gridpp Namespace Reference

Typedefs
Short-hand notation for vectors of different dimensions sizes
typedef std::vector< float >	vec
	1D float vector More...
typedef std::vector< vec >	vec2
	2D float vector More...
typedef std::vector< vec2 >	vec3
	3D float vector More...
typedef std::vector< int >	ivec
	1D integer vector More...
typedef std::vector< ivec >	ivec2
	2D integer vector More...
typedef std::vector< ivec2 >	ivec3
	3D integer vector More...
typedef std::vector< double >	dvec
	1D double vector More...
typedef std::vector< dvec >	dvec2
	2D double vector More...

Functions
Data assimilation methods
Functions that merge observations with a background field
vec2	optimal_interpolation (const Grid &bgrid, const vec2 &background, const Points &obs_points, const vec &obs, const vec &variance_ratios, const vec &background_at_points, const StructureFunction &structure, int max_points, bool allow_extrapolation=true)
	Optimal interpolation for a deterministic gridded field. More...
vec	optimal_interpolation (const Points &bpoints, const vec &background, const Points &obs_points, const vec &obs, const vec &variance_ratios, const vec &background_at_points, const StructureFunction &structure, int max_points, bool allow_extrapolation=true)
	Optimal interpolation for a deterministic vector of points. More...
vec2	optimal_interpolation_full (const Grid &bgrid, const vec2 &background, const vec2 &bvariance, const Points &obs_points, const vec &obs, const vec &obs_variance, const vec &background_at_points, const vec &bvariance_at_points, const StructureFunction &structure, int max_points, vec2 &analysis_variance, bool allow_extrapolation=true)
	Optimal interpolation for a deterministic gridded field including analysis variance. More...
vec	optimal_interpolation_full (const Points &bpoints, const vec &background, const vec &bvariance, const Points &obs_points, const vec &obs, const vec &obs_variance, const vec &background_at_points, const vec &bvariance_at_points, const StructureFunction &structure, int max_points, vec &analysis_variance, bool allow_extrapolation=true)
	Optimal interpolation for a deterministic vector of points including analysis variance. More...
vec3	optimal_interpolation_ensi (const Grid &bgrid, const vec3 &background, const Points &obs_points, const vec &obs, const vec &obs_standard_deviations, const vec2 &background_at_points, const StructureFunction &structure, int max_points, bool allow_extrapolation=true)
	Optimal interpolation for an ensemble gridded field See Lussana et al 2019 (DOI: 10.1002/qj.3646) More...
vec2	optimal_interpolation_ensi (const Points &bpoints, const vec2 &background, const Points &obs_points, const vec &obs, const vec &obs_standard_deviations, const vec2 &background_at_points, const StructureFunction &structure, int max_points, bool allow_extrapolation=true)
	Optimal interpolation for an ensemble point field See Lussana et al 2019 (DOI: 10.1002/qj.3646) More...
vec2	local_distribution_correction (const Grid &bgrid, const vec2 &background, const Points &obs_points, const vec &obs, const vec &background_at_points, const StructureFunction &structure, float min_quantile, float max_quantile, int min_points=0)
	Correction of a gridded field ensuring the distribution of values nearby match that of observations. More...
vec2	local_distribution_correction (const Grid &bgrid, const vec2 &background, const Points &obs_points, const vec2 &obs, const vec2 &background_at_points, const StructureFunction &structure, float min_quantile, float max_quantile, int min_points=0)
	Correction of a gridded field ensuring the distribution of values nearby match that of observations. More...
vec2	fill (const Grid &igrid, const vec2 &input, const Points &points, const vec &radii, float value, bool outside)
	Fill in values inside or outside a set of circles (useful for masking) More...
vec2	doping_square (const Grid &igrid, const vec2 &background, const Points &points, const vec &observations, const ivec &halfwidths, float max_elev_diff=gridpp::MV)
	Insert observations into gridded field using a square box. More...
vec2	doping_circle (const Grid &igrid, const vec2 &background, const Points &points, const vec &observations, const vec &radii, float max_elev_diff=gridpp::MV)
	Insert observations into gridded field using a circle. More...
Distributions
Functions that extract values from probability distributions
vec	gamma_inv (const vec &levels, const vec &shape, const vec &scale)
	Extract quantiles from a gamma distribution. More...
Spatial neighbourhood filters
Functions that apply neighbourhood filters on a gridded field
vec2	neighbourhood (const vec2 &input, int halfwidth, Statistic statistic)
	Spatial neighbourhood filter, computing a statistic for a sliding square window. More...
vec2	neighbourhood (const vec3 &input, int halfwidth, Statistic statistic)
	Spatial neighbourhood filter for an ensemble of fields. More...
vec2	neighbourhood_quantile (const vec2 &input, float quantile, int halfwidth)
	Computes a quantile in a sliding square neighbourhood. More...
vec2	neighbourhood_quantile (const vec3 &input, float quantile, int halfwidth)
	Computes a quantile in a sliding square neighbourhood across ensemble members. More...
vec2	neighbourhood_quantile_fast (const vec2 &input, float quantile, int halfwidth, const vec &thresholds)
	Fast and approximate neighbourhood quantile. More...
vec2	neighbourhood_quantile_fast (const vec3 &input, float quantile, int halfwidth, const vec &thresholds)
	Fast and approximate neighbourhood quantile for ensemble of fields. More...
vec2	neighbourhood_quantile_fast (const vec2 &input, const vec2 &quantile, int halfwidth, const vec &thresholds)
	Fast and approximate neighbourhood quantile, with spatially varying quantile levels. More...
vec2	neighbourhood_quantile_fast (const vec3 &input, const vec2 &quantile, int halfwidth, const vec &thresholds)
	Fast and approximate neighbourhood quantile for ensemble of fields, with spatially varying quantile. More...
vec2	neighbourhood_brute_force (const vec2 &input, int halfwidth, Statistic statistic)
	Spatial neighbourhood filter without any shortcuts. More...
vec2	neighbourhood_brute_force (const vec3 &input, int halfwidth, Statistic statistic)
	Spatial ensemble neighbourhood filter without any shortcuts. More...
vec	get_neighbourhood_thresholds (const vec2 &input, int num_thresholds)
	Calculate appropriate approximation thresholds for fast neighbourhood quantile. More...
vec	get_neighbourhood_thresholds (const vec3 &input, int num_thresholds)
	Calculate appropriate approximation thresholds for neighbourhood quantile based on an ensemble. More...
vec2	calc_gradient (const vec2 &base, const vec2 &values, GradientType gradient_type, int halfwidth, int min_num=2, float min_range=gridpp::MV, float default_gradient=0)
	Computes gradients based on values in neighbourhood. More...
vec2	neighbourhood_search (const vec2 &array, const vec2 &search_array, int halfwidth, float search_target_min, float search_target_max, float search_delta, const ivec2 &apply_array=ivec2())
	Find suitable value in neighbourhood that match a search criteria. More...
vec2	neighbourhood_ens (const vec3 &input, int halfwidth, Statistic statistic)
	Deprecated: Compute neighbourhood statistic on ensemble field. More...
vec2	neighbourhood_quantile_ens (const vec3 &input, float quantile, int halfwidth)
	Deprecated: Compute neighbourhood quantiles on ensemble field. More...
vec2	neighbourhood_quantile_ens_fast (const vec3 &input, float quantile, int radius, const vec &thresholds)
	Deprecated: Compute neighbourhood quantiles fast on ensemble field. More...
Calibration methods
Functions that apply statistical methods to data
vec	quantile_mapping_curve (const vec &ref, const vec &fcst, vec &output_fcst, vec quantiles=vec())
	Create quantile mapping calibration curve. More...
vec	metric_optimizer_curve (const vec &ref, const vec &fcst, const vec &thresholds, Metric metric, vec &output_fcst)
	Create calibration curve that optimizes a metric. More...
float	apply_curve (float fcst, const vec &curve_ref, const vec &curve_fcst, Extrapolation policy_below, Extrapolation policy_above)
	Apply arbitrary calibration curve to a single value. More...
vec	apply_curve (const vec &fcst, const vec &curve_ref, const vec &curve_fcst, Extrapolation policy_below, Extrapolation policy_above)
	Apply arbitrary calibration curve to 1D forecasts. More...
vec2	apply_curve (const vec2 &fcst, const vec &curve_ref, const vec &curve_fcst, Extrapolation policy_below, Extrapolation policy_above)
	Apply arbitrary calibration curve to 2D forecasts. More...
vec2	apply_curve (const vec2 &fcst, const vec3 &curve_ref, const vec3 &curve_fcst, Extrapolation policy_below, Extrapolation policy_above)
	Apply arbitrary calibration curve to 2D forecasts with spatially varying QQ map. More...
vec	monotonize_curve (vec curve_ref, vec curve_fcst, vec &output_fcst)
	Ensure calibration curve is monotonic, by removing points on the curve. More...
float	get_optimal_threshold (const vec &curve_ref, const vec &curve_fcst, float threshold, Metric metric)
	Compute the optimal threshold to remap an input threshold to. More...
float	calc_score (float a, float b, float c, float d, Metric metric)
	Compute the score for a 2x2 contingency table. More...
float	calc_score (const vec &ref, const vec &fcst, float threshold, Metric metric)
	Compute the score for a 2x2 contingency table by thresholding. More...
float	calc_score (const vec &ref, const vec &fcst, float threshold, float fthreshold, Metric metric)
	Compute the score for a 2x2 contingency table by thresholding and allowing a different threshold for the forecast. More...
Downscaling methods
Functions that interpolate data from one grid to another
vec2	downscaling (const Grid &igrid, const Grid &ogrid, const vec2 &ivalues, Downscaler downscaler)
	Downscale a gridded field. More...
vec3	downscaling (const Grid &igrid, const Grid &ogrid, const vec3 &ivalues, Downscaler downscaler)
	Downscale a gridded ensemble field. More...
vec	downscaling (const Grid &igrid, const Points &opoints, const vec2 &ivalues, Downscaler downscaler)
	Downscale a gridded field to points. More...
vec2	downscaling (const Grid &igrid, const Points &opoints, const vec3 &ivalues, Downscaler downscaler)
	Downscale a gridded ensemble field to points. More...
vec2	nearest (const Grid &igrid, const Grid &ogrid, const vec2 &ivalues)
	Nearest neighbour dowscaling grid to grid for a deterministic field. More...
vec3	nearest (const Grid &igrid, const Grid &ogrid, const vec3 &ivalues)
	Nearest neighbour dowscaling grid to grid for an ensemble field. More...
vec	nearest (const Grid &igrid, const Points &opoints, const vec2 &ivalues)
	Nearest neighbour dowscaling grid to point for a deterministic field. More...
vec2	nearest (const Grid &igrid, const Points &opoints, const vec3 &ivalues)
	Nearest neighbour dowscaling grid to point for an ensemble field. More...
vec	nearest (const Points &ipoints, const Points &opoints, const vec &ivalues)
	Nearest neighbour dowscaling point to point for a deterministic field. More...
vec2	nearest (const Points &ipoints, const Points &opoints, const vec2 &ivalues)
	Nearest neighbour dowscaling point to point for an ensemble field. More...
vec2	nearest (const Points &ipoints, const Grid &ogrid, const vec &ivalues)
	Nearest neighbour dowscaling point to grid for a deterministic field. More...
vec3	nearest (const Points &ipoints, const Grid &ogrid, const vec2 &ivalues)
	Nearest neighbour dowscaling point to grid for an ensemble field. More...
vec2	downscale_probability (const Grid &igrid, const Grid &ogrid, const vec3 &ivalues, const vec2 &threshold, const ComparisonOperator &comparison_operator)
	Nearest neighbour downscaling grid to grid and probability in one. More...
vec2	mask_threshold_downscale_consensus (const Grid &igrid, const Grid &ogrid, const vec3 &ivalues_true, const vec3 &ivalues_false, const vec3 &theshold_values, const vec2 &threshold, const ComparisonOperator &comparison_operator, const Statistic &statistic)
	Masked ensemble downscaling and consensus. More...
vec2	mask_threshold_downscale_quantile (const Grid &igrid, const Grid &ogrid, const vec3 &ivalues_true, const vec3 &ivalues_false, const vec3 &theshold_values, const vec2 &threshold, const ComparisonOperator &comparison_operator, const float quantile_level)
	Masked ensemble downscaling and quantile extraction. More...
vec2	bilinear (const Grid &igrid, const Grid &ogrid, const vec2 &ivalues)
	Bilinear downscaling grid to grid for a deterministic field. More...
vec3	bilinear (const Grid &igrid, const Grid &ogrid, const vec3 &ivalues)
	Bilinear dowscaling grid to grid for an ensemble field. More...
vec	bilinear (const Grid &igrid, const Points &opoints, const vec2 &ivalues)
	Bilinear dowscaling grid to point for a deterministic field. More...
vec2	bilinear (const Grid &igrid, const Points &opoints, const vec3 &ivalues)
	Bilinear dowscaling grid to point for an ensemble field. More...
vec2	simple_gradient (const Grid &igrid, const Grid &ogrid, const vec2 &ivalues, float elev_gradient, Downscaler downscaler=Nearest)
	Elevation correction dowscaling grid to grid for a deterministic field. More...
vec3	simple_gradient (const Grid &igrid, const Grid &ogrid, const vec3 &ivalues, float elev_gradient, Downscaler downscaler=Nearest)
	Elevation correction dowscaling grid to grid for an ensemble field. More...
vec	simple_gradient (const Grid &igrid, const Points &opoints, const vec2 &ivalues, float elev_gradient, Downscaler downscaler=Nearest)
	Elevation correction dowscaling grid to point for a deterministic field. More...
vec2	simple_gradient (const Grid &igrid, const Points &opoints, const vec3 &ivalues, float elev_gradient, Downscaler downscaler=Nearest)
	Elevation correction dowscaling grid to point for an ensemble field. More...
vec2	full_gradient (const Grid &igrid, const Grid &ogrid, const vec2 &ivalues, const vec2 &elev_gradient, const vec2 &laf_gradient=vec2(), Downscaler downscaler=Nearest)
	Compute Downscale. More...
vec3	full_gradient (const Grid &igrid, const Grid &ogrid, const vec3 &ivalues, const vec3 &elev_gradient, const vec3 &laf_gradient, Downscaler downscaler=Nearest)
	Spatially varying gradient dowscaling grid to grid for an ensemble field. More...
vec	full_gradient (const Grid &igrid, const Points &opoints, const vec2 &ivalues, const vec2 &elev_gradient, const vec2 &laf_gradient, Downscaler downscaler=Nearest)
	Spatially varying gradient dowscaling grid to point for a deterministic field. More...
vec2	full_gradient (const Grid &igrid, const Points &opoints, const vec3 &ivalues, const vec3 &elev_gradient, const vec3 &laf_gradient, Downscaler downscaler=Nearest)
	Spatially varying gradient dowscaling grid to point for an ensemble field. More...
vec3	full_gradient_debug (const Grid &igrid, const Grid &ogrid, const vec2 &ivalues, const vec2 &elev_gradient, const vec2 &laf_gradient=vec2(), Downscaler downscaler=Nearest)
vec2	smart (const Grid &igrid, const Grid &ogrid, const vec2 &ivalues, int num, const StructureFunction &structure)
	Smart neighbour downscaling grid to grid for a deterministic field. More...
Grid calculations
Functions that calculate statistics on a grid
vec	count (const Grid &grid, const Points &points, float radius)
	For each point, counts the number of gridpoints within the radius. More...
vec2	count (const Grid &igrid, const Grid &ogrid, float radius)
	For each gridpoint, counts the number of gridpoints within the radius. More...
vec2	count (const Points &points, const Grid &grid, float radius)
	For each gridpoint, counts the number of points within the radius. More...
vec	count (const Points &ipoints, const Points &opoints, float radius)
	For each point, counts the number of points within the radius. More...
vec	distance (const Grid &grid, const Points &points, int num=1)
	For each point, calculates the distance to nearest gridpoint. More...
vec2	distance (const Grid &igrid, const Grid &ogrid, int num=1)
	For each output gridpoint, calculate the distance to nearest input gridpoint. More...
vec2	distance (const Points &points, const Grid &grid, int num=1)
	For each gridpoint, calculates the distance to nearest point. More...
vec	distance (const Points &ipoints, const Points &opoint, int num=1)
	For each output point, calculates the distance to nearest input point. More...
vec2	fill_missing (const vec2 &values)
	Fill in missing values based on nearby values. More...
vec2	gridding (const Grid &grid, const Points &points, const vec &values, float radius, int min_num, Statistic statistic)
	Aggregate points onto a grid. More...
vec	gridding (const Points &opoints, const Points &ipoints, const vec &values, float radius, int min_num, Statistic statistic)
	Aggregate points onto a points. More...
vec2	gridding_nearest (const Grid &grid, const Points &points, const vec &values, int min_num, gridpp::Statistic statistic)
	Assign each point to nearest neighbour in grid and aggregate values. More...
vec	gridding_nearest (const Points &opoints, const Points &ipoints, const vec &values, int min_num, gridpp::Statistic statistic)
	Assign each ipoint to nearest neighbour in opoints and aggregate values. More...
vec2	neighbourhood_score (const Grid &grid, const Points &points, const vec2 &fcst, const vec &ref, int half_width, gridpp::Metric metric, float threshold)
	Compute a score for a metric of all points within a radius. More...
Diagnosing meteorological variables
Functions that diagnose a meteorological variable based on other variables
float	dewpoint (float temperature, float relative_humidity)
	Calculate dewpoint temperature from temperature and relative humidity. More...
vec	dewpoint (const vec &temperature, const vec &relative_humidity)
	Vector version of dewpoint calculation. More...
float	pressure (float ielev, float oelev, float ipressure, float itemperature=288.15)
	Calculate pressure at a new elevation. More...
vec	pressure (const vec &ielev, const vec &oelev, const vec &ipressure, const vec &itemperature)
	Calculate Vector version of pressure calculation. More...
float	sea_level_pressure (float ps, float altitude, float temperature, float rh=gridpp::MV, float dewpoint=gridpp::MV)
	Convert Surface Pressure to Sea Level Pressure. More...
vec	sea_level_pressure (const vec &ps, const vec &altitude, const vec &temperature, const vec &rh, const vec &dewpoint)
	Vector version of Convert Surface Pressure to Sea Level Pressure. More...
float	qnh (float pressure, float altitude)
	Diagnose QNH from pressure and altitude. More...
vec	qnh (const vec &pressure, const vec &altitude)
	Vector version of QNH calculation. More...
float	relative_humidity (float temperature, float dewpoint)
	Calculate relative humidity from temperature and dewpoint temperature. More...
vec	relative_humidity (const vec &temperature, const vec &dewpoint)
	Vector version of relative humidity calculation. More...
float	wetbulb (float temperature, float pressure, float relative_humidity)
	Calculate wetbulb temperature from temperature, pressure, and relative humidity. More...
vec	wetbulb (const vec &temperature, const vec &pressure, const vec &relative_humidity)
	Vector version of wetbulb calculation. More...
float	wind_speed (float xwind, float ywind)
	Diagnose wind speed from its components. More...
vec	wind_speed (const vec &xwind, const vec &ywind)
	Vector version of wind speed calculation. More...
float	wind_direction (float xwind, float ywind)
	Diagnose wind direction from its components. More...
vec	wind_direction (const vec &xwind, const vec &ywind)
	Vector version of wind direction calculation. More...
OpenMP settings
Functions that configure OpenMP
void	set_omp_threads (int num)
	Set the number of OpenMP threads to use. More...
int	get_omp_threads ()
	Get the number of OpenMP threads currently set. More...
void	initialize_omp ()
	Sets the number of OpenMP threads to 1 if OMP_NUM_THREADS undefined. More...

