---
title: "adw"
output:
html_document:
highlight: textmate
keep_md: true
number_sections: no
theme: united
toc: yes
toc_float: yes
vignette: >
%\VignetteIndexEntry{adw: Angular Distance Weighting Interpolation}
%\VignetteEngine{knitr::rmarkdown}
%\VignetteEncoding{UTF-8}
---
```{r, include = FALSE, message=FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>"
)
```
## Angular Distance Weighting Interpolation
The irregularly-spaced data are interpolated onto regular latitude-longitude grids by weighting each station according to its distance and angle from the center of a search radius.
In addition to this, we also provide a simple way (Jones and Hulme, 1996) to grid the irregularly-spaced data points onto regular latitude-longitude grids by averaging all stations in grid-boxes.
## Reference
Caesar, J., L. Alexander, and R. Vose, 2006: Large-scale changes in observed daily maximum and minimum temperatures: Creation and analysis of a new gridded data set. Journal of Geophysical Research, 111, .
Jones, P. D., and M. Hulme, 1996: Calculating regional climatic time series for temperature and precipitation: Methods and illustrations. Int. J. Climatol., 16, 361–377, 3.0.CO;2-F>.
## Installation
Install the latest CRAN release via command:
```{r,eval=FALSE}
install.packages("adw")
```
## generate data and plot scatter
```{r, message=FALSE}
library(sf)
library(ggplot2)
library(adw)
library(cnmap)
set.seed(1)
tavg <- data.frame(lon = runif(100, min = 110, max = 117),
lat = runif(100, min = 31, max = 37),
value = runif(100, min = 20, max = 35))
hmap <- getMap(name = "河南省", returnClass = "sf")
ggplot() +
geom_point(data = tavg, aes(x = lon, y = lat, colour = value),
pch = 17, size = 2.5) +
geom_sf(data = st_cast(hmap, 'MULTILINESTRING')) +
scale_colour_fermenter(palette = "YlOrRd",
direction = 1,
breaks = seq(from = 25, to = 32, by = 1),
limits = c(0, 100),
name = expression("\u00B0C")) +
ggtitle("The irregularly-spaced data") +
theme_bw() +
theme(axis.title = element_blank(),
legend.key.width = unit(0.5,"cm"),
legend.key.height = unit(1.5, "cm"),
plot.title = element_text(hjust = 0.5, size = 11))
```
## Interpolation
### Usage 1
The parameter *extent* in the **adw** function is a *sf* class (sf package), and the coordinate reference system of the object is WGS1984 (EPSG: 4326).
```{r}
library(adw)
hmap_sf <- getMap(name = "河南省", returnClass = "sf") |> st_make_valid()
dg <- adw(tavg, extent = hmap_sf, gridsize = 0.1, cdd = 400)
head(dg)
ggplot() +
geom_tile(data = dg, aes(x = lon, y = lat, fill = value)) +
geom_sf(data = st_cast(hmap_sf, 'MULTILINESTRING')) +
scale_fill_fermenter(palette = "YlOrRd",
direction = 1,
breaks = seq(from = 25, to = 32, by = 1),
limits = c(0, 100),
name = expression("\u00B0C"),
na.value = "white") +
ggtitle("Angular distance weighting interpolation") +
theme_bw() +
theme(axis.title = element_blank(),
legend.key.width = unit(0.5,"cm"),
legend.key.height = unit(1.5, "cm"),
plot.title = element_text(hjust = 0.5, size = 11))
```
### Usage 2
The parameter *extent* in the **adw** function is a *SpatVector* class (terra packag), and the coordinate reference system of the object is WGS1984 (EPSG: 4326).
```{r}
library(adw)
library(terra)
hmap_sv <- getMap(name = "河南省", returnClass = "sv")
dg <- adw(tavg, extent = hmap_sv, gridsize = 0.1, cdd = 400)
head(dg)
ggplot() +
geom_tile(data = dg, aes(x = lon, y = lat, fill = value)) +
geom_sf(data = st_cast(hmap_sf, 'MULTILINESTRING')) +
scale_fill_fermenter(palette = "YlOrRd",
direction = 1,
breaks = seq(from = 25, to = 32, by = 1),
limits = c(0, 100),
name = expression("\u00B0C"),
na.value = "white") +
ggtitle("Angular distance weighting interpolation") +
theme_bw() +
theme(axis.title = element_blank(),
legend.key.width = unit(0.5,"cm"),
legend.key.height = unit(1.5, "cm"),
plot.title = element_text(hjust = 0.5, size = 11))
```
### Usage 3
The parameter *extent* in the **adw** function is a extent *vector* of length 4 in the order [xmin, xmax, ymin, ymax]
```{r}
library(adw)
interpExtent <- c(110.36, 116.65, 31.38, 36.37) # [xmin, xmax, ymin, ymax]
dg <- adw(tavg, extent = interpExtent, gridsize = 0.1, cdd = 400)
head(dg)
ggplot() +
geom_tile(data = dg, aes(x = lon, y = lat, fill = value)) +
geom_sf(data = st_cast(hmap_sf, 'MULTILINESTRING')) +
scale_fill_fermenter(palette = "YlOrRd",
direction = 1,
breaks = seq(from = 25, to = 32, by = 1),
limits = c(0, 100),
name = expression("\u00B0C"),
na.value = "white") +
ggtitle("Angular distance weighting interpolation") +
theme_bw() +
theme(axis.title = element_blank(),
legend.key.width = unit(0.5,"cm"),
legend.key.height = unit(1.5, "cm"),
plot.title = element_text(hjust = 0.5, size = 11))
```
### Usage 4
The irregularly-spaced data of points are converted onto regular latitude-longitude grids by averaging all stations in grid-boxes. The parameter *extent* in the **point2grid** function is a extent *vector* of length 4 in the order [xmin, xmax, ymin, ymax], or a simple fearture object, or a SpatVect object.
```{r}
library(adw)
interpExtent <- c(110.36, 116.65, 31.38, 36.37) # [xmin, xmax, ymin, ymax]
dg <- points2grid(tavg, extent = interpExtent, gridsize = 0.5)
head(dg)
ggplot() +
geom_tile(data = dg, aes(x = lon, y = lat, fill = value)) +
geom_sf(data = st_cast(hmap_sf, 'MULTILINESTRING')) +
scale_fill_fermenter(palette = "YlOrRd",
direction = 1,
breaks = seq(from = 25, to = 32, by = 1),
limits = c(0, 100),
name = expression("\u00B0C"),
na.value = "white") +
ggtitle("Averaging all stations in grid-boxes") +
theme_bw() +
theme(axis.title = element_blank(),
legend.key.width = unit(0.5,"cm"),
legend.key.height = unit(1.5, "cm"),
plot.title = element_text(hjust = 0.5, size = 11))
```
### Area weight average
The large area, or hemispheric, or global averages can be calculated dependent on the area represented by the grid-point or grid-box. The weight of latitude-longitude grid-points-boxes should be the cosine of the latitude of the ith grid-point-box.
```{r}
dg <- na.omit(dg)
awa(dg$value, dg$lat)
```