Spectral reflectance from a broccoli crop with vegetation or soil as background: influence on immigration by Brevicoryne brassicae and Myzus persicae

Light reflectance in five wavebands of the spectrum was measured from broccoli (Brassicae oieracea var. botrytis [L].) interplanted with leguminous cover crops (cover crop background) or broccoli grown as monoculture (bare soil background), and fertilized with compost or synthetic fertilizer. Alate Brevicoryne brassicae (L.) and Myzus persicae (Sulzer) (Homoptera: Aphididae) were monitored in yellow pan water traps and on broccoli leaves. Reflectance intensity was higher with a background of bare soil at all wavebands except blue (400-455 nm) in the early-season. Intensity decreased as broccoli canopy cover increased at all wavebands except blue and green (515-550 nm), declining-most dramatically in the yellow (550-590 nm). Highest late-season intensities were in plots with bare soil background and fertilized with compost (those stressed for nitrogen). Few differences in spectral composition, expressed for each waveband as a percentage of total intensity, were recorded. Numbers of alatae were lowest in cover crop background plots in the early season, reached equivalency with bare soil background by mid-season, and showed highest positive correlations with intensity in the yellow (550-590 nm). Results correspond to laboratory findings that aphids are attracted to higher intensity light, especially in the yellow waveband, and support a phototactic explanation for aphid orientation in the field. Introduction (Smith, 1976; McKinlay, 1985; Ogenga-Latigo, 1992; Bottenberg & Irwin, 1992b), which affect aphid opto­ Interplanting annual crops with alternate crops (inter­ motor response (Kennedy et ai., 1961) or phototaxis cropping) or non-crop vegetation (living mulches) (Moericke, 1955). often reduces numbers of immigrating alate aphids Evidence that spectral light quality influences ori­ compared to annual crops grown as monocultures entation and host plant finding by Homoptera is abun­ (Horn 1981; Tukahirwa & Coaker, 1982; Andow et ai., dant. Peak attraction to yellow (560-590 nm in the light 1986; Cartwright et at., 1990; Bottenberg & Irwin, spectrum) has been found for whiteflies (Vaisham­ 1992a). Moreover, for most aphid species, fewer alatae payan et aI., 1975; Coombe, 1981), leatboppers (Todd are trapped on monocultures with open canopies (i.e., et ai., 1990), psyllids (Mensah & Madden, 1992) having a high ratio of exposed soil to crop foliage) and aphids (Kring, 1972; Kieckhefer, 1976; Burrows than closed canopies (A'Brook, 1964, 1968; Gonza­ et aI., 1983; Campbell, et at., 1991), although some lez & Rawlins, 1968; Ogenga-Latigo, 1992; but see aphid species are attracted most strongly to green (500­ A'Brook, 1968; Halbert & Irwin, 1981). Such pat­ 560 nm) (Kieckhefer 1976; Nottingham et al., 1991). terns in aphid abundance are often attributed to the Repellence by aluminized and other surfaces reflect­ greater visual contrast between a crop and a back­ ing a high proportion of UV light has been shown ground of soil versus a background of vegetation for leatboppers (Zalom, 1981) and aphids (Burton & Krenzer, 1985; Schalk & Robbins, 1987; Kring & Schuster, 1992). Aphids in the initial phase of migration respond positively to shortwave skylight (<500 nm), but become more attracted to longwave light (<500 urn) reflected from plants and soils when in the alightment phase (Kennedy et a1., 1961). Despite these findings, I have found no studies on aphids in interplanting systems which measured the degree ofcrop canopy-to-backgroundcontrast in terms of light reflectance. Presented here are results of a study which attempts to draw an association between patterns of spectral light reflectance from broccoli with bare soil or vegetation as background and immigration by M. persicae and B. brassicae. Materials and methods The study site was located in the Salinas Valley, Cali­ fornia USA; Soil type was a Hanford series sandy clay loam. The experiment took place between 18 