Big Data Mining for Climate Change

Cover
BIG DATA
MINING FOR CLIMATE CHANGE
Copyright
Contents
Preface
1 Big climate data
	1.1 Big data sources
		1.1.1 Earth observation big data
		1.1.2 Climate simulation big data
	1.2 Statistical and dynamical downscaling
	1.3 Data assimilation
		1.3.1 Cressman analysis
		1.3.2 Optimal interpolation analysis
		1.3.3 Three-dimensional variational analysis
		1.3.4 Four-dimensional variational analysis
	1.4 Cloud platforms
		1.4.1 Cloud storage
		1.4.2 Cloud computing
	Further reading
2 Feature extraction of big climate data
	2.1 Clustering
		2.1.1 K-means clustering
		2.1.2 Hierarchical clustering
	2.2 Hidden Markov model
	2.3 Expectation maximization
	2.4 Decision trees and random forests
	2.5 Ridge and lasso regressions
	2.6 Linear and quadratic discriminant analysis
		2.6.1 Bayes classifier
		2.6.2 Linear discriminant analysis
		2.6.3 Quadratic discriminant analysis
	2.7 Support vector machines
		2.7.1 Maximal margin classifier
		2.7.2 Support vector classifiers
		2.7.3 Support vector machines
	2.8 Rainfall estimation
	2.9 Flood susceptibility
	2.10 Crop recognition
	Further reading
3 Deep learning for climate patterns
	3.1 Structure of neural networks
	3.2 Back propagation neural networks
		3.2.1 Activation functions
		3.2.2 Back propagation algorithms
	3.3 Feedforward multilayer perceptrons
	3.4 Convolutional neural networks
	3.5 Recurrent neural networks
		3.5.1 Input-output recurrent model
		3.5.2 State-space model
		3.5.3 Recurrent multilayer perceptrons
		3.5.4 Second-order network
	3.6 Long short-term memory neural networks
	3.7 Deep networks
		3.7.1 Deep learning
		3.7.2 Boltzmann machine
		3.7.3 Directed logistic belief networks
		3.7.4 Deep belief nets
	3.8 Reinforcement learning
	3.9 Dendroclimatic reconstructions
	3.10 Downscaling climate variability
	3.11 Rainfall-runoff modeling
	Further reading
4 Climate networks
	4.1 Understanding climate systems as networks
	4.2 Degree and path
	4.3 Matrix representation of networks
	4.4 Clustering and betweenness
	4.5 Cut sets
	4.6 Trees and planar networks
	4.7 Bipartite networks
	4.8 Centrality
		4.8.1 Degree centrality
		4.8.2 Closeness centrality
		4.8.3 Betweenness centrality
	4.9 Similarity
		4.9.1 Cosine similarity
		4.9.2 Pearson similarity
	4.10 Directed networks
	4.11 Acyclic directed networks
	4.12 Weighted networks
		4.12.1 Vertex strength
		4.12.2 Weight-degree/weight-weight correlation
		4.12.3 Weighted clustering
		4.12.4 Shortest path
	4.13 Random walks
	4.14 El Niño southern oscillation
	4.15 North Atlantic oscillation
	Further reading
5 Random climate networks and entropy
	5.1 Regular networks
		5.1.1 Fully connected networks
		5.1.2 Regular ring-shaped networks
		5.1.3 Star-shaped networks
	5.2 Random networks
		5.2.1 Giant component
		5.2.2 Small component
	5.3 Configuration networks
		5.3.1 Edge probability and common neighbor
		5.3.2 Degree distribution
		5.3.3 Giant components
		5.3.4 Small components
		5.3.5 Directed random network
	5.4 Small-world networks
		5.4.1 Main models
		5.4.2 Degree distribution
		5.4.3 Clustering
		5.4.4 Mean distance
	5.5 Power-law degree distribution
		5.5.1 Price's models
		5.5.2 Barabasi-Albert models
	5.6 Dynamics of random networks