Variables
Constants
Functions that assimilate observations onto a gridded background
static const float	MV = NAN
	Missing value indicator. More...
static const float	MV_CML = -999
	Missing value indicator in gridpp command-line tool. More...
static const float	pi = 3.14159265
	Mathematical constant pi. More...
static const double	radius_earth = 6.378137e6
	Radius of the earth [m]. More...
static const float	lapse_rate =0.0065
	Constant Lapse Rate moist air standard atmosphere [K/m]. More...
static const float	standard_surface_temperature = 288.15
	Temperature at surface in standard atmosphere [K]. More...
static const float	gravit = 9.80665
	Gravitational acceleration [m/s^2]. More...
static const float	molar_mass = 0.0289644
	Molar Mass of Dry Air [kg/mol]. More...
static const float	gas_constant_mol = 8.31447
	Universal Gas Constant [kg*m^2*s^-2/(K*mol)]. More...
static const float	gas_constant_si = 287.05
	Universal Gas Constant [J/(kg*K)]. More...

Enumerations
enum	ComparisonOperator { Lt = 0, Leq = 10, Gt = 20, Geq = 30 }
	Types of comparison operators. More...
enum	CoordinateType { Geodetic = 0, Cartesian = 1 }
	Types of coordinates for position of points. More...
enum	CorrectionType { Qq = 0, Multiplicative = 10, Additive = 20 }
	Method for statistical correction. More...
enum	Downscaler { Nearest = 0, Bilinear = 1 }
	Types of simple downscaling methods. More...
enum	Extrapolation {   OneToOne = 0, MeanSlope = 10, NearestSlope = 20, Zero = 30,   Unchanged = 40 }
	Methods for extrapolating outside a curve. More...
enum	GradientType { MinMax = 0, LinearRegression = 10 }
	Types of methods to calculate the gradient. More...
enum	Metric {   Ets = 0, Ts = 1, Kss = 20, Pc = 30,   Bias = 40, Hss = 50 }
	Binary verification metrics. More...
enum	Statistic {   Mean = 0, Min = 10, Median = 20, Max = 30,   Quantile = 40, Std = 50, Variance = 60, Sum = 70,   Count = 80, RandomChoice = 90, Unknown = -1 }
	Statistical operations to reduce a vector to a scalar. More...

