1 the modis land cover and land cover dynamics products a.h. strahler (pi), mark friedl, xiaoyang...
TRANSCRIPT
11
The MODIS Land Cover and Land Cover The MODIS Land Cover and Land Cover Dynamics ProductsDynamics Products
A.H. Strahler (PI), Mark Friedl, Xiaoyang Zhang, John Hodges,
Crystal Schaaf, Amanda Cooper, and Alessandro Baccini
http://geography.bu.edu/landcover/Center for Remote Sensing and Dept. of Geography
Boston University
22
MODIS Land Cover: Five Sets of LabelsMODIS Land Cover: Five Sets of Labels
• IGBP:International Geosphere-Biosphere Project labels
– 17 classes of vegetation life-form
• UMD: University of Maryland land cover class labels
– 14 classes without mosaic classes
• LAI/FPAR: Classes for LAI/FPAR Production
– 6 labels including broadleaf and cereal crops
• BGC: Biome BGC Model Classes– 6 labels: leaf type, leaf longevity,
plant persistence
IGBPIGBP
UMDUMD LAI/FPARLAI/FPAR BGCBGC
Plant Functional TypesPlant Functional Types
• Plant Functional Types (Future)– Plant functional types to be used
with the community land model (NCAR, Bonan)
– Exact classes TBD
33
MODIS Land Cover: Where Does it Come From?MODIS Land Cover: Where Does it Come From?
• MODIS Data– 16-day Nadir BRDF-Adjusted Reflectances (NBARs)
assembled over one year of observations– 7 spectral bands, 0.4–2.4 m, similar to Landsat– 16-day Enhanced Vegetation Index (EVI)
• Training Data– >1,500 training sites delineated from high resolution
satellite imagery (largely Landsat)• Classifier
– Uses decision tree classifier with boosting
44
MODIS Land BandsMODIS Land Bands
Band Spatial Resolution Wavelength, nm
1 250 m 654–664
2 250 m 860–870
3 500 m 465–475
4 500 m 550–560
5 500 m 1234–1246
6 500 m 1632–1648
7 500 m 2120–2140
55
MODIS GeolocationMODIS Geolocation
• Geolocation accuracy specification is 300 m (2 ) and goal is 100 m (2 ) at nadir
• Geolocation goal driven by Land 250 m change product requirements
• Goal is currently being met
Land: 550 CPs from 126 TM Scenes
Ocean: 4600 island points from SeaWifs library
Ground Control Points—Land
66
MODIS Data LevelsMODIS Data Levels
• Level 1– Radiometrically corrected, geolocated radiances
• Level 2– Products derived from Level 1 data without geometric resampling
• Level 2G (MODIS Land)– Forward-binned into integerized sinusoidal projection without
resampling
• Level 3– Products resampled using geolocation information to a standard
family of map grids; often multitemporal or composited
• Level 4– Products derived from multiple data sources by modeling
77
IGBP Land Cover Units (17)IGBP Land Cover Units (17)
• Natural Vegetation (11)– Evergreen Needleleaf
Forests– Evergreen Broadleaf Forests– Deciduous Needleleaf Forests– Deciduous Broadleaf Forests– Mixed Forests– Closed Shrublands– Open Shrublands– Woody Savannas– Savannas– Grasslands– Permanent Wetlands
• Developed and Mosaic Lands (3)– Croplands– Urban and Built-Up Lands– Cropland/Natural Vegetation
Mosaics
• Nonvegetated Lands (3)– Snow and Ice– Barren– Water Bodies
88
The Land Cover Input DatabaseThe Land Cover Input Database
• 242 Features From MODIS: – Temporal and spectral information; 16-day composites
• Uses Surface Reflectance (NBAR)– View-angle corrected surface reflectance, 7 land bands
• And Enhanced Vegetation Index (EVI)
• Plus (in the future)….– Spatial Texture from 250-m Band 2
• Standard deviation-to-mean ratio in Band 2 (near-infrared)– Snow Cover
• MODIS Snow Cover Product, number of days with snow cover– Land Surface Temperature
• MODIS Land Surface Temperature, maximum value composite– Directional Information
• Bidirectional reflectance information from BRDF product
Global Composite Map of Nadir BRDF-Adjusted Reflectance (NBAR)April 7–22 2001
10 km resolution, Hammer-Aitoff projection,produced by MODIS BRDF/Albedo Team
no data
MODLAND/Strahler et al.
No data True color, MODIS Bands 2, 4, 3
9
MODIS Nadir BRDF-Adjusted Reflectance
May 25–June 9 2001False Color ImageNIR–Red–Green
10
NBAR Time TrajectoriesNBAR Time Trajectories
11
NDVINDVI
EVIEVI
MODIS 500 m Vegetation Indices
MOD13A1 16 dayComposite
(September 30 – October 15, 2000
MODLAND/Huete et al 1212
NDVI EVI
EVI shows better dynamic range, less saturation1313
1414
Advanced Technology ClassifiersAdvanced Technology Classifiers
• Supervised Mode– Use of supervised mode with training sites– Allows rapid reclassifications for tuning
• Decision Trees—C4.5 Univariate Decision Tree– Fast algorithm– Uses boosting to create multiple trees and improve accuracy,
estimate confidence
• Neural Networks—Fuzzy ARTMAP– Uses Adaptive Resonance Theory in building network– Presently not in use. Too slow; does not handle missing data well.
1515
Decision Tree ClassificationDecision Tree Classification
• Goal:– Optimal prediction of class labels from a set
of feature values• Basic Approach
– Supervised learning using training data• Key Attributes:
– Nonparametric– Able to handle noisy or missing features– Adept at capturing nonlinear, hierarchical
patterns
1616
DTs: Basic TheoryDTs: Basic Theory
• Tree Structure– Root node (all data), internal nodes and
terminal or leaf nodes (predictions)• Building the Decision Tree
– Recursive partitioning of training data into successively more homogeneous subsets
• Multiple Leaf Nodes per Class– Leaf nodes identify class assignment– Sub-classes allocated individual leaves
Root
Leaf nodes
Internal nodes
1717
Postclassification ProcessingPostclassification Processing
• Application of Prior Probabilities– Use of priors to remove training site count biases (sample
equalization)– Application of global and moving-window priors from earlier
products• Increases accuracies, reduces speckle
– Use of external maps of prior probabilities to resolve confusions• Agriculture/natural vegetation confusion in some regions• Use of city lights DMSP data to enhance urban class accuracy
(to come)• Filling of Cloud-Covered Pixels from Earlier Maps
– Use of at-launch (EDC DISCover v. 2) or provisional product when there are not sufficient values to classify a pixel with confidence
1818
MODIS Map of Broadleaf Crops in Continental United StatesMODIS Map of Cereal Crops in the Continental United States
Broadleaf Crop Intensity from USDA Statistics Cereal Crop Intensity from USDA Statistics
Using Priors to Classify Cereal and Broadleaf CropsUsing Priors to Classify Cereal and Broadleaf Crops
1919
Provisional Land Cover Product June 01Provisional Land Cover Product June 01
MODIS data from Jul 00–Jan 01
1919
2020
Consistent Year Land Cover Product June 02—IGBPConsistent Year Land Cover Product June 02—IGBP
MODIS data from Nov 00–Oct 01
2121
Consistent Year Land Cover Product, Nov 00–Oct 01Consistent Year Land Cover Product, Nov 00–Oct 01
Cropland
Cropland/Natural Vegetation Mosaic
Mixed Forest
Evergreen Needleleaf Forest
Deciduous Broadleaf Forest
Urban
2222
Classification Confidence Classification Confidence MapMap
Second Most-Likely Second Most-Likely ClassClass
High Confidence
Lower Confidence
Second choice omittedwith very highconfidence level
2222
2323
Rondonia ComparisonRondonia Comparison
Consistent Year
EDC DISCover v.2
Confidence
Provisional Product
• Note better delineation of land cover pattern
2424
Consistent Year
EDC DISCover v.2
Confidence
Provisional Product
2525
Consistent Year
EDC DISCover v.2
Confidence
Provisional Product
2626
Land Cover ValidationLand Cover Validation
• Validation Plan Utilizes Multiple Approaches• Level 1: Comparisons with existing data sources
– Examples• Global AVHRR land cover datasets: DISCover, UMd• Humid Tropics: Landsat Pathfinder• Forest Cover: FAO Forest Resources Assessment• Western Europe: CORINE• United States: USGS/EPA MLRC• United States: California Timber Maps (McIver and Woodcock)• MODIS and Bigfoot test site comparisons
2727
Validation Levels, Cont.Validation Levels, Cont.
