Ground Cover Supplement : GC Supplement - Precision Agriculture 2005
GROUND COVER PRECISION AGRICULTURE 18 THE NEW SOUTH WALES Department of Primary Industries has been running a precision agriculture project on two sites for one cropping season with one initial aim: to see what information could be obtained quickly to help define manage- ment zones. The project, part of the GRDC PA Initiative, is based near Trangie in central NSW. One site, near Nyngan, has a rela- tively uniform red soil; the other, between Trangie and Tottenham, has a mix of grey soils, a distinct red area and other gra- dational soils. Results from this site are reported here. Variation in yield could mainly be in response to available water, and differenc- es in soil types and their moisture-holding capacity. We were therefore interested in any techniques that might reveal variation in soil type across each site and how that might relate to yield. One method of classifying the soil in this region is by colour. An aerial pho- tograph of the bare soil can be used to reveal colour variation across a paddock. Aerial photographs for NSW are readily obtainable but, with the recent trend to maintaining surface cover at all times, photographs showing bare soil are likely to be some years old. This was the case for the Trangie-Tottenham site, where the most recent bare-soil photograph dated from March 2001. The top picture in Figure 1 (facing page) is the bare-soil aerial photograph of this site clipped to the original paddock boundary. It has been computer classi- fied to enhance differences in soil colour and overlaid in green with boundaries of potential management zones. Different zones of soil colour are shown as red, brown, orange and pink. Trees and gilgais are dark green. Some of the uncleared land, which appears as the black area of the photograph, had been cleared by the time the project started, as can be seen in the lower picture in Figure 1. Only the area to the east of the road- way is being studied in detail. Other site parameters that have been measured and mapped during the year are the draught of a sowing point, apparent electrical con- ductivity as indicated by both EM31 and EM38 and moisture content and bulk den- sity from core samples at sowing, anthesis and harvest. The crop was monitored by the use of airborne multi-spectral imagery at anthesis, sampling for yield and harvest index, and using a yield monitor on the harvester. The roughly circular red area on the photograph indicates a distinct lens of red soil and was an obvious choice for one management zone. The remainder of the eastern half of the paddock was split into two zones, as indicated by the green line around the southern zone marked on the photograph in Figure 1. An important factor in differentiating these areas is the difference in the scale of variation within the zones as a result of gilgais (natural gentle depression) in the north-eastern part of the grey soil area. The gilgais are shown as dark green in Figure 1. The draught of the seeder point tended to be higher in the north-eastern portion, as shown in Figure 2, and this area contains a large, relatively uniform area of soil shown as brown in the classified aerial photograph in Figure 1. Subsequent examination of the soil profiles in these zones showed differences in structure and texture. The lens of red soil does not stand out strongly on the map of seeder point draught in Figure 2, although the areas of soil shown as brown in Figure 1 are similar to areas where the seeder point draught is high, mapped as dark red. The seeder point draught is lower in the areas most recently cleared. This could be due to compaction of the soil or other structural degradation in the areas that have been cropped longest, or to loosening of the soil in the newly cleared area during the clearing process. In either case, since the paddock is now cropped using controlled traffic, this dif- ference may persist for some years. The site was sown to canola in 2004. Because of the colour of the canola flow- ers, analysis of the multispectral imagery needs to be carried out differently than for crops with less dominant flower colours. An index of red versus blue reflectance on the airborne imagery best matched the intensity of flowering in canola. The analysed image showing Red-Blue index Paddocks with a past: previous crops are a vital part of the map KNOWING WHAT HAS GROWN IN A PADDOCK BEFORE CAN HAVE A BIG EFFECT ON YIELD VARIATION, WRITE ALAN PALMER AND IAN McGOWEN 'Variation in yield could mainly be in response to available water, and differences in soil types'
GC Supplement - Nutrient Management 2006
GC Supplement - Grain Storage 2005