Classes
class	BarnesStructure
	Simple structure function based on distance, elevation, and land area fraction. More...
class	BoxCox
	Box-Cox transformation. More...
class	CressmanStructure
	Simple structure function based on distance, elevation, and land area fraction. More...
class	CrossValidation
class	Gamma
	Gamma transformation. More...
class	Grid
	Represents a 2D grid of locations and their metadata. More...
class	Identity
	Identity transform, i.e. More...
class	KDTree
	Helper class for Grid and Points representing a tree of points. More...
class	Log
	Log transformation: output = log(input) More...
class	MultipleStructure
class	not_implemented_exception
class	Point
	Represents a single point in some coordinate system. More...
class	Points
	Represents a vector of locations and their metadata. More...
class	StructureFunction
	Covariance structure function. More...
class	Transform

Utilities
Helper functions
static int	_debug_level = 0
void	set_debug_level (int level)
	Set the verbosity of debug messages. More...
int	get_debug_level ()
	Get the currently set level of debug messagess. More...
Statistic	get_statistic (std::string name)
	Convert name of a statistic enums. More...
std::string	version ()
	The gridpp version. More...
double	clock ()
	The current time. More...
void	debug (std::string string)
	Writes a debug message to standard out. More...
void	warning (std::string string)
	Writes a warning message to standard out. More...
void	error (std::string string)
	Writes an error message to standard out. More...
void	future_deprecation_warning (std::string function, std::string other="")
	Writes an deprecation warning to standard out. More...
bool	is_valid (float value)
	Checks if a value is valid. More...
float	calc_statistic (const vec &array, Statistic statistic)
	Compute a statistic on a 1D vector. More...
float	calc_quantile (const vec &array, float quantile_level)
	Compute a quantile from a 1D vector. More...
vec	calc_statistic (const vec2 &array, Statistic statistic)
	Compute a statistic on a 2D vector. More...
vec	calc_quantile (const vec2 &array, float quantile=MV)
	Compute a quantile from a 2D vector. More...
vec2	calc_quantile (const vec3 &array, const vec2 &quantile_levels)
	Compute quantile with 2D varying quantile level. More...
bool	convert_coordinates (const vec &lats, const vec &lons, CoordinateType type, vec &x_coords, vec &y_coords, vec &z_coords)
	Convert lats/lons or 2D x/y coordinates to 3D cartesian coordinates with the centre of the earth as the origin. More...
bool	convert_coordinates (float lat, float lon, CoordinateType type, float &x_coord, float &y_coord, float &z_coord)
	Convert lat/lon or 2D x/y coordinate to 3D cartesian coordinate with the centre of the earth as the origin. More...
bool	is_valid_lat (float lat, CoordinateType type)
	Checks that a lat-coordinate is valid (based on the coordinate type) More...
bool	is_valid_lon (float lon, CoordinateType type)
	Checks that a lon-coordinate is valid (based on the coordinate type) More...
int	num_missing_values (const vec2 &iArray)
	Count the number of missing values in a vector. More...
int	get_lower_index (float iX, const vec &iValues)
	Find the index in a vector that is equal to or just below a value. More...
int	get_upper_index (float iX, const vec &iValues)
	Find the index in a vector that is equal to or just above a value. More...
float	interpolate (float x, const vec &iX, const vec &iY)
	Piecewise linear interpolation If x is outside the range of iX, then the min/max value of iY is used. More...
vec	interpolate (const vec &x, const vec &iX, const vec &iY)
	Piecewise linear interpolation, vectorised version. More...
ivec2	init_ivec2 (int Y, int X, int value)
	Initialize a 2D integer vector of size Y, X, with a given value. More...
vec2	init_vec2 (int Y, int X, float value=MV)
	Initialize a 2D float vector of size Y, X, with a given value. More...
ivec3	init_ivec3 (int Y, int X, int E, int value)
	Initialize a 3D integer vector of size Y, X, E, with a given value. More...
vec3	init_vec3 (int Y, int X, int E, float value=MV)
	Initialize a 3D float vector of size Y, X, E, with a given value. More...
vec	calc_even_quantiles (const vec &values, int num)
	Get reasonably spaced quantile levels from a vector of values, ignoring duplicate values but including the first number after duplicated values. More...
vec2	window (const vec2 &array, int length, gridpp::Statistic statistic, bool before=false, bool keep_missing=false, bool missing_edges=true)
	Compute window statistics across time, independently for each case. More...
bool	compatible_size (const Grid &grid, const vec2 &v)
	Check if the grid is the same size as the 2D vector. More...
bool	compatible_size (const Grid &grid, const vec3 &v)
	Check if the grid is compatible with 3D vector. More...
bool	compatible_size (const Points &points, const vec &v)
	Check if the points are compatible with a 1D vector. More...
bool	compatible_size (const Points &points, const vec2 &v)
	Check if the points are compatible with a 2D vector. More...
bool	compatible_size (const vec2 &a, const vec2 &b)
	Check if two 2D vectors are compatible. More...
bool	compatible_size (const vec2 &a, const vec3 &b)
	Check if a 2D and 3D vectors are compatible. More...
bool	compatible_size (const vec3 &a, const vec3 &b)
	Check if two 3D vectors are compatible. More...
bool	point_in_rectangle (const Point &A, const Point &B, const Point &C, const Point &D, const Point &m)
	Checks if a point is located inside a rectangle formed by 4 points. More...

SWIG testing functions
Functions for testing the SWIG interface. Not useful for any other purpose.
static const float	swig_default_value = -1
	Default value used to fill array in SWIG testing functions. More...
float *	test_array (float *v, int n)
	Special function whose presense is needed for SWIG. More...
float	test_vec_input (const vec &input)
	Testing function for 1D input vector. More...
int	test_ivec_input (const ivec &input)
	Testing function for 1D input vector. More...
float	test_vec2_input (const vec2 &input)
	Testing function for 2D input vector. More...
float	test_vec3_input (const vec3 &input)
	Testing function for 3D input vector. More...
vec	test_vec_output ()
	Testing function for 1D output vector. More...
ivec	test_ivec_output ()
vec2	test_vec2_output ()
	Testing function for 2D output vector. More...
ivec2	test_ivec2_output ()
vec3	test_vec3_output ()
	Testing function for 3D output vector. More...
ivec3	test_ivec3_output ()
float	test_vec_argout (vec &distances)
	Testing function for 1D vector treated as output. More...
float	test_vec2_argout (vec2 &distances)
	Testing function for 2D vector treated as output. More...
void	test_not_implemented_exception ()

Typedef Documentation

dvec

typedef std::vector<double> gridpp::dvec

1D double vector

dvec2

typedef std::vector<dvec> gridpp::dvec2

2D double vector

ivec

typedef std::vector<int> gridpp::ivec

1D integer vector

ivec2

typedef std::vector<ivec> gridpp::ivec2

2D integer vector

ivec3

typedef std::vector<ivec2> gridpp::ivec3

3D integer vector

vec

typedef std::vector<float> gridpp::vec

1D float vector

vec2

typedef std::vector<vec> gridpp::vec2

2D float vector

vec3

typedef std::vector<vec2> gridpp::vec3

3D float vector

Enumeration Type Documentation

ComparisonOperator

enum gridpp::ComparisonOperator

Types of comparison operators.

Enumerator
Lt	Lower than, <.
Leq	Lower or equal than, <=.
Gt	Greater than, >
Geq	Greater or equal than, >=.

CoordinateType

enum gridpp::CoordinateType

Types of coordinates for position of points.

Enumerator
Geodetic	Latitude and longitude.
Cartesian	X and Y.

CorrectionType

enum gridpp::CorrectionType

Method for statistical correction.

Enumerator
Qq	Quantile mapping.
Multiplicative	Multiplicative.
Additive	Additive.

Downscaler

enum gridpp::Downscaler

Types of simple downscaling methods.

Enumerator
Nearest	Nearest neighour downscaler.
Bilinear	Bilinear downscaler.

Extrapolation

enum gridpp::Extrapolation

Methods for extrapolating outside a curve.

Enumerator
OneToOne	Continue past the end-points using a slope of 1.
MeanSlope	Continue past the end-points using the mean slope of the curve.
NearestSlope	Continue past the end-points using the slope of the two lowermost or uppermost points in the curve.
Zero	Continue past the end-points using a slope of 0.
Unchanged	Keep values the way they were.

GradientType

enum gridpp::GradientType

Types of methods to calculate the gradient.

Enumerator
MinMax
LinearRegression

Metric

enum gridpp::Metric

Binary verification metrics.

Enumerator
Ets	Equitable threat score.
Ts	Threat score.
Kss	Hannsen-Kuiper skill score.
Pc	Proportion correct.
Bias	Bias.
Hss	Heidke skill score.

Statistic

enum gridpp::Statistic

Statistical operations to reduce a vector to a scalar.

Enumerator
Mean	Mean of values.
Min	Minimum of values.
Median	Mean of values.
Max	Maximum of values.
Quantile	A quantile from values.
Std	Standard deviation of values.
Variance	Population variance of values.
Sum	Sum of values.
Count	Count of values.
RandomChoice	Randomly pick a non-nan value.
Unknown	Unknown statistic.

Function Documentation

apply_curve() [1/4]

float gridpp::apply_curve	(	float	fcst,
		const vec &	curve_ref,
		const vec &	curve_fcst,
		gridpp::Extrapolation	policy_below,
		gridpp::Extrapolation	policy_above
	)

Apply arbitrary calibration curve to a single value.

Parameters

fcst	1D vector of forecast values to correct
curve_ref	1D vector of reference curve values
curve_fcst	1D vector of forecast curve values
policy_below	Extrapolation policy below curve
policy_above	Extrapolation policy above curve

Returns

Calibrated forecasts

apply_curve() [2/4]

vec gridpp::apply_curve	(	const vec &	fcst,
		const vec &	curve_ref,
		const vec &	curve_fcst,
		gridpp::Extrapolation	policy_below,
		gridpp::Extrapolation	policy_above
	)

Apply arbitrary calibration curve to 1D forecasts.

Parameters

fcst	1D vector of forecast values to correct
curve_ref	1D vector of reference curve values
curve_fcst	1D vector of forecast curve values
policy_below	Extrapolation policy below curve
policy_above	Extrapolation policy above curve

Returns

1D vector of calibrated forecasts

apply_curve() [3/4]

vec2 gridpp::apply_curve	(	const vec2 &	fcst,
		const vec &	curve_ref,
		const vec &	curve_fcst,
		gridpp::Extrapolation	policy_below,
		gridpp::Extrapolation	policy_above
	)

Apply arbitrary calibration curve to 2D forecasts.

Parameters

fcst	2D vector of forecast values to correct
curve_ref	1D vector of reference curve values
curve_fcst	1D vector of forecast curve values
policy_below	Extrapolation policy below curve
policy_above	Extrapolation policy above curve

Returns

2D array of calibrated forecasts (Y, X)

apply_curve() [4/4]

vec2 gridpp::apply_curve	(	const vec2 &	fcst,
		const vec3 &	curve_ref,
		const vec3 &	curve_fcst,
		gridpp::Extrapolation	policy_below,
		gridpp::Extrapolation	policy_above
	)

Apply arbitrary calibration curve to 2D forecasts with spatially varying QQ map.

Parameters

fcst	2D vector of forecast values to correct
curve_ref	3D vector of reference curve values (Y, X, Curve)
curve_fcst	3D vector of Forecast curve values (Y, X, Curve)
policy_below	Extrapolation policy below curve
policy_above	Extrapolation policy above curve

Returns

2D vector of calibrated forecasts (Y, X)

bilinear() [1/4]

vec2 gridpp::bilinear	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec2 &	ivalues
	)

Bilinear downscaling grid to grid for a deterministic field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	2D vector of input values (Yi, Xi)

Returns

2D vector of output values (Yo, Xo)

bilinear() [2/4]

vec3 gridpp::bilinear	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec3 &	ivalues
	)

Bilinear dowscaling grid to grid for an ensemble field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	3D vector of input values (E, Yi, Xi)

Returns

3D vector of output values (E, Yo, Xo)

bilinear() [3/4]

vec gridpp::bilinear	(	const Grid &	igrid,
		const Points &	opoints,
		const vec2 &	ivalues
	)

Bilinear dowscaling grid to point for a deterministic field.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	2D vector of input values grid (Yi, Xi)

Returns

1D vector of output values

bilinear() [4/4]

vec2 gridpp::bilinear	(	const Grid &	igrid,
		const Points &	opoints,
		const vec3 &	ivalues
	)

Bilinear dowscaling grid to point for an ensemble field.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	3D vector of input values (E, Yi, Xi)

Returns

2D vector of output values (E, Points)

calc_even_quantiles()

vec gridpp::calc_even_quantiles	(	const vec &	values,
		int	num
	)

Get reasonably spaced quantile levels from a vector of values, ignoring duplicate values but including the first number after duplicated values.

