Introduction to the localsp package
1.1 This package can be used to address following issues:
Identify local power of individual determinants.
Explore how spatial stratified association changes spatially and in local regions.
1.2 How to install this package:
- Install the stable version, please use:
- Install the development version, please use:
install.packages("localsp",
repos = c("https://ausgis.r-universe.dev",
"https://cloud.r-project.org"),
dep = TRUE)1.3 General calculation process of LISP model:
Figure 1 shows the process of LISP, comprising three sequential steps. First, the optimal extent for local is identified using a spatial variogram to satisfy the criterion of a sufficiently small range and ensure that the data within the extent exhibit adequate heterogeneity and association. The second step is to analyze the PD of individual variables on the response variable through GD modeling with optimal spatial discretization algorithms and parameters at the local scale. The final step is to quantify the local PD of interaction variables, including the interaction of a pair of spatial variables and the interaction of multiple variables, using a tree-based spatial discretization approach and stratified heterogeneity approaches.
1.4 Example data in this package:
The gtc.csv dataset documents glacier thickness changes
(GTC) from 2000 to 2020, along with potential influencing factors in the
Greater Himalayas, encompassing the Hengