Ground Cover Supplement : GC Supplement - Root and crown diseases
8 GROUND COVER ROOT & CROWN DISEASES DISEASE IDENTIFICATION Data layers for identifying soil-borne disease NEW GUIDELINES FOR the best method of identifying differences in inoculum level between Precision Agricultural (PA) zones are the result of our latest research on sampling strategies for soil-borne diseases such as take-all, Rhizoctonia and crown rot, which often have a patchy distribution across paddocks. Research supported by the South Australian Research and Development Institute (SARDI) and the GRDC has shown this distribution is not totally random, and that patches are often segregated between PA production zones. This result is not surprising as the PA zones are often correlated with variation in soil attributes, topography and plant growth (egg yield, biomass) that also affect soil-borne pathogen levels. Knowledge of which zones are at risk from diseases is valuable for designing management strategies, so targeting soil sampling to specific zones is encouraged. The PA zones can be mapped using combinations of different data layers (for example, yield, elevation, biomass and so on). These data layers can be collected proximally (on or near ground) and remotely (by satellite, for example). About 30 spatial data layers per paddock were collected across five paddocks to determine which combinations are suitable for defining zones to manage soil-borne diseases. Six zone models were generated, using different combinations of data layers (see Table 1). Inoculum was measured for points on a grid over the whole paddock, and subsets of this grid were used to calculate the level within each zone. All zone maps (except (f), Table 1) combined multiple data layers using a clustering technique in JMP (‘Jump’) computer software. Zone models (a) and (b) were also compared over an additional eight paddocks. Averaged over the five paddocks, proximal (a), satellite (b) and custom MLR (c) zone models were found to be equally useful for mapping variation in distribution of soil-borne pathogens. Since proximal and satellite data worked well, use of custom MLR analysis to produce PA zones was not warranted. ‘Biological’ (d) and ‘geological’ (e) models were equally reliable but less useful than options (a) and (b). Maps based on ECa (f) alone were least useful. When proximal (a) and satellite (b) models were compared over 13 sites (Figure 1) the results suggested that they were equally useful to map the major diseases take-all, Rhizoctonia and cereal cyst nematode (CCN). Satellite NDVI data appeared to be better at predicting crown rot distribution, while the proximal model was better for common root rot, Pratylenchus neglectus, and P. thornei. Based on our work to date, PA zones using satellite or proximal data are equally useful for managing the broad range of soil-borne pathogens. Where growers have access to these PA zone maps, targeting soil sampling to specific zones should improve knowledge of the risk from soil-borne diseases. GRDC Research Codes DAS311, DAS00035 More information: Dr John Heap, 08 8303 9444, email@example.com TABLE 1 ZONE MODELS FOR EACH PADDOCK. INOCULUM LEVELS WERE MEASURED FROM POINTS WITHIN EACH ZONE Model Data layers used for zone map (a) Proximally-sensed data (yield, ECa [soil electrical conductivity], elevation) (b) Satellite NDVI (Normalised Difference Vegetation Index) biomass data (c) Custom disease zones (using correlation matrices and Forward Stepwise Multiple Linear Regression to choose layers -- 'Custom MLR' -- selected from all available data layers) (d) 'Biological' layers (yield, NDVI biomass, aerial photography, N-Sensor) (e) 'Geological' layers (ECa, elevation, slope, gamma-radiometric and magnetic susceptibility) (f) ECa (for example, EM38) alone 1.4 0.8 1.2 1.0 0.4 0.6 0.2 0 --0.2 Rs TA CCN CRp Pn Pt CRR Total n=9 n=5 n=3 n=4 n=4 n=6 n=2 n=33 FIGURE 1 COMPARISON OF USEFULNESS (PARTITION INDEX) FOR ZONES BASED ON PROXIMALLY-SENSED AND SATELLITE NDVI DATA FOR A RANGE OF SOIL-BORNE DISEASES OVER 13 SITES (n=OBSERVATIONS) Partition index (mean) Proximally-sensed Satellite NDVI CCN = cereal cyst nematode Pt = Pratylenchus thornei CRp = crown rot (Fusarium pseudograminearum) Rs = rhizoctonia CRR = common root rot TA = take-all Pn = Pratylenchus neglectus GUIDELINES HAVE BEEN DEFINED FOR IDENTIFYING DIFFERENCES IN SOIL-BORNE DISEASE INOCULUM LEVELS BETWEEN PRECISION AGRICULTURAL ZONES BY JOHN HEAP AND ALAN McKAY Knowledge of which zones are at risk from diseases is valuable for designing management strategies, so targeting soil sampling to specific zones is encouraged.
GC Supplement - Adapting to climate change
GC Supplement - Grains nutrition