temperature and irrigation levels

Abstract

Minimizing seed loss is one of the main goals for a successful crop establishment. The main purpose of this research was to model seed loss of tropically relevant crops, measured both at germination and at emergence, as a function of temperature and irrigation level. Four bean varieties (catarina, ervilha, manteiga and sondeyombua) and two maize varieties (matuba and sam3) were studied. Experiments were performed in a controlled environment using a thermogradient plate. Temperatures were monitored with Cu-CuNi thermocouples. Seed loss under optimal hydric conditions was simulated at different temperatures, depending on the variety. Eight irrigation levels (6, 7, 8, 9, 10, 12, 15 and 18 mm) were used to simulate the effects of irrigation on losses under optimal thermal conditions. Temperature ranges suitable to the thermal response of each crop were used. A sandy loam-clay textured soil was used in both germination and emergence experiments whereas filter papers in Petri dishes were used for germination experiments only.

Seed loss was minimal along wide thermal or irrigation ranges and increased toward the lowest and highest temperatures and, in most cases, also towards the extreme irrigation levels. The use of even-degree polynomial models successfully identified thermal [T1, T2] and irrigation [I1, I2] ranges over which losses did not exceed the maximum acceptable losses (m.a.l.). The endpoints and length of those ranges were found to be crop-dependent. Differences between estimated polynomial model parameters, for both germination and emergence (Lmin, Tmin or Imin, and ct or ci), highlighted the role of soil in rising the seedling towards the surface. Germination size as a function of temperature also depended on the substrate used.

Both the range lengths and endpoints can be of interest for decision-making regarding crops or varieties to be used or in identifying sowing times for a given crop or variety. Farm weather forecast and warning systems of different nature and agro-climatic zoning may also benefit with the knowledge of these parameters.