Include the lowest and highest values.

Parameters

values	1D vector of values (unsorted, and no invalid values)
num	number of thresholds to get

Returns

1D vector of sorted quantile levels

calc_gradient()

vec2 gridpp::calc_gradient	(	const vec2 &	base,
		const vec2 &	values,
		GradientType	gradient_type,
		int	halfwidth,
		int	min_num = 2,
		float	min_range = gridpp::MV,
		float	default_gradient = 0
	)

Computes gradients based on values in neighbourhood.

Parameters

base	2D vector of dependent values, missing values are ignored (Y, X)
values	2D vector of independent values, missing values are ignored (Y, X)
gradient_type	What gradient type to compute
halfwidth	Neighbourhood half width in number of gridpoints
min_nim	Minimum number of points required to compute gradient
min_range	Minimum range of base to compute gradient
default_gradient	Use this gradient if minimum number of points not available

Returns

2D vector of gradients (Y, X)

calc_quantile() [1/3]

float gridpp::calc_quantile	(	const vec &	array,
		float	quantile_level
	)

Compute a quantile from a 1D vector.

Parameters

array	1D vector of values
quantile_level	Quantile level to compute

Returns

Computed quantile

calc_quantile() [2/3]

vec gridpp::calc_quantile	(	const vec2 &	array,
		float	quantile = MV
	)

Compute a quantile from a 2D vector.

Parameters

array	2D vector of values (Y, X)
quantile_level	Quantile level to compute

Returns

1D vector of computed quantiles (Y)

calc_quantile() [3/3]

vec2 gridpp::calc_quantile	(	const vec3 &	array,
		const vec2 &	quantile_levels
	)

Compute quantile with 2D varying quantile level.

Parameters

array	3D input vector (T, Y, X)
quantile	2D vector of Quantile levels (Y, X)

Returns

Extracted quantiles (Y, X)

calc_score() [1/3]

float gridpp::calc_score	(	float	a,
		float	b,
		float	c,
		float	d,
		gridpp::Metric	metric
	)

Compute the score for a 2x2 contingency table.

Parameters

a	Fraction of hits
b	Fraction of false alarms
c	Fraction of misses
d	Fraction of correct rejections
metric	Metric to conpute score for

Returns

Score

calc_score() [2/3]

float gridpp::calc_score	(	const vec &	ref,
		const vec &	fcst,
		float	threshold,
		gridpp::Metric	metric
	)

Compute the score for a 2x2 contingency table by thresholding.

Parameters

ref	1D vector of reference values
fcst	1D vector forecast values
threshold	Threshold to compute score for
metric	Metric to conpute score for

Returns

Score

calc_score() [3/3]

float gridpp::calc_score	(	const vec &	ref,
		const vec &	fcst,
		float	threshold,
		float	fthreshold,
		gridpp::Metric	metric
	)

Compute the score for a 2x2 contingency table by thresholding and allowing a different threshold for the forecast.

Parameters

ref	1D vector of reference values
fcst	1D vector forecast values
threshold	Threshold for the reference values
fthreshold	Threshold for the forecast values
metric	Metric to conpute score for

Returns

Score

calc_statistic() [1/2]

float gridpp::calc_statistic	(	const vec &	array,
		gridpp::Statistic	statistic
	)

Compute a statistic on a 1D vector.

Parameters

array	1D vector of values
statistic	Statistic to compute

Returns

Computed value

calc_statistic() [2/2]

vec gridpp::calc_statistic	(	const vec2 &	array,
		gridpp::Statistic	statistic
	)

Compute a statistic on a 2D vector.

Parameters

array	2D vector of values (Y, X)
statistic	Statistic to compute

Returns

1D vector of computed values (Y)

clock()

double gridpp::clock

(

)

The current time.

Returns

The current time in seconds since 1870-01-01 00:00:00Z [s]

compatible_size() [1/7]

bool gridpp::compatible_size	(	const Grid &	grid,
		const vec2 &	v
	)

Check if the grid is the same size as the 2D vector.

Parameters

grid	Grid (Y, X)
v	2D vector (Y, X)

Returns

True if compatible

compatible_size() [2/7]

bool gridpp::compatible_size	(	const Grid &	grid,
		const vec3 &	v
	)

Check if the grid is compatible with 3D vector.

Parameters

grid	Grid (Y, X)
v	3D vector (T, Y, X)

Returns

True if compatible

compatible_size() [3/7]

bool gridpp::compatible_size	(	const Points &	points,
		const vec &	v
	)

Check if the points are compatible with a 1D vector.

Parameters

grid	Points
v	1D vector

Returns

True if compatible

compatible_size() [4/7]

bool gridpp::compatible_size	(	const Points &	points,
		const vec2 &	v
	)

Check if the points are compatible with a 2D vector.

Parameters

grid	Points (P)
v	2D vector (T, P)

Returns

True if compatible

compatible_size() [5/7]

bool gridpp::compatible_size	(	const vec2 &	a,
		const vec2 &	b
	)

Check if two 2D vectors are compatible.

Parameters

a	2D vector (Y, X)
b	2D vector (Y, X)

Returns

True if compatible

compatible_size() [6/7]

bool gridpp::compatible_size	(	const vec2 &	a,
		const vec3 &	b
	)

Check if a 2D and 3D vectors are compatible.

Parameters

a	2D vector (Y, X)
b	3D vector (Y, X, E)

Returns

True if compatible

compatible_size() [7/7]

bool gridpp::compatible_size	(	const vec3 &	a,
		const vec3 &	b
	)

Check if two 3D vectors are compatible.

Parameters

a	3D vector (Y, X, E)
b	3D vector (Y, X, E)

Returns

True if compatible

convert_coordinates() [1/2]

bool gridpp::convert_coordinates	(	const vec &	lats,
		const vec &	lons,
		CoordinateType	type,
		vec &	x_coords,
		vec &	y_coords,
		vec &	z_coords
	)

Convert lats/lons or 2D x/y coordinates to 3D cartesian coordinates with the centre of the earth as the origin.

Parameters

lats	vector of latitudes [deg] or 2D y-coordinates [m]
lons	vector of longitudes [deg] or 2D x-coordinates [m]
type	Coordinate type for lats and lons
x_coords	vector of x-coordinates [m]
y_coords	vector of y-coordinates [m]
z_coords	vector of z-coordinates [m]

convert_coordinates() [2/2]

bool gridpp::convert_coordinates	(	float	lat,
		float	lon,
		CoordinateType	type,
		float &	x_coord,
		float &	y_coord,
		float &	z_coord
	)

Convert lat/lon or 2D x/y coordinate to 3D cartesian coordinate with the centre of the earth as the origin.

Parameters

lat	vector of latitudes [deg] or 2D y-coordinates [m]
lon	vector of longitudes [deg] or 2D x-coordinates [m]
type	Coordinate type for lat and lon
x_coord	x-coordinate [m]
y_coord	y-coordinate [m]
z_coord	z-coordinate [m]

count() [1/4]

vec gridpp::count	(	const Grid &	grid,
		const Points &	points,
		float	radius
	)

For each point, counts the number of gridpoints within the radius.

Parameters

grid	Input grid
points	Output points
radius	Radius [m]

Returns

1D vector of number of gridpoints

count() [2/4]

vec2 gridpp::count	(	const Grid &	igrid,
		const Grid &	ogrid,
		float	radius
	)

For each gridpoint, counts the number of gridpoints within the radius.

Parameters

igrid	Input grid
ogrid	Output grid
radius	Radius [m]

Returns

2D vector of number of gridpoints

count() [3/4]

vec2 gridpp::count	(	const Points &	points,
		const Grid &	grid,
		float	radius
	)

For each gridpoint, counts the number of points within the radius.

Parameters

points	Input points
grid	Output grid
radius	Radius [m]

Returns

2D vector of number of points

count() [4/4]

vec gridpp::count	(	const Points &	ipoints,
		const Points &	opoints,
		float	radius
	)

For each point, counts the number of points within the radius.

Parameters

ipoints	Input points (Pi)
opoints	Output points (Po)
radius	Radius [m]

Returns

1D vector of number of points (Po)

debug()

void gridpp::debug

(

std::string

string

)

Writes a debug message to standard out.

Parameters

string

Write this string

dewpoint() [1/2]

float gridpp::dewpoint	(	float	temperature,
		float	relative_humidity
	)

Calculate dewpoint temperature from temperature and relative humidity.

Parameters

temperature	Temperature [K]
relative_humidity	Relative humidity [1]

Returns

Dewpoint temperature [K]

dewpoint() [2/2]

vec gridpp::dewpoint	(	const vec &	temperature,
		const vec &	relative_humidity
	)

Vector version of dewpoint calculation.

Parameters

temperature	1D vector of temperature [K]
relative_humidity	1D vector of relative humidity [1]

Returns

1D vector of dewpoint temperature [K]

distance() [1/4]

vec gridpp::distance	(	const Grid &	grid,
		const Points &	points,
		int	num = 1
	)

For each point, calculates the distance to nearest gridpoint.

Parameters

grid	Grid
points	Points
num	Number of points

Returns

1D vector of distance [m] to nearest gridpoint for each point

distance() [2/4]

vec2 gridpp::distance	(	const Grid &	igrid,
		const Grid &	ogrid,
		int	num = 1
	)

For each output gridpoint, calculate the distance to nearest input gridpoint.

Parameters

grid	Grid (Yi, Xi)
ogrid	Output grid /Yo, Xo)
num	Number of points

Returns

2D vector of distance [m] to nearest gridpoint for each gridpoint (Yo, Xo)

distance() [3/4]

vec2 gridpp::distance	(	const Points &	points,
		const Grid &	grid,
		int	num = 1
	)

For each gridpoint, calculates the distance to nearest point.

Parameters

points	Points
grid	Grid
num	Number of points

Returns

2D vector of distance [m] to nearest point for each gridpoint

distance() [4/4]

vec gridpp::distance	(	const Points &	ipoints,
		const Points &	opoint,
		int	num = 1
	)

For each output point, calculates the distance to nearest input point.

Parameters

ipoints	Input points (Pi)
opoints	Output points (Po)
num	Number of points

Returns

1D vector of distance [m] to nearest point for each point (Po)

doping_circle()

vec2 gridpp::doping_circle	(	const Grid &	igrid,
		const vec2 &	background,
		const Points &	points,
		const vec &	observations,
		const vec &	radii,
		float	max_elev_diff = gridpp::MV
	)

Insert observations into gridded field using a circle.

Parameters

grid	Grid of background
background	2D vector of background values with (Y, X)
points	Points of observations
observations	1D vector of observations
radii	Radius of circle where observations are inserted for each point [m]
max_elev_diff	Only insert where elevation difference between grid and point is less than this value

Returns

2D vector of final field

doping_square()

vec2 gridpp::doping_square	(	const Grid &	igrid,
		const vec2 &	background,
		const Points &	points,
		const vec &	observations,
		const ivec &	halfwidths,
		float	max_elev_diff = gridpp::MV
	)

Insert observations into gridded field using a square box.

