/
North American Croplands North American Croplands

North American Croplands - PowerPoint Presentation

chiquity
chiquity . @chiquity
Follow
342 views
Uploaded On 2020-06-17

North American Croplands - PPT Presentation

Richard Massey Dr Teki Sankey Crop type mapping for the US at 250m Crop type maps for the US for 20082014 We used two approaches Yearspecific and Temporal average Validation Pixeltopixel comparison with USDA CDL ID: 780593

type crop specific temporal crop type temporal specific aaez approach year pixel average mapping accuser time series cropland 2008

Share:

Link:

Embed:

Download Presentation from below link

Download The PPT/PDF document "North American Croplands" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.


Presentation Transcript

Slide1

North American Croplands

Richard Massey

Dr

Teki

Sankey

Slide2

Crop type mapping for the US at 250m

Crop type maps for the US for 2008-2014 We used two approaches: Year-specific and Temporal average

Validation:

Pixel-to-pixel comparison with USDA CDL

Independent accuracy assessment using validation samples derived from USDA CDL

Slide3

Comparison between USDA CDL and cropland products for 2008

USDA CDL

Cropland extent

Crop type classification using year specific approach

Crop type classification using temporal average approach

Slide4

Year-specific approach for 2008: Producer’s and User’s accuracies of major cropland categories

 

Total cropland area

 

Area of major

crops

CORN-SOYBEAN

WHEAT-BARLEY

ALFALFA

OTHER-CROP

FALLOW

Producer's

AccUser's AccProducer's AccUser's AccProducer's AccUser's AccProducer's AccUser's AccProducer's AccUser's AccAEZ Mha%%%%%%%%%%%AEZ 20.5 96a n/abn/a70.271.791.671.372.468.985.299.4AEZ 33.988n/an/a87.581.885.085.0n/an/a73.591.7AEZ 412.38079.878.079.875.770.391.0n/an/a70.565.5AEZ 512.68084.185.387.774.5n/an/an/an/a63.478.1AEZ 613.88390.284.884.578.4n/an/an/an/an/an/aAEZ 711.18490.982.888.779.7n/an/an/an/an/an/aAEZ 811.88692.986.380.774.5n/an/an/an/an/an/aAEZ 913.18585.985.776.193.1n/an/a79.369.7n/an/aAEZ 1022.08689.474.9n/an/an/an/an/an/an/an/aAEZ 116.68690.281.3n/an/an/an/a81.277.9n/an/aAEZ 121.27086.280.8n/an/an/an/a75.277.6n/an/aAEZ 130.18892.8100.0n/an/an/an/a90.697.7n/an/a

Note:

These accuracies are derived from independent accuracy assessment

a = indicates 96% of 0.5

Mha

of croplands in AEZ2 are major crops: (a) Wheat-Barley, (b) Alfalfa, (c ) Other crops, and (d) Fallow

b = indicates non-major crops

Slide5

Spatial distribution of overall accuracies for 2008

Crop type classification using year specific approach

Crop type classification using temporal average approach

Accuracy map for temporal average approach derived using pixel-to-pixel comparison

Accuracy map for year specific approach derived using pixel-to-pixel comparison

Slide6

Crop type mapping for Canada at 250m

Preprocessing:

MODIS MOD09Q1 data for 2009-2015

NDVI time-series stack for each year

Reference layer:

Agriculture

and

Agri

-Food

Canada

(AAFC)

or Ag-Canada crop layer for 2009-2015Cropland extent: Union overlay of AAFC crop layer for 2009-2015Sub-zones: Cropland extent divided in 4 AEZ (AEZ 7-10)2-3 sub-zones in each AEZSamples: Training/Validation samples for each sub-zone from center of fieldsFuture work: NDVI spectral samples and supervised classificationAAFC crop layer for 2012

Slide7

Crop type mapping at 30mConstruction of high spatial and temporal resolution time-series data using Landsat-MODIS data fusion (

e.g STARFM)NDVI time-series and spectral samples analysis and selection for each AEZ

Extraction of image objects using image segmentation

Decision tree classification of image objects based on spectra and NDVI time series

Slide8

Moving ForwardAccuracy assessment for year-specific and temporal-average classifications of

2008, 2010, 2011, and 2012Year-Specific and Temporal average algorithm for crop type mapping in Canada at 250m

Crop type mapping in

C

entral America at 250m

Crop type mapping for the US at 30m using Landsat and MODIS time-series

Slide9

Thank you