• Level 2: Quantitative studies of output and training data– Per-pixel confidence statistics
• Aggregated by land cover type and region• Describe the accuracy of the classification process
– Test site cross-comparisons• Confusion matrices globally and by region• Provides estimates of errors of omission and commission
• Level 3: Sample-based statistical studies– Random stratified sampling according to proper statistical principles– Costly, but needed for making proper accuracy statements
• CEOS Cal-Val Land Product Validation Land Cover Activity
2828
Confidence Values by Land Cover Type (Preliminary)Confidence Values by Land Cover Type (Preliminary)
IGBP Class Confidence
1 Evergreen Needleleaf 68.3
2 Evergreen Broadleaf 89.3
3 Deciduous Needleleaf 66.7
4 Deciduous Broadleaf 65.9
5 Mixed Forest 65.4
6 Closed Shrubland 60.0
7 Open Shrubland 75.3
8 Woody Savanna 64.0
IGBP Class Confidence
9 Savanna 67.8
10 Grasslands 70.6
11 Permanent Wetlands 52.3
12 Cropland 76.4
14 Cropland/Nat. Veg’n. 60.7
15 Snow and Ice 87.2
16 Barren 90.0
Overall Confidence 76.3
Includes adjustment for prior probabilities. Urban and Built-Up (13), Water(17) classes omitted. Pixels filled from prior data omitted. Based on preliminary data, subject to change.
2929
Confidence Values by Continental Region (Preliminary)Confidence Values by Continental Region (Preliminary)
Region Confidence, percent
Africa 79.4
Australia/Pacific 83.2
Eurasia 76.8
North America 71.9
South America 78.5
Overall Confidence 76.3
Includes adjustment for prior probabilities. Urban and Built-Up (13), Water(17) classes omitted. Pixels filled from prior data omitted. Based on preliminary data, subject to change.
3030
Cross Validation with Training SitesCross Validation with Training Sites
• Cross-Validation Procedure– Hide 10 percent of training sites, classify with remaining
90 percent; repeat ten times for ten unique sets of all sites
– Provides “confusion matrix” based on unseen pixels where whole training site is unseen
– Not a stratified random sample, but a reasonable indication of within-class accuracy
3131
Confusion Matrix (Preliminary)Confusion Matrix (Preliminary)
Global Test Site Confusion Matrix—Consistent Year Product, After Priors
Site ClassClass Name 1 2 3 4 5 6 7 8 9 10 11 12 14 15 16 Total
1 Evergreen Needleleaf 1460 42 18 11 266 7 9 17 23 10 15 21 2 0 0 1901
2 Evergreen Broadleaf 31 4889 0 14 14 11 18 79 23 17 4 38 10 0 1 5149
3 Deciduous Needleleaf 87 0 104 25 118 0 0 4 0 0 0 10 0 0 0 348
4 Deciduous Broadleaf 22 56 16 384 278 0 3 11 1 3 0 47 82 0 0 903
5 Mixed Forest 405 63 94 148 1355 3 1 27 7 8 40 41 17 0 0 2209
6 Closed Shrubland 34 35 2 12 5 140 124 29 15 30 2 158 19 0 8 613
7 Open Shrubland 10 12 3 9 1 41 1002 33 45 203 0 210 6 0 213 1788
8 Woody Savanna 62 133 0 16 110 11 104 577 141 71 0 221 22 0 3 1471
9 Savanna 10 53 1 0 21 18 48 93 440 43 1 252 79 0 16 1075
10 Grasslands 2 16 0 2 20 4 179 6 101 632 0 249 13 0 363 1587
11 Pmnt Wtlnd 63 24 0 5 28 23 1 2 36 2 89 1 7 0 0 281
12 Cropland 6 75 2 7 16 8 61 42 132 133 2 5168 183 0 18 5853
14 Cropland/Natural Vegn 2 133 0 48 28 2 8 16 66 8 1 320 832 0 7 1471
15 Snow+ice 1 0 0 0 0 1 2 0 0 0 5 1 0 1297 5 1312
16 Barren 0 2 1 0 0 1 162 4 5 126 3 56 5 14 3537 3916
Total 2195 5533 241 681 2260 270 1722 940 1035 1286 162 6793 1277 1311 4171 29877
Classification Outcome
3232
Dataset Training Site Accuracy
Before priors 78.6 %