	5.7 Entropy and joint entropy
	5.8 Conditional entropy and mutual information
	5.9 Entropy rate
	5.10 Entropy-based climate network
	5.11 Entropy-based decision tree
	Further reading
6 Spectra of climate networks
	6.1 Understanding atmospheric motions via network spectra
	6.2 Adjacency spectra
		6.2.1 Maximum degree
		6.2.2 Diameter
		6.2.3 Paths of length k
	6.3 Laplacian spectra
		6.3.1 Maximum degree
		6.3.2 Connectivity
		6.3.3 Spanning tree
		6.3.4 Degree sequence
		6.3.5 Diameter
	6.4 Spectrum centrality
		6.4.1 Eigenvector centrality
		6.4.2 Katz centrality
		6.4.3 Pagerank centrality
		6.4.4 Authority and hub centralities
	6.5 Network eigenmodes
	6.6 Spectra of complete networks
	6.7 Spectra of small-world networks
	6.8 Spectra of circuit and wheel network
	6.9 Spectral density
	6.10 Spectrum-based partition of networks
	Further reading
7 Monte Carlo simulation of climate systems
	7.1 Random sampling
		7.1.1 Uniform distribution
		7.1.2 Nonuniform distribution
		7.1.3 Normal distribution
	7.2 Variance reduction technique
		7.2.1 Control variable method
		7.2.2 Control vector method
	7.3 Stratified sampling
	7.4 Sample paths for Brownian motion
		7.4.1 Cholesky and Karhounen-Loève expansions
		7.4.2 Brownian bridge
	7.5 Quasi-Monte Carlo method
		7.5.1 Discrepancy
		7.5.2 Koksma-Hlawka inequality
		7.5.3 Van der Corput sequence
		7.5.4 Halton sequence
		7.5.5 Faure sequence
	7.6 Markov chain Monte Carlo
	7.7 Gibbs sampling
	Further reading
8 Sparse representation of big climate data
	8.1 Global positioning
		8.1.1 Multidimensional scaling
		8.1.2 Local rigid embedding
	8.2 Embedding rules
		8.2.1 Attractors and fractal dimension
		8.2.2 Delay embedding
		8.2.3 Multichannel singular spectrum analysis
		8.2.4 Recurrence networks
	8.3 Sparse recovery
		8.3.1 Sparse interpolation
		8.3.2 Sparse approximation
		8.3.3 Greedy algorithms
	8.4 Sparse representation of climate modeling big data
	8.5 Compressive sampling of remote sensing big data
		8.5.1 s-Sparse approximation
		8.5.2 Minimal samples
		8.5.3 Orthogonal matching pursuit
		8.5.4 Compressive sampling matching pursuit
		8.5.5 Iterative hard thresholding
	8.6 Optimality
		8.6.1 Optimization algorithm for compressive sampling
		8.6.2 Chambolle and Pock's primal-dual algorithm
	Further reading
9 Big-data-driven carbon emissions reduction
	9.1 Precision agriculture
	9.2 Oil exploitation
	9.3 Smart buildings
	9.4 Smart grids
	9.5 Smart cities
	Further reading
10 Big-data-driven low-carbon management
	10.1 Large-scale data envelopment analysis
	10.2 Natural resource management
	10.3 Roadway network management
	10.4 Supply chain management
	10.5 Smart energy management
	Further reading
11 Big-data-driven Arctic maritime transportation
	11.1 Trans-Arctic routes
	11.2 Sea-ice remote-sensing big data
		11.2.1 Arctic sea-ice concentration
		11.2.2 Melt ponds
		11.2.3 Arctic sea-ice extent
		11.2.4 Arctic sea-ice thickness
		11.2.5 Arctic sea-ice motion
		11.2.6 Comprehensive integrated observation system
	11.3 Sea-ice modeling big data
	11.4 Arctic transport accessibility model
	11.5 Economic and risk assessments of Arctic routes
	11.6 Big-data-driven dynamic optimal trans-Arctic route system
	11.7 Future prospects
	Further reading
Index
Back Cover