Eileen M. Perry

Managing Wheat From Space: Linking MODIS NDVI and Crop Models for Predicting Australian Dryland Wheat Biomass

This study explored the relationships between moderate resolution imaging spectroradiometer (MODIS) NDVI observations with both measured and simulated fractional green cover (FGrC), leaf area index (LAI), and above ground biomass (AGB) for dryland wheat in Australia. A total of 37 paddocks in north-western Victoria, Australia, were sampled during 2003-2006 for AGB at anthesis, and for FGrC, NDVI (from an active optical sensor), and AGB during 2012. The 2012 FGrC and NDVI measurements were fitted to MODIS NDVI, resulting in positive, linear relationships when the MODIS NDVI values were ≤ 0.80. Measured AGB was also positively, linearly related to MODIS summed NDVI, resulting in an overall R2 of 0.81 and root mean square error (RMSE) of 1397 kg/ha. Crop simulations were run for the fourteen paddocks from 2003 to 2006, and six paddocks from 2012. Four crop phenological points were selected to extract corresponding NDVI and simulated crop parameters: emergence, peak LAI, the mid-point between emergence and peak LAI, and anthesis. Linear models were fit between the MODIS NDVI and simulated values of FGrC, LAI, and AGB. Overall, the highest R2 values corresponded to using all of the dates for FGrC (R2 = 0.82) and AGB (R2 = 0.92), and anthesis dates for LAI (R2 = 0.74). For FGrC and AGB, the RMSE with simulated parameters were comparable or better than the equivalent results from the in situ measurements (note that there were no LAI in situ measurements to compare with). The results support the notion for extending the value of the MODIS NDVI using crop simulation models. The combination of remotely sensed and simulation data might offer regional maps of spatial AGB and ultimately grain yield, which would have high value for research, resource management, policy, and potentially, crop management.

In-field methods for rapid detection of frost damage in Australian dryland wheat during the reproductive and grain-filling phase

Abstract. Frost damage causes significant production losses and costs to Australian dryland wheat, and frost impacts are not expected to decline in the near future, despite global warming. Rapid estimation of frost damage to crops on a spatial basis would allow for timely management decisions to reduce the economic impact of frost events. In this paper, we take a first step in evaluating the utility of hyperspectral reflectance and active light fluorescence for detecting frost damage to wheat during its reproductive phase. Two experiments were conducted immediately after the first observation of frost damage, (i) in 2006, five plots in an existing trial were opportunistically subdivided to take spectral reflectance measurements on frost damaged plants along with yield measurements, and (ii) in 2015, a transect across 31 rows within a commercial paddock was established to evaluate spectral reflectance, fluorometer measurements, and yield along a gradient from non-frosted to frost damaged plants. The results of the hyperspectral reflectance data appeared variable in response across the two experimental sites where frost was observed in-crop. In 2006, hyperspectral-derived indices showed significant differences (P < 0.05) between measurements of frosted and non-frosted canopies, but this was not the case for observations taken in 2015, where the mean response was reversed between experimental sites for several of the indices. In contrast, fluorometer measurements in the 2015 trial resulted in higher correlations with yield and observed frost damage compared with the reflectance measurements. Seven of the nine fluorometer indices evaluated were correlated with yield (used as an indicator of frost damage) at P < 0.01. An index of compounds which absorbs at 375 nm, FLAV, had the best correlation coefficients of 0.91 and 0.90 for the two dates in 2015. The fluorescence index FLAV was selected to evaluate whether it could be used to classify the canopy as frost affected or not, using discriminant analysis for the 2015 transect data. The overall classification accuracy, defined as the number of correctly classified measurements (57) divided by the total number (62) was 92%. The present study was not able to provide insight into how rapidly the sensors could detect frost damage before detection with the naked eye, as the survey data constituted a transect based on early visual symptoms, however this study does provide important insight into what sensors and/or indices may be sensitive to ‘seeing’ early frost damage in-crop. The next steps, which build on this work and need to be resolved are (i) what is the nominal scale of measurements required, and for which portions of the plant canopy? (ii) How robust (over space and time) are any relationships between frost damage and index response? (iii) Can frost damage be detected before the onset of visual damage?

Crop traceability and remote sensing in tree fruit

Fresh market fruit crops such as apples have not employed precision agriculture tools, partially due to the labor intensive nature of the cropping systems. In this paper we describe new research in the development of precision agriculture tools for tree fruit, including the ability to track spatially variable orchard data before harvest through to the packing plant. Remote sensing is a key component of this system, and remote sensing products are being evaluated for their usefulness in guiding orchard management.