Parameters

grid	Grid of background
background	2D vector of background values (Y, X)
points	Points of observations
observations	1D vector of observations
halfwidths	Half width of square (in number of grid points) where observations are inserted for each point
max_elev_diff	Only insert where elevation difference between grid and point is less than this value [m]

Returns

2D vector of final field

downscale_probability()

vec2 gridpp::downscale_probability	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec3 &	ivalues,
		const vec2 &	threshold,
		const ComparisonOperator &	comparison_operator
	)

Nearest neighbour downscaling grid to grid and probability in one.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	3D vector of input values (Yi, Xi, E)
threshold	2D vector of threshold values (Yo, Xo)
comparison_operator	Comparison operator to create probability

Returns

2D vector of output values (Yo, Xo)

downscaling() [1/4]

vec2 gridpp::downscaling	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec2 &	ivalues,
		Downscaler	downscaler
	)

Downscale a gridded field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	2D vector of values on the input grid (Yi, Xi)
downscaler	Downscaling method

Returns

2D vector of output values (Yo, Xo)

downscaling() [2/4]

vec3 gridpp::downscaling	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec3 &	ivalues,
		Downscaler	downscaler
	)

Downscale a gridded ensemble field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	3D vector of values on the input grid (E, Yi, Xi)
downscaler	Downscaling method

Returns

3D vector of output values (E, Yo, Xo)

downscaling() [3/4]

vec gridpp::downscaling	(	const Grid &	igrid,
		const Points &	opoints,
		const vec2 &	ivalues,
		Downscaler	downscaler
	)

Downscale a gridded field to points.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	2D vector of values on the input grid (Yi, Xi)
downscaler	Downscaling method

Returns

1D vector of output values

downscaling() [4/4]

vec2 gridpp::downscaling	(	const Grid &	igrid,
		const Points &	opoints,
		const vec3 &	ivalues,
		Downscaler	downscaler
	)

Downscale a gridded ensemble field to points.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	3D vector of values on the input grid (E, Yi, Xi)
downscaler	Downscaling method

Returns

2D vector of output values (E, Points)

error()

void gridpp::error

(

std::string

string

)

Writes an error message to standard out.

Parameters

string

Write this string

fill()

vec2 gridpp::fill	(	const Grid &	igrid,
		const vec2 &	input,
		const Points &	points,
		const vec &	radii,
		float	value,
		bool	outside
	)

Fill in values inside or outside a set of circles (useful for masking)

Parameters

igrid	Grid of input
input	2D vector of values (Y, X)
points	Points to fill in
radii	1D vector of circle radii for each point [m]
value	Value to fill in
outside	If True, fill outside circles, if False, fill inside circles

Returns

2D vector of final field

fill_missing()

vec2 gridpp::fill_missing

(

const vec2 &

values

)

Fill in missing values based on nearby values.

Parameters

values

2D vector of values

Returns

2D vector of values without any missing values

full_gradient() [1/4]

vec2 gridpp::full_gradient	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec2 &	ivalues,
		const vec2 &	elev_gradient,
		const vec2 &	laf_gradient = vec2(),
		Downscaler	downscaler = Nearest
	)

Compute Downscale.

Parameters

igrid	input grid
ogrid	output grid
ivalues	values from igrid
elev_gradient	elevation gradient
laf_gradient	land area fraction gradientSpatially varying gradient dowscaling grid to grid for a deterministic field
igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	2D vector of input values (Yi, Xi)
elev_gradient	2D vector of elevation gradients (Yi, Xi) [input unit/m]
laf_gradient	2D vector of land-area-fraction gradients (Yi, Xi) [input unit/1]
downscaler	Downscaling method

Returns

2D vector of output values (Yo, Xo)

full_gradient() [2/4]

vec3 gridpp::full_gradient	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec3 &	ivalues,
		const vec3 &	elev_gradient,
		const vec3 &	laf_gradient,
		Downscaler	downscaler = Nearest
	)

Spatially varying gradient dowscaling grid to grid for an ensemble field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	3D vector of input values (E, Yi, Xi)
elev_gradient	3D vector of elevation gradients (Yi, Xi, E) [input unit/m]
laf_gradient	3D vector of land-area-fraction gradients (Yi, Xi, E) [input unit/1]
downscaler	Downscaling method

Returns

3D vector of output values (E, Yo, Xo)

full_gradient() [3/4]

vec gridpp::full_gradient	(	const Grid &	igrid,
		const Points &	opoints,
		const vec2 &	ivalues,
		const vec2 &	elev_gradient,
		const vec2 &	laf_gradient,
		Downscaler	downscaler = Nearest
	)

Spatially varying gradient dowscaling grid to point for a deterministic field.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	2D vector of input values grid (Yi, Xi)
elev_gradient	2D vector of elevation gradients (Yi, Xi) [input unit/m]
laf_gradient	2D vector of land-area-fraction gradients (Yi, Xi) [input unit/1]
downscaler	Downscaling method

Returns

1D vector of output values

full_gradient() [4/4]

vec2 gridpp::full_gradient	(	const Grid &	igrid,
		const Points &	opoints,
		const vec3 &	ivalues,
		const vec3 &	elev_gradient,
		const vec3 &	laf_gradient,
		Downscaler	downscaler = Nearest
	)

Spatially varying gradient dowscaling grid to point for an ensemble field.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	3D vector of input values (E, Yi, Xi)
elev_gradient	3D vector of elevation gradients (E, Yi, Xi) [input unit/m]
laf_gradient	3D vector of land-area-fraction gradients (E, Yi, Xi) [input unit/1]
downscaler	Downscaling method

Returns

2D vector of output values (E, Points)

full_gradient_debug()

vec3 gridpp::full_gradient_debug	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec2 &	ivalues,
		const vec2 &	elev_gradient,
		const vec2 &	laf_gradient = vec2(),
		Downscaler	downscaler = Nearest
	)

future_deprecation_warning()

void gridpp::future_deprecation_warning	(	std::string	function,
		std::string	other = ""
	)

Writes an deprecation warning to standard out.

Parameters

string	Name of the deprecated function
other	Additional message to write

gamma_inv()

vec gridpp::gamma_inv	(	const vec &	levels,
		const vec &	shape,
		const vec &	scale
	)

Extract quantiles from a gamma distribution.

Parameters

levels	1D vector of quantile levels to retrieve
shape	1D vector of shape parameter of gamma distribution
scale	1D vector of scale parameter of gamma distribution

Returns

1D vector of quantiles

get_debug_level()

int gridpp::get_debug_level

(

)

Get the currently set level of debug messagess.

Returns

Currently set debug level

get_lower_index()

int gridpp::get_lower_index	(	float	iX,
		const vec &	iValues
	)

Find the index in a vector that is equal to or just below a value.

Parameters

iX	Lookup value
iValues	1D vector of lookup values. Must be sorted.

Returns

The index into iValues that is equal to or just below iX

get_neighbourhood_thresholds() [1/2]

vec gridpp::get_neighbourhood_thresholds	(	const vec2 &	input,
		int	num_thresholds
	)

Calculate appropriate approximation thresholds for fast neighbourhood quantile.

Parameters

input	2D vector of values (Y, X)
num_thresholds	Number of thresholds

Returns

1D vector of approximation thresholds

get_neighbourhood_thresholds() [2/2]

vec gridpp::get_neighbourhood_thresholds	(	const vec3 &	input,
		int	num_thresholds
	)

Calculate appropriate approximation thresholds for neighbourhood quantile based on an ensemble.

Parameters

input	3D vector of values (Y, X, T)
num_thresholds	Number of thresholds

Returns

1D vector of approximation thresholds

get_omp_threads()

int gridpp::get_omp_threads

(

)

Get the number of OpenMP threads currently set.

Returns

Number of threads

get_optimal_threshold()

float gridpp::get_optimal_threshold	(	const vec &	curve_ref,
		const vec &	curve_fcst,
		float	threshold,
		gridpp::Metric	metric
	)

Compute the optimal threshold to remap an input threshold to.

Parameters

curve_ref	1D vector of reference curve values
curve_fcst	1D vector of reference curve values
threshold	Input threshold
metric	Metric to optimize

Returns

Optimal remapping threshold

get_statistic()

gridpp::Statistic gridpp::get_statistic

(

std::string

name

)

Convert name of a statistic enums.

Parameters

name	Name of a statistic

Returns

Statistic

get_upper_index()

int gridpp::get_upper_index	(	float	iX,
		const vec &	iValues
	)

Find the index in a vector that is equal to or just above a value.

Parameters

iX	Lookup value
iValues	1D vector of lookup values. Must be sorted.

Returns

The index into iValues that is equal to or just above iX

gridding() [1/2]

vec2 gridpp::gridding	(	const Grid &	grid,
		const Points &	points,
		const vec &	values,
		float	radius,
		int	min_num,
		gridpp::Statistic	statistic
	)

Aggregate points onto a grid.

Writes gridpp::MV where there are not enough points.

Parameters

grid	Output grid to aggregate to
points	Input points
values	1D vector of input values
radius	Circle radius for aggregating points [m]
min_num	Minimum number of points in radius to create a value
statistic	Statistic to compute on points within radius

Returns

2D vector of output values

gridding() [2/2]

vec gridpp::gridding	(	const Points &	opoints,
		const Points &	ipoints,
		const vec &	values,
		float	radius,
		int	min_num,
		gridpp::Statistic	statistic
	)

Aggregate points onto a points.

Writes gridpp::MV where there are not enough points.

Parameters

opoints	Output points to aggregate to
ipoints	Input points
values	1D vector of input Values
radius	Circle radius for aggregate points [m]
min_num	Minimum number of points in radius to create a value
statistic	Statistic to compute on points within radius

Returns

1D vector of output values

gridding_nearest() [1/2]

vec2 gridpp::gridding_nearest	(	const Grid &	grid,
		const Points &	points,
		const vec &	values,
		int	min_num,
		gridpp::Statistic	statistic
	)

Assign each point to nearest neighbour in grid and aggregate values.

Writes gridpp::MV where there are not enough observations.

Parameters

grid	Output grid to aggregate to
points	Input points
values	Input values
min_num	Minimum number of points in a gridpoint to create a value
statistic	Statistic to compute on points within gridpoint

Returns

2D vector of output values

gridding_nearest() [2/2]

vec gridpp::gridding_nearest	(	const Points &	opoints,
		const Points &	ipoints,
		const vec &	values,
		int	min_num,
		gridpp::Statistic	statistic
	)

Assign each ipoint to nearest neighbour in opoints and aggregate values.

Writes gridpp::MV where there are not enough observations.

Parameters

opoints	Output points to aggregate to
ipoints	Input points
values	Input values
min_num	Minimum number of points in a gridpoint to create a value
statistic	Statistic to compute on points within gridpoint

Returns

1D vector of output values

init_ivec2()

ivec2 gridpp::init_ivec2	(	int	Y,
		int	X,
		int	value
	)

Initialize a 2D integer vector of size Y, X, with a given value.

Parameters

Y	Size of first dimension
X	Size of second dimension value Initialize with this value

Returns

2D interger vector

init_ivec3()

ivec3 gridpp::init_ivec3	(	int	Y,
		int	X,
		int	E,
		int	value
	)

Initialize a 3D integer vector of size Y, X, E, with a given value.

Parameters

Y	Size of first dimension
X	Size of second dimension
E	Size of third dimension value Initialize with this value

Returns

3D integer vector

init_vec2()

vec2 gridpp::init_vec2	(	int	Y,
		int	X,
		float	value = MV
	)

Initialize a 2D float vector of size Y, X, with a given value.

Parameters

Y	Size of first dimension
X	Size of second dimension value Initialize with this value

Returns

2D float vector

init_vec3()

vec3 gridpp::init_vec3	(	int	Y,
		int	X,
		int	E,
		float	value = MV
	)

Initialize a 3D float vector of size Y, X, E, with a given value.

Parameters

Y	Size of first dimension
X	Size of second dimension
E	Size of third dimension value Initialize with this value

Returns

3D float vector

initialize_omp()

void gridpp::initialize_omp

(

)

Sets the number of OpenMP threads to 1 if OMP_NUM_THREADS undefined.

interpolate() [1/2]

float gridpp::interpolate	(	float	x,
		const vec &	iX,
		const vec &	iY
	)

Piecewise linear interpolation If x is outside the range of iX, then the min/max value of iY is used.