May and 8 July 1991 and was designed as a 4 x 2 factorial in a randomized complete block, split plot design with blocks replicated four times and plot size 10 x 10 m. The main plot (background) factor consisted of four levels: three levels of cover crop and one level of a no cover (bare soil) control. The purpose for includ­ ing different cover crops in this study was to compare their competitiveness and compatibility with broccoli as a living mulch, the results of which are presented in a separate paper (Costello, 1994). Cover crops used were white clover (Trifolium repens L.), strawberry clover (Trifolium fragiferum L. cv. O'Connors) and a mixture of trefoil (Lotus corniculatus L. cv. Kalo) and red clover (Trifolium praetense L.). The sub-plot (fertilizer) factor consisted of two levels: synthetic fertilizer (high available nitrogen) or compost (low available nitrogen). Fertilization with compost low­ ered midto late-season broccoli leaf area, leaf water content and leaf nitrogen (Costello, 1994), and there­ fore enabled comparison of late-season reflectances from plants with less than full canopy cover and low nitrogen/water levels. On 18 May 1991 broccoli (cv. 'Arcata') was transplanted into cover crop plots and bare soil plots in 10 cm-wide rows at a row spacing of 0.6 m and an intra-row spacing of 22 cm. These spacings provided 83% cover at planting time in cov­ er crop plots. Other methodology has been described earlier (Costello, 1994). Light measurements were taken using a light meter (LI-185, LICOR, Lincoln, Nebraska, USA) with a sensor having a range of 400-700 nanometers (urn) and equally sensitive to all wavelengths of light. The light meter was modified to a spectral radiometer using a series of glass interference filters (Long Pass Fil­ ters, Oriel Co., Stratford, Connecticutt, USA), each of which limited transmittance of light to that greater than a specified wavelength. The filters were placed sequentially over the sensor, which was ensheathed in a 5 em-long tube of black polyurethene to prevent angled light from entering. Spectral reflectance curves were generated by taking readings with each filter from three randomly selected broccoli plants per subplot, excluding the outer 2 meters to eliminate interference from the edges. The sensor was held 30 cm above the canopy, which recorded reflectance from an area of 784 cm2 (a circle with diameter of 31.6 cm). Read­ ings were taken from the same plants with each filter, and were taken on 22, 32 and 52 days after transplant­ ing (DAP) the broccoli. To minimize changes in light quality, readings were taken under cloudless skies and when the soil surface was uniformly dry, generally between 11 A.M. and 3 P.M.. The period most favor­ able for aphid activity in the Salinas Valley was from dawn until midor late-afternoon, at which point wind­ speeds increased to greater than 2m/sec. Five filters were used, which transmitted incoming light greater than 400 urn, 455 nm, 515 urn, 590 urn, and 645 nm, respectively. Light transmittance for each filter was recorded with a spectrophotometer (Shimadzu Co., Kyoto, Japan; Fig. 1). This created wavelength bands (hereafter referred to as wavebands) of 400-455 nm (blue), 455-515 urn (blue-green), 515-550nm (green), 550-590 urn (yellow), and 590-640 nm (orange). UV light was not measured in this study; however, UV light is not a major component of spectral reflectance from plants or soils (Kennedy et al., 1961). Alatae were trapped with 12 x 8 x 8 em water­ filled aluminum pans painted yellow on the inside and black on the outside. The pans were filled with water and a small amount of non-sudsing detergent was mixed in to break the surface tension. Two pans were used per subplot, maintained at broccoli canopy height. Samples were taken on 12, 22, 32, 42 and 52 DAP. Alatae from broccoli leaves were collected by heat-extraction using a combination of the meth­ ods described by Hughes (1963) and Pielou (1961). Samples were taken on 22, 32, 42, 52 and 62 DAP. 100 ....-::::::._.......-:.;.; ",--­ 80 w : I 0 i I • I : I z