array with 0 dimensions i.e. a simple number, character, etc.
vector
array with 1 dimension.
matrix
array with 2 dimensions.
dimension-size
number of values along a dimension.
shape
vector of dimension-sizes of array.
rank
number of dimensions (a.k.a. dimensionality) i.e. shape of shape.
row
final (least-significant) dimension
columns
number of elements in each row
dimension-name
name given to dimension e.g. "latitude".
coordinate-variable (CV)
vector (usually sorted) associated with a dimension of the same size.
CVs are
often used to map an array's dimensions to physical dimensions such as length and time,
thus locating the array elements in physical space and time.
Special Numeric Values
missing-value (MV)
numeric value of data which is abnormal in some way such as:
not applicable (e.g. land point for ocean data)
not available (e.g. instrument failure or delay in obtaining data)
result of some illegal operation such as dividing 0 by 0
undefined for some other reason
infinity
floating-point value representing a value which is too large
(or small in the case of negative infinity) to represent.
This can result from operations such as dividing by 0.
NaN (not-a-number)
floating-point value resulting from an illegal operation such as dividing 0 by 0.