Evaluating satellite remote sensing as a method for measuring yield variability in Avocado and Macadamia tree crops. [email protected]Macadamia Avocado Andrew J Robson, Muhammad Moshiur Rahman, Jasmine Muir, Ashley Saint, Chad Simpson and Chris Searle
28
Embed
Evaluating satellite remote sensing as a method for ... · Evaluating satellite remote sensing as a method for measuring yield variability in Avocado and Macadamia tree crops....
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Evaluating satellite remote sensing as a method for measuring yield variability in Avocado and
Indication of cost (100 km2)8 band 50 cm multispectral ~$ 3,300 (USD)8 band 30 cm multispectral ~$ 5,800 (USD)8 band DEM ~$ 12,000 (USD)16 band 30 cm multispectral ~$ 8,000 (USD)
100 km2 area includes 7 farms @ ~ $5,800 (USD)= $830 (USD)/ farm Or ~ $8 (USD) per orchard block.
Timing of image captures:
2 Sep 2015 to correspond with flowering / pollination surveys. (PhD student Bryony Wilcox)
22 Jan 2016 to correspond with final fruit set.
7 Apr 2016 to correspond with pre‐ harvest.
Identifying General Variability to Coordinate Targeted Sampling: NDVI maps
Classified NDVI of orchard block.
High NDVI
Mid NDVI
Low NDVI
Targeted sampling (18 trees per block)
Blocks Sampled
Classified NDVI images derived from WV‐2 image captured 2 September 2015 of two Macadamia blocks: 19.4 ha (a) and 5.5 ha (b) and the three Avocado blocks: 14.4 ha (c), 6.8 ha (d) and 11.3 ha (e). The yellow markers indicate the
locations of the individual trees sampled for yield parameters.
Block 4 Block 5
Block 1 Block 2
Block 3
• Pollinations surveys • Final fruit set counts• Harvesting • Post harvest fruit quality assessments • Targeted soil sampling• Weather stations (temp/ RH) on 12 trees per block
Field Sampling‐ Avocados
• Pollinations surveys• Nut drop counts• Harvest• Post harvest Nut Quality
Other measures• Weather stations (temp/ RH) on 12 trees per Macadamia blocks.
Field Sampling‐ Macadamia
Results
Macadamia Yield results by colour zone replicate
Avocado Yield results by colour zone replicate
Avocado Yield results by colour zone replicate
Extracting canopy reflectance data from each sampled tree
Differential GPS positioning of each tree
2 m buffer applied around each GPS point.
20 structural and pigment based vegetation indices were derived from the 8 band WV‐3 image (7 April)
Linear relationship between measured yield and corresponding canopy spectral data (VI)
Extraction of canopy data for all trees
Calculation of average/ total block yield and derivation yield
map
Derived Yield Maps (Macs)
Derived Yield Maps (Avos)
Improvement on current methods
0
100
200
300
400
500
600
700
1 2 3 4 5 6
Fruit n
umbe
r
Block number
Actual fruit no.
Predicted fruit no.
Estimated fruit no.
Comparison between the actual Vs predicted (extrapolated from imagery/ 18 tree sample) and Growers eyeball estimate fruit number data in 2016
Macadamia Quality• TKR %: Total Kernel Recovery: total kernel wt (g)/ nut is shell wt (g) *100.• Growers are paid on the ratio kernel to shell• SKR %: Sound kernel ratio• Hexanal: a measure of rancidity, a value of 100 + faint off taste
R G Bave temp 24.7 24.1 23.9max temp 34.5 32.6 31.4min temp 18.1 18.3 18.4