If there are multiple identical x values, then the average of the y values at each end of the x-interval that is constant is used. The exception is if the constant interval is on one (only) of the edges of the interpolation curve. In that case, the y-value at the end of the interval further away from the boundary of the curve is used.

Parameters

x	Interpolation to this point
iX	1D vector of x-axis values. Vector must be sorted.
iY	1D vector of y-axis values corresponding to iX.

Returns

Y value corresponding to x

interpolate() [2/2]

vec gridpp::interpolate	(	const vec &	x,
		const vec &	iX,
		const vec &	iY
	)

Piecewise linear interpolation, vectorised version.

Parameters

x	1D vector of values to interpolate to
iX	1D vector of x-axis values. Vector must be sorted.
iY	1D vector of y-axis values corresponding to iX.

Returns

1D vector of y-axis values corresponding to x

is_valid()

bool gridpp::is_valid

(

float

value

)

Checks if a value is valid.

Parameters

value

Value to check

Returns

True if the value is valid, False if it is NaN, Inf, or other invalid value

is_valid_lat()

bool gridpp::is_valid_lat	(	float	lat,
		CoordinateType	type
	)

Checks that a lat-coordinate is valid (based on the coordinate type)

Parameters

lat	latitude [deg] or y-coordinate [m]
type	Coordinate type for lat

Returns

True if lat is valid latitude or y-coordinate

is_valid_lon()

bool gridpp::is_valid_lon	(	float	lon,
		CoordinateType	type
	)

Checks that a lon-coordinate is valid (based on the coordinate type)

Parameters

lat	longitude [deg] or x-coordinate [m]
type	Coordinate type for lon

Returns

True if lon is valid latitude or x-coordinate

local_distribution_correction() [1/2]

vec2 gridpp::local_distribution_correction	(	const Grid &	bgrid,
		const vec2 &	background,
		const Points &	obs_points,
		const vec &	obs,
		const vec &	background_at_points,
		const StructureFunction &	structure,
		float	min_quantile,
		float	max_quantile,
		int	min_points = 0
	)

Correction of a gridded field ensuring the distribution of values nearby match that of observations.

This is an experimental method.

Parameters

bgrid	Grid of background field
background	2D vector of background values (Y, X)
obs_points	Points of observations
obs	1D vector of observations
background_at_points	1D vector of background values at observation points
structure	Structure function
min_quantile	Truncate quantile map below this quantile level
max_quantile	Truncate quantile map above this quantile level
max_points	Maximum number of points used within localization radius (not used at moment)

Returns

2D vector of corrected values

local_distribution_correction() [2/2]

vec2 gridpp::local_distribution_correction	(	const Grid &	bgrid,
		const vec2 &	background,
		const Points &	obs_points,
		const vec2 &	obs,
		const vec2 &	background_at_points,
		const StructureFunction &	structure,
		float	min_quantile,
		float	max_quantile,
		int	min_points = 0
	)

Correction of a gridded field ensuring the distribution of values nearby match that of observations.

This is an experimental method. Version with multiple number of timesteps. Pool all observations across time in when computing the calibration curve.

Parameters

bgrid	Grid of background
background	2D vector of background values (Y, X)
obs_points	Points of observations
obs	2D vector of observations (Time, Points)
background_at_points	2D vector of background values at observation points (Time, Points)
structure	structure function
min_quantile	Truncate quantile map below this quantile
max_quantile	Truncate quantile map above this quantile
max_points	Maximum number of points used within localization radius (not used at moment)

Returns

2D vector of corrected values

mask_threshold_downscale_consensus()

vec2 gridpp::mask_threshold_downscale_consensus	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec3 &	ivalues_true,
		const vec3 &	ivalues_false,
		const vec3 &	theshold_values,
		const vec2 &	threshold,
		const ComparisonOperator &	comparison_operator,
		const Statistic &	statistic
	)

Masked ensemble downscaling and consensus.

This method computes a high-resolution consensus from a low-resolution ensemble, without downscaling all members. See the wiki for more details.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues_true	3D vector of input values to use when condition is true (Yi, Xi, E)
ivalues_false	3D vector of input values to use when condition is false (Yi, Xi, E)
threshold_values	3D vector of values (Yi, Xi, E) defining the mask for ivalues after applying the theshold
threshold	2D vector of thresholds on output grid that is upscaled to input grid (Yo, Xo)
comparison_operator	Comparison operator to create probability
statistic	statistic to compute over the ensemble dimension after downscaling

Returns

2D vector of output values (Yo, Xo)

mask_threshold_downscale_quantile()

vec2 gridpp::mask_threshold_downscale_quantile	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec3 &	ivalues_true,
		const vec3 &	ivalues_false,
		const vec3 &	theshold_values,
		const vec2 &	threshold,
		const ComparisonOperator &	comparison_operator,
		const float	quantile_level
	)

Masked ensemble downscaling and quantile extraction.

This method extractes a quantile for a high-resolution grid from a low-resolution ensemble, without downscaling all members. See the wiki for more details.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues_true	3D vector of input values to use when condition is true (Yi, Xi, E)
ivalues_false	3D vector of input values to use when condition is false (Yi, Xi, E)
threshold_values	3D vector of values (Yi, Xi, E) defining the mask for ivalues after applying the theshold
threshold	2D vector of thresholds on output grid that is upscaled to input grid (Yo, Xo)
comparison_operator	Comparison operator to create probability
quantile_level	Quantile level (between 0 and 1) to extract

Returns

2D vector of output values (Yo, Xo)

metric_optimizer_curve()

vec gridpp::metric_optimizer_curve	(	const vec &	ref,
		const vec &	fcst,
		const vec &	thresholds,
		gridpp::Metric	metric,
		vec &	output_fcst
	)

Create calibration curve that optimizes a metric.

Parameters

ref	1D vector of reference values (observations)
fcst	1d vector of forecast values
thresholds	1D vector of thresholds for computing optimal values for
metric	A Metric to optimize for
output_fcst	1D vector of output forecast quantiles

Returns

1D vector of output reference quantiles

monotonize_curve()

vec gridpp::monotonize_curve	(	vec	curve_ref,
		vec	curve_fcst,
		vec &	output_fcst
	)

Ensure calibration curve is monotonic, by removing points on the curve.

Parameters

curve_ref	1D vector of reference curve values
curve_fcst	1D vector of forecast curve values
output_fcst	1D vector of output forecast curve values

Returns

1D vector of output reference curve values

nearest() [1/8]

vec2 gridpp::nearest	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec2 &	ivalues
	)

Nearest neighbour dowscaling grid to grid for a deterministic field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	2D vector of input values (Yi, Xi)

Returns

2D vector of output values (Yo, Xo)

nearest() [2/8]

vec3 gridpp::nearest	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec3 &	ivalues
	)

Nearest neighbour dowscaling grid to grid for an ensemble field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	3D vector of input values (E, Yi, Xi)

Returns

3D vector of output values (E, Yo, Xo)

nearest() [3/8]

vec gridpp::nearest	(	const Grid &	igrid,
		const Points &	opoints,
		const vec2 &	ivalues
	)

Nearest neighbour dowscaling grid to point for a deterministic field.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	2D vector of input values grid (Yi, Xi)

Returns

1D vector of output values

nearest() [4/8]

vec2 gridpp::nearest	(	const Grid &	igrid,
		const Points &	opoints,
		const vec3 &	ivalues
	)

Nearest neighbour dowscaling grid to point for an ensemble field.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	3D vector of input values (E, Yi, Xi)

Returns

2D vector of output values (E, Points)

nearest() [5/8]

vec gridpp::nearest	(	const Points &	ipoints,
		const Points &	opoints,
		const vec &	ivalues
	)

Nearest neighbour dowscaling point to point for a deterministic field.

Parameters

ipoints	Points of input
opoints	Points of downscaled output
ivalues	1D vector of input values

Returns

Values 1D vector of output values

nearest() [6/8]

vec2 gridpp::nearest	(	const Points &	ipoints,
		const Points &	opoints,
		const vec2 &	ivalues
	)

Nearest neighbour dowscaling point to point for an ensemble field.

Parameters

ipoints	Points of input (Pi)
opoints	Points of downscaled output Po)
ivalues	2D vector of input values (E, Pi)

Returns

2D vector of output values (E, Po)

nearest() [7/8]

vec2 gridpp::nearest	(	const Points &	ipoints,
		const Grid &	ogrid,
		const vec &	ivalues
	)

Nearest neighbour dowscaling point to grid for a deterministic field.

Parameters

ipoints	Points of input
ogrid	Grid of downscaled output
ivalues	1D vector of values for the input points

Returns

2D vector of output values (Y, X)

nearest() [8/8]

vec3 gridpp::nearest	(	const Points &	ipoints,
		const Grid &	ogrid,
		const vec2 &	ivalues
	)

Nearest neighbour dowscaling point to grid for an ensemble field.

Parameters

ipoints	Points of input
ogrid	Grid of downscaled output
ivalues	2D vector input values (E, Points)

Returns

3D vector of output values (E, Y, X)

neighbourhood() [1/2]

vec2 gridpp::neighbourhood	(	const vec2 &	input,
		int	halfwidth,
		gridpp::Statistic	statistic
	)

Spatial neighbourhood filter, computing a statistic for a sliding square window.

Parameters

input	2D vector of input values (Y, X)
halfwidth	Filter halfwidth in number of gridpoints
statistic	Statistic to compute

Returns

2D vector of final field (Y, x)

neighbourhood() [2/2]

vec2 gridpp::neighbourhood	(	const vec3 &	input,
		int	halfwidth,
		gridpp::Statistic	statistic
	)

Spatial neighbourhood filter for an ensemble of fields.

Computes static across neighbourhood and all members

Parameters

input	3D vector of input values (Y, X, E)
halfwidth	Filter halfwidth in number of gridpoints
statistic	Statistic to compute

Returns

2D vector of final field (Y, x)

neighbourhood_brute_force() [1/2]

vec2 gridpp::neighbourhood_brute_force	(	const vec2 &	input,
		int	halfwidth,
		gridpp::Statistic	statistic
	)

Spatial neighbourhood filter without any shortcuts.

This is quite slow and is only useful for testing.

Parameters

input	2D vector of values (Y, X)
halfwidth	Filter halfwidth in number of gridpoints
statistic	Statistic to compute

Returns

2D vector of final field (Y, X)

neighbourhood_brute_force() [2/2]

vec2 gridpp::neighbourhood_brute_force	(	const vec3 &	input,
		int	halfwidth,
		gridpp::Statistic	statistic
	)

Spatial ensemble neighbourhood filter without any shortcuts.

This is quite slow and is only useful for testing.

Parameters

input	3D vector of values (Y, X, E)
halfwidth	Filter halfwidth in number of gridpoints
statistic	Statistic to compute

Returns

2D vector of final field (Y, X)

neighbourhood_ens()

vec2 gridpp::neighbourhood_ens	(	const vec3 &	input,
		int	halfwidth,
		gridpp::Statistic	statistic
	)

Deprecated: Compute neighbourhood statistic on ensemble field.

Deprecated:

Use neighbourhood() function

neighbourhood_quantile() [1/2]

vec2 gridpp::neighbourhood_quantile	(	const vec2 &	input,
		float	quantile,
		int	halfwidth
	)

Computes a quantile in a sliding square neighbourhood.

