Top Banner
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

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....

May 23, 2020

Download

Documents

dariahiddleston
Welcome message from author
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
Page 1: 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. andrew.robson@une.edu.au

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 

Page 2: 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. andrew.robson@une.edu.au

Collaborators

Page 3: 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. andrew.robson@une.edu.au

Field Sites: Bundaberg, Qld

N

3 Avocado orchard blocks2 Macadamia orchid blocks

Page 4: 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. andrew.robson@une.edu.au

Worldview Satellite Imagery

8 bandscoastal (400‐ 450 nm)blue (450 – 510 nm) green (510 – 580 nm) yellow (585 – 625 nm)red (630 – 690 nm) red‐edge (705 – 745 nm) NIR 1 (770 – 895 nm) NIR 2 (860 – 1040 nm)

Worldview 3 31 cm PS, 1.24 m multispectral 

Orbit : 770 km above earthRepeat time: 1.1. days

Worldview 2 46 cm PS, 1.84 m Multispectral

Orbit : 617 km above earthRepeat time: 1 days

16 bands (405‐ 2245 nm)coastal (400‐ 450 nm)blue (450 – 510 nm) green (510 – 580 nm) yellow (585 – 625 nm)red (630 – 690 nm) red‐edge (705 – 745 nm) NIR 1 (770 – 895 nm) NIR 2 (860 – 1040 nm)(Image Copyright © DigitalGlobe)

Page 5: 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. andrew.robson@une.edu.au

http://www.markelowitz.com/Hyperspectral.html

Page 6: 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. andrew.robson@une.edu.au

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.

Page 7: 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. andrew.robson@une.edu.au

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.

Page 8: 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. andrew.robson@une.edu.au

Identifying General Variability to Coordinate Targeted Sampling: NDVI maps

Classified NDVI of orchard block. 

High NDVI

Mid NDVI

Low NDVI

Page 9: 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. andrew.robson@une.edu.au

Targeted sampling (18 trees per block)

Page 10: 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. andrew.robson@une.edu.au

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

Page 11: 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. andrew.robson@une.edu.au

• 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

Page 12: 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. andrew.robson@une.edu.au

• 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

Page 13: 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. andrew.robson@une.edu.au

Results

Page 14: 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. andrew.robson@une.edu.au

Macadamia Yield results by colour zone replicate

Page 15: 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. andrew.robson@une.edu.au

Avocado Yield results by colour zone replicate

Page 16: 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. andrew.robson@une.edu.au

Avocado Yield results by colour zone replicate

Page 17: 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. andrew.robson@une.edu.au

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) 

Page 18: 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. andrew.robson@une.edu.au

Macadamia: VI vs Yield Parameter

Page 19: 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. andrew.robson@une.edu.au

Avocado: VI vs Yield Parameter

Page 20: 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. andrew.robson@une.edu.au

Mapping block yield

False colour WV image with sample locations

Area: 11.3 ha (9*6 spacing)Ave. 3.7 T/ha (20 kg/ tree)Total: 42 tonnes

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

Page 21: 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. andrew.robson@une.edu.au

Derived Yield Maps (Macs)

Page 22: 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. andrew.robson@une.edu.au

Derived Yield Maps (Avos)

Page 23: 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. andrew.robson@une.edu.au

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

Page 24: 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. andrew.robson@une.edu.au

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

Page 25: 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. andrew.robson@une.edu.au

R G Bave temp 24.7 24.1 23.9max temp 34.5 32.6 31.4min temp 18.1 18.3 18.4

ave RH 78.8 80.5 80.9max RH 100.3 100.0 99.7min RH 46.9 50.5 53.2

Temperature data

Low NDVI (Red) trees reach a higher max temp (~3+degC), and lower average relative humidity than the High NDVI (Blue ) trees.

Block 2 R G Bave temp 24.2 23.8 23.6max temp 33.0 31.4 30.1min temp 18.6 18.8 19.1

ave RH 78.6 79.2 80.7max RH 99.3 98.4 97.8min RH 47.2 50.2 54.9

(20 Nov 2015 to 28 May 2016)

12 sensors per block4‐ low NDVI4‐ mid NDVI4‐ high NDVI

Avocado

Mac

Page 26: 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. andrew.robson@une.edu.au

Soil Nutrition influencing Yield (6 blocks)

Page 27: 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. andrew.robson@une.edu.au

2016/ 17 remote sensing/ Field work

Avocado: Bundaberg/ Chiders, Qld and Renmark S.A. 

Macadamia: Bundaberg/ Chiders, Qld and Northern Rivers N.S.W. 

Mango: Bundaberg/ Chiders, Qld;  Darwin N.T. (P.A. and Magpie geese), and Mareeba (pollination) Qld 

Banana: Tully, QLD; Brisbane (PhD/ QDAF)

Page 28: 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. andrew.robson@une.edu.au

Acknowledgments

Growers:Chad Simpson (Simpson Farms)

David Depaoli (Austchilli)Tom Redfern (Donovan Family Investments)

Ben Schaefer (Costa S.A.)

Growers:Clayton Mattiazzi (Hinkler Park

Plantations)Andrew Wallis

Ray Norris