After priors 71.0 %
After priors, first two classes 84.0 %
Accuracies—Consistent Year Product (Preliminary)Accuracies—Consistent Year Product (Preliminary)
Based on Global Test Site Confusion Matrix
3333
Overall AccuraciesOverall Accuracies
• Proper accuracy statements require proper statistical sampling
• AVHRR state of the art has been 60–70 percent, depending on class and region
• MODIS accuracies are falling in 70–80 percent range• Most “mistakes” are between similar classes
• Land cover change should NOT be inferred from comparing successive land cover maps
3434
Land Cover DynamicsLand Cover Dynamics
• Primary Objectives:– Quantify interannual change
• Uses change vectors comparing successive years• Identifies regions of short-term climate variation• Under development with Eric Lambin, Frederic Lupo at UCL,
Belgium– Quantify phenology
• Greenup, maturity, senescence, dormancy• Values of VI, EVI at greenup and peak, plus annual integrated
values• Uses logistic functions fit to time trajectories of EVI
3535
Land Cover Dynamics:Land Cover Dynamics:Defining Phenological AttributesDefining Phenological Attributes
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 50 100 150 200 250 300 350 400
Julian day
Maturity stabilitySenescence onset
Dormancy onset
Dormancy stability
Duration of greenness
Duration of maturity
Maximum Greenness
Greenup onset
Greenup stability
Maturity onset
Senescence stability
3636
QuickTime™ and aVideo decompressor
are needed to see this picture.
3737
QuickTime™ and aVideo decompressor
are needed to see this picture.
3838
Web Site: http://geography.bu.edu/landcoverWeb Site: http://geography.bu.edu/landcover
3939
ReferencesReferences
• Friedl, M.A., D. Muchoney, D.K. McIver, A.H. Strahler, and J.C.F. Hodges 2000: Characterization of North American land cover from AVHRR Data, Geophysical Research Letters, vol. 27, no. 7, pp. 977-980.
• Friedl, M.A., C. Woodcock, S. Gopal, D. Muchoney, A.H.Strahler, and C. Barker-Schaaf 2000. A note on procedures used for accuracy assessment in land cover maps derived from AVHRR data, International Journal of Remote Sensing, vol. 21, pp.1073-1077.
• Friedl, M.A., Brodley, C.E. and A.H. Strahler 1999: Maximizing land cover classification accuracies at continental to global scales, IEEE Transactions on Geoscience and Remote Sensing, vol. 37, pp. 969-977.
• Friedl, M.A. and C.E. Brodley 1997: Decision tree classification of land cover from remotely sensed data, Remote Sensing of Environment, vol. 61, pp. 399-409.
• Mciver, D.K. and M.A. Friedl 2002. Using prior probabilities in decision-tree classification of remotely sensed data, Remote Sensing of Environment, Vol. 81, pp. 253-261.
• McIver, D.K. and M.A. Friedl 2001. Estimating pixel-scale land cover classification confidence using non-parametric machine learning methods, IEEE Transactions on Geoscience and Remote Sensing. Vol 39(9), pp. 1959-1968.
• Muchoney, D., Borak, J, Chi, H., Friedl, M.A., Hodges, J. Morrow, N. and A.H. Strahler 1999: Application of the MODIS global supervised classification model to vegetation and land cover mapping of Central America, International Journal of Remote Sensing, Vol 21, no 6 & 7, pp. 1115-1138.
• Muchoney, D. M., and Strahler, A. H., 2001, Pixel and site-based calibration and validation methods for evaluating supervised classification of remotely sensed data, Remote Sens. Environ., in press.