Parameters

input	2D vector of values
quantile	Quantile to compute (between 0 and 1)
halfwidth	Filter halfwidth in number of gridpoints

Returns

2D vector of final field (Y, x)

neighbourhood_quantile() [2/2]

vec2 gridpp::neighbourhood_quantile	(	const vec3 &	input,
		float	quantile,
		int	halfwidth
	)

Computes a quantile in a sliding square neighbourhood across ensemble members.

Parameters

input	3D vector of values (Y, X, E)
quantile	Quantile to compute (between 0 and 1)
halfwidth	Filter halfwidth in number of gridpoints

Returns

2D vector of final field (Y, x)

neighbourhood_quantile_ens()

vec2 gridpp::neighbourhood_quantile_ens	(	const vec3 &	input,
		float	quantile,
		int	halfwidth
	)

Deprecated: Compute neighbourhood quantiles on ensemble field.

Deprecated:

Use neighbourhood_quantile() function

neighbourhood_quantile_ens_fast()

vec2 gridpp::neighbourhood_quantile_ens_fast	(	const vec3 &	input,
		float	quantile,
		int	radius,
		const vec &	thresholds
	)

Deprecated: Compute neighbourhood quantiles fast on ensemble field.

Deprecated:

Use neighbourhood_quantile_fast() function

neighbourhood_quantile_fast() [1/4]

vec2 gridpp::neighbourhood_quantile_fast	(	const vec2 &	input,
		float	quantile,
		int	halfwidth,
		const vec &	thresholds
	)

Fast and approximate neighbourhood quantile.

Parameters

input	2D vector of values (Y, X)
quantile	Quantile to compute (between 0 and 1)
halfwidth	Filter halfwidth in number of gridpoints
thresholds	1D vector of thresholds to use to approximate value

Returns

2D vector of final field (Y, X)

neighbourhood_quantile_fast() [2/4]

vec2 gridpp::neighbourhood_quantile_fast	(	const vec3 &	input,
		float	quantile,
		int	halfwidth,
		const vec &	thresholds
	)

Fast and approximate neighbourhood quantile for ensemble of fields.

Parameters

input	3D vector of values (Y, X, E)
quantile	Quantile to compute (between 0 and 1)
halfwidth	Filter halfwidth in number of gridpoints
thresholds	1D vector of thresholds to use to approximate value

Returns

2D vector of final field (Y, X)

neighbourhood_quantile_fast() [3/4]

vec2 gridpp::neighbourhood_quantile_fast	(	const vec2 &	input,
		const vec2 &	quantile,
		int	halfwidth,
		const vec &	thresholds
	)

Fast and approximate neighbourhood quantile, with spatially varying quantile levels.

Parameters

input	2D vector of values (Y, X)
quantile	2D vector quantile levels to compute (between 0 and 1) (Y, X)
halfwidth	Filter halfwidth in number of gridpoints
thresholds	1D vector of thresholds to use to approximate value

Returns

2D vector of final field (Y, X)

neighbourhood_quantile_fast() [4/4]

vec2 gridpp::neighbourhood_quantile_fast	(	const vec3 &	input,
		const vec2 &	quantile,
		int	halfwidth,
		const vec &	thresholds
	)

Fast and approximate neighbourhood quantile for ensemble of fields, with spatially varying quantile.

Parameters

input	3D vector of values with dimensions (Y, X, E)
quantile	2D vector of quantile levels to compute (between 0 and 1) (Y, X)
halfwidth	Filter halfwidth in number of gridpoints
thresholds	1D vector of thresholds to use to approximate value

Returns

2D vector of final field (Y, X)

neighbourhood_score()

vec2 gridpp::neighbourhood_score	(	const Grid &	grid,
		const Points &	points,
		const vec2 &	fcst,
		const vec &	ref,
		int	half_width,
		gridpp::Metric	metric,
		float	threshold
	)

Compute a score for a metric of all points within a radius.

This is an experimental method.

neighbourhood_search()

vec2 gridpp::neighbourhood_search	(	const vec2 &	array,
		const vec2 &	search_array,
		int	halfwidth,
		float	search_target_min,
		float	search_target_max,
		float	search_delta,
		const ivec2 &	apply_array = ivec2()
	)

Find suitable value in neighbourhood that match a search criteria.

If values are found within the range, then the average of the values are used. If not, the closest point to the range is used, provided this is within search_delta of the criteria range. if no point fulfills this, the original value is not modified.

Parameters

input	2D vector of input values, e.g. temperatures (Y, X)
search_array	2D vector to search in, e.g. land area fraction (Y, X)
halfwidth	Neighbourhood halfwidth to search in
search_target_min	Lower bound of criteria, e.g. a land area fraction value
search_target_Max	Upper bound of criteria, e.g. a land area fraction value
search_delta	Max distance outside target range where the most suitable point is still used
apply_array	2D integer vector of 1 (correct this grid point) or 0 (don't correct)

Returns

2D vector of output values (Y, X)

num_missing_values()

int gridpp::num_missing_values

(

const vec2 &

iArray

)

Count the number of missing values in a vector.

Parameters

array

2D vector

Returns

Number of invalid values

optimal_interpolation() [1/2]

vec2 gridpp::optimal_interpolation	(	const Grid &	bgrid,
		const vec2 &	background,
		const Points &	obs_points,
		const vec &	obs,
		const vec &	variance_ratios,
		const vec &	background_at_points,
		const StructureFunction &	structure,
		int	max_points,
		bool	allow_extrapolation = true
	)

Optimal interpolation for a deterministic gridded field.

Parameters

bgrid	Grid of background field
background	2D field of background values
obs_points	Points of observations
obs	1D vector of observations
variance_ratios	1D vector of ratio of observation to background error variance
background_at_points	1D vector of background at observation points
structure	Structure function
max_points	Maximum number of observations to use inside localization zone; Use 0 to disable
allow_extrapolation	Allow OI to extrapolate increments outside increments at observations points

Returns

2D vector of analised values

optimal_interpolation() [2/2]

vec gridpp::optimal_interpolation	(	const Points &	bpoints,
		const vec &	background,
		const Points &	obs_points,
		const vec &	obs,
		const vec &	variance_ratios,
		const vec &	background_at_points,
		const StructureFunction &	structure,
		int	max_points,
		bool	allow_extrapolation = true
	)

Optimal interpolation for a deterministic vector of points.

Parameters

bpoints	Points of background field
background	1D field of background values
obs_points	Points of observations
obs	1D vector of observations
variance_ratios	1D vector of ratio of observation to background error variance
background_at_points	1D vector of background at observation points
structure	Structure function
max_points	Maximum number of observations to use inside localization zone; Use 0 to disable
allow_extrapolation	Allow OI to extrapolate increments outside increments at observations points

Returns

1D vector of analised values

optimal_interpolation_ensi() [1/2]

vec3 gridpp::optimal_interpolation_ensi	(	const Grid &	bgrid,
		const vec3 &	background,
		const Points &	obs_points,
		const vec &	obs,
		const vec &	obs_standard_deviations,
		const vec2 &	background_at_points,
		const StructureFunction &	structure,
		int	max_points,
		bool	allow_extrapolation = true
	)

Optimal interpolation for an ensemble gridded field See Lussana et al 2019 (DOI: 10.1002/qj.3646)

Parameters

bgrid	Grid of background field
background	3D vector of background values (Y, X, E)
obs_points	observation points
obs	1D vector of observations
obs_standard_deviations	1D vector of observation standard deviations
background_at_points	2D vector of background at observation points (L, E)
structure	Structure function
max_points	Maximum number of observations to use inside localization zone; Use 0 to disable
allow_extrapolation	Allow OI to extrapolate increments outside increments at observations

Returns

3D vector of analised values (Y, X, E)

optimal_interpolation_ensi() [2/2]

vec2 gridpp::optimal_interpolation_ensi	(	const Points &	bpoints,
		const vec2 &	background,
		const Points &	obs_points,
		const vec &	obs,
		const vec &	obs_standard_deviations,
		const vec2 &	background_at_points,
		const StructureFunction &	structure,
		int	max_points,
		bool	allow_extrapolation = true
	)

Optimal interpolation for an ensemble point field See Lussana et al 2019 (DOI: 10.1002/qj.3646)

Parameters

bpoints	Points of background field
background	2D vector of background values (L, E)
obs_points	Observation points
obs	1D vector of observations
obs_standard_deviations	1D vector of observation standard deviations
background_at_points	2D vector of background at observation points (L, E)
structure	Structure function
max_points	Maximum number of observations to use inside localization zone; Use 0 to disable
allow_extrapolation	Allow OI to extrapolate increments outside increments at observations

Returns

2D vector of analised values (L, E)

optimal_interpolation_full() [1/2]

vec2 gridpp::optimal_interpolation_full	(	const Grid &	bgrid,
		const vec2 &	background,
		const vec2 &	bvariance,
		const Points &	obs_points,
		const vec &	obs,
		const vec &	obs_variance,
		const vec &	background_at_points,
		const vec &	bvariance_at_points,
		const StructureFunction &	structure,
		int	max_points,
		vec2 &	analysis_variance,
		bool	allow_extrapolation = true
	)

Optimal interpolation for a deterministic gridded field including analysis variance.

Parameters

bgrid	Grid of background field
background	2D vector of background values
bvariance	2D vector of background variances
obs_points	Points of observations
obs	1D vector of observations
obs_variance	1D vector of observation variances
background_at_points	1D vector of background at observation points
bvariance_at_points	1D vector of background variance at observation points
structure	Structure function
max_points	Maximum number of observations to use inside localization zone; Use 0 to disable
analysis_variance	2D output vector of analysis variance
allow_extrapolation	Allow OI to extrapolate increments outside increments at observations

Returns

2D vector of analysed values

optimal_interpolation_full() [2/2]

vec gridpp::optimal_interpolation_full	(	const Points &	bpoints,
		const vec &	background,
		const vec &	bvariance,
		const Points &	obs_points,
		const vec &	obs,
		const vec &	obs_variance,
		const vec &	background_at_points,
		const vec &	bvariance_at_points,
		const StructureFunction &	structure,
		int	max_points,
		vec &	analysis_variance,
		bool	allow_extrapolation = true
	)

Optimal interpolation for a deterministic vector of points including analysis variance.

Parameters

bpoints	Points of background field
background	1D vector of background values
bvariance	1D vector of background variance
obs_points	Points of observations
obs	1D vector of observations
obs_variance	1D vector of observation variances
background_at_points	1D vector of background at observation points
bvariance_at_points	1D vector of background variance at observation points
structure	Structure function
max_points	Maximum number of observations to use inside localization zone; Use 0 to disable
analysis_variance	1D output vector of analysis variance
allow_extrapolation	Allow OI to extrapolate increments outside increments at observations

Returns

2D vector of analised values

point_in_rectangle()

bool gridpp::point_in_rectangle	(	const Point &	A,
		const Point &	B,
		const Point &	C,
		const Point &	D,
		const Point &	m
	)

Checks if a point is located inside a rectangle formed by 4 points.

The 4 points must be provided in an order that draws out a rectangle (either clockwise or counter-clockwise).

Parameters

A	A point in the rectangle
B	A point in the rectangle
C	A point in the rectangle
D	A point in the rectangle
m	The point to test if it is inside

Returns

True if the point is inside, False otherwise

pressure() [1/2]

float gridpp::pressure	(	float	ielev,
		float	oelev,
		float	ipressure,
		float	itemperature = 288.15
	)

Calculate pressure at a new elevation.

Parameters

ielev	Elevation at start point [m]
oelev	Elevation at new point [m]
ipressure	Pressure at start point [Pa]
itemperature	Temperature at start point [K]

Returns

Pressure at new point [Pa]

pressure() [2/2]

vec gridpp::pressure	(	const vec &	ielev,
		const vec &	oelev,
		const vec &	ipressure,
		const vec &	itemperature
	)

Calculate Vector version of pressure calculation.

Parameters

ielev	1D vector of elevation at start point [m]
oelev	1D vector of elevation at new point [m]
ipressure	1D vector of pressure at start point [Pa]
itemperature	1D vector of temperature at start point [K]

Returns

1D vector of pressure at new points [Pa]

qnh() [1/2]

float gridpp::qnh	(	float	pressure,
		float	altitude
	)

Diagnose QNH from pressure and altitude.

Parameters

pressure	Pressure at point [Pa]
altitude	Altitude of point [m]

Returns

QNH [Pa]

qnh() [2/2]

vec gridpp::qnh	(	const vec &	pressure,
		const vec &	altitude
	)

Vector version of QNH calculation.

Parameters

pressure	1D vector of Pressure [Pa]
altitude	1D vector of altitude [m]

Returns

1D vector of QNH [Pa]

quantile_mapping_curve()

vec gridpp::quantile_mapping_curve	(	const vec &	ref,
		const vec &	fcst,
		vec &	output_fcst,
		vec	quantiles = vec()
	)

Create quantile mapping calibration curve.

Parameters

ref	1D vector of reference values (observations)
fcst	1D vector of forecast values
output_fcst	1D vector of output forecast quantiles
quantiles	1D vector of quantile levels to extract. If empty, use all values.

Returns

1D vector of output reference quantiles

relative_humidity() [1/2]

float gridpp::relative_humidity	(	float	temperature,
		float	dewpoint
	)

Calculate relative humidity from temperature and dewpoint temperature.

Parameters

temperature	Temperature [K]
dewpoint	Dewpoint temperature [K]

Returns

Relative humidity [1]

relative_humidity() [2/2]

vec gridpp::relative_humidity	(	const vec &	temperature,
		const vec &	dewpoint
	)

Vector version of relative humidity calculation.

Parameters

temperature	1D vector of temperature [K]
dewpoint	1D vector of dewpoint temperature [K]

Returns

1D vector of relative humidity [1]

sea_level_pressure() [1/2]

float gridpp::sea_level_pressure	(	float	ps,
		float	altitude,
		float	temperature,
		float	rh = gridpp::MV,
		float	dewpoint = gridpp::MV
	)

Convert Surface Pressure to Sea Level Pressure.

Parameters

ps	Surface pressure [Pa]
altitude	Station altitude above sea level [m]
temperature	2m temperature [K]
rh	2m Relative humidity [1]
dewpoint	2m Dewpoint Temperature at station [K]

Returns

Sea Level Pressure [Pa]

sea_level_pressure() [2/2]

vec gridpp::sea_level_pressure	(	const vec &	ps,
		const vec &	altitude,
		const vec &	temperature,
		const vec &	rh,
		const vec &	dewpoint
	)

Vector version of Convert Surface Pressure to Sea Level Pressure.

Parameters

ps	1D vector of Surface pressures [Pa]
altitude	1D vector of station altitudes above sea level [m]
temperature	1D vector of temperature [K]
rh	1D vector of relative humidity [1]
dewpoint	1D vector of dewpoint temperature [K]

Returns

1D vector of sea Level pressure [Pa]

set_debug_level()

void gridpp::set_debug_level

(

int

level

)

Set the verbosity of debug messages.

Use 0 for no messages.s

Parameters

level

Debug level

set_omp_threads()

void gridpp::set_omp_threads

(

int

num

)

Set the number of OpenMP threads to use.

Overrides OMP_NUM_THREAD env variable.

Parameters

num	Number of threads

simple_gradient() [1/4]

vec2 gridpp::simple_gradient	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec2 &	ivalues,
		float	elev_gradient,
		Downscaler	downscaler = Nearest
	)

Elevation correction dowscaling grid to grid for a deterministic field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	2D vector of input values (Yi, Xi)
elev_gradient	Elevation gradient [input unit/m]
downscaler	Downscaling method

Returns

2D vector of output values (Yo, Xo)

simple_gradient() [2/4]

vec3 gridpp::simple_gradient	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec3 &	ivalues,
		float	elev_gradient,
		Downscaler	downscaler = Nearest
	)

Elevation correction dowscaling grid to grid for an ensemble field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	3D vector of input values (E, Yi, Xi)
elev_gradient	Elevation gradient [input unit/m]
downscaler	Downscaling method

Returns

3D vector of output values (E, Yo, Xo)

simple_gradient() [3/4]

vec gridpp::simple_gradient	(	const Grid &	igrid,
		const Points &	opoints,
		const vec2 &	ivalues,
		float	elev_gradient,
		Downscaler	downscaler = Nearest
	)

Elevation correction dowscaling grid to point for a deterministic field.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	2D vector of input values grid (Yi, Xi)
elev_gradient	Elevation gradient [input unit/m]
downscaler	Downscaling method

Returns

1D vector of output values

simple_gradient() [4/4]

vec2 gridpp::simple_gradient	(	const Grid &	igrid,
		const Points &	opoints,
		const vec3 &	ivalues,
		float	elev_gradient,
		Downscaler	downscaler = Nearest
	)

Elevation correction dowscaling grid to point for an ensemble field.

Parameters

igrid	Grid of input (Yi, Xi)
opoints	Points of downscaled output
ivalues	3D vector of input values (E, Yi, Xi)
elev_gradient	Elevation gradient [input unit/m]
downscaler	Downscaling method

Returns

2D vector of output values (E, Points)

smart()

vec2 gridpp::smart	(	const Grid &	igrid,
		const Grid &	ogrid,
		const vec2 &	ivalues,
		int	num,
		const StructureFunction &	structure
	)

Smart neighbour downscaling grid to grid for a deterministic field.

Parameters

igrid	Grid of input (Yi, Xi)
ogrid	Grid of downscaled output (Yo, Xo)
ivalues	2D vector of input values (Yi, Xi)
num	Number of neighbours to average
structure	Structure function for determining how similar neighbour gridpoints are

Returns

2D vector of output values (Yo, Xo)

test_array()

float * gridpp::test_array	(	float *	v,
		int	n
	)

Special function whose presense is needed for SWIG.

test_ivec2_output()

ivec2 gridpp::test_ivec2_output

(

)

test_ivec3_output()

ivec3 gridpp::test_ivec3_output

(

)

test_ivec_input()

int gridpp::test_ivec_input

(

const ivec &

input

)

Testing function for 1D input vector.

test_ivec_output()

ivec gridpp::test_ivec_output

(

)

test_not_implemented_exception()

void gridpp::test_not_implemented_exception

(

)

test_vec2_argout()

float gridpp::test_vec2_argout

(

vec2 &

distances

)

Testing function for 2D vector treated as output.

test_vec2_input()

float gridpp::test_vec2_input

(

const vec2 &

input

)

Testing function for 2D input vector.

test_vec2_output()

vec2 gridpp::test_vec2_output

(

)

Testing function for 2D output vector.

test_vec3_input()

float gridpp::test_vec3_input

(

const vec3 &

input

)

Testing function for 3D input vector.

test_vec3_output()

vec3 gridpp::test_vec3_output

(

)

Testing function for 3D output vector.

test_vec_argout()

float gridpp::test_vec_argout

(

vec &

distances

)

Testing function for 1D vector treated as output.

test_vec_input()

float gridpp::test_vec_input

(

const vec &

input

)

Testing function for 1D input vector.

test_vec_output()

vec gridpp::test_vec_output

(

)

Testing function for 1D output vector.

version()

std::string gridpp::version

(

)

The gridpp version.

Returns

The gridpp version

warning()

void gridpp::warning

(

std::string

string

)

Writes a warning message to standard out.

Parameters

string

Write this string

wetbulb() [1/2]

float gridpp::wetbulb	(	float	temperature,
		float	pressure,
		float	relative_humidity
	)

Calculate wetbulb temperature from temperature, pressure, and relative humidity.

Parameters

temperature	Temperature [K]
pressure	Air pressure [Pa]
relative_humidity	Relative humidity [1]

Returns

Wetbulb temperature [K]

wetbulb() [2/2]

vec gridpp::wetbulb	(	const vec &	temperature,
		const vec &	pressure,
		const vec &	relative_humidity
	)

Vector version of wetbulb calculation.

Parameters

temperature	1D vector of temperature [K]
pressure	1D vector of air pressure [pa]
relative_humidity	1D vector of Relative humidity [1]

Returns

Wetbulb temperatures [K]

wind_direction() [1/2]

float gridpp::wind_direction	(	float	xwind,
		float	ywind
	)

Diagnose wind direction from its components.

If both xwind and ywind are 0, then direction is 180

Parameters

xwind	X-component of wind [any unit]
ywind	Y-component of wind [any unit]

Returns

Wind direction [degrees]

wind_direction() [2/2]

vec gridpp::wind_direction	(	const vec &	xwind,
		const vec &	ywind
	)

Vector version of wind direction calculation.

Parameters

xwind	1D vector of X-component of wind [any unit]
ywind	1D vector of Y-component of wind [any unit]

Returns

1D vector of wind direction [degrees]

wind_speed() [1/2]

float gridpp::wind_speed	(	float	xwind,
		float	ywind
	)

Diagnose wind speed from its components.

Parameters

xwind	X-component of wind [any unit]
ywind	Y-component of wind [any unit]

Returns

Wind speed [any unit]

wind_speed() [2/2]

vec gridpp::wind_speed	(	const vec &	xwind,
		const vec &	ywind
	)

Vector version of wind speed calculation.

Parameters

xwind	1D vector of X-component of wind [any unit]
ywind	1D vector of Y-component of wind [any unit]

Returns

1D vector of wind speeds [any unit]

window()

vec2 gridpp::window	(	const vec2 &	array,
		int	length,
		gridpp::Statistic	statistic,
		bool	before = false,
		bool	keep_missing = false,
		bool	missing_edges = true
	)

Compute window statistics across time, independently for each case.

Parameters

array	2D vector of input values (Case, Time)
length	Window length in number of timesteps
statistic	Statistic to apply to window
before	If true, make the window end at the particular time. If false, centre it.
keep_missing	If true, window value will be missing if one or more values in window are missing
missing_edges	If true put missing values at the edges, where window overshoots the edge

Returns

2D vector of output values (Case, Time)

Variable Documentation

_debug_level

int gridpp::_debug_level = 0

static

gas_constant_mol

const float gridpp::gas_constant_mol = 8.31447

static

Universal Gas Constant [kg*m^2*s^-2/(K*mol)].

gas_constant_si

const float gridpp::gas_constant_si = 287.05

static

Universal Gas Constant [J/(kg*K)].

gravit

const float gridpp::gravit = 9.80665

static

Gravitational acceleration [m/s^2].

lapse_rate

const float gridpp::lapse_rate =0.0065

static

Constant Lapse Rate moist air standard atmosphere [K/m].

molar_mass

const float gridpp::molar_mass = 0.0289644

static

Molar Mass of Dry Air [kg/mol].

MV

const float gridpp::MV = NAN

static

Missing value indicator.

MV_CML

const float gridpp::MV_CML = -999

static

Missing value indicator in gridpp command-line tool.

pi

const float gridpp::pi = 3.14159265

static

Mathematical constant pi.

radius_earth

const double gridpp::radius_earth = 6.378137e6

static

Radius of the earth [m].

standard_surface_temperature

const float gridpp::standard_surface_temperature = 288.15

static

Temperature at surface in standard atmosphere [K].

swig_default_value

const float gridpp::swig_default_value = -1

static

Default value used to fill array in SWIG testing functions.

Not useful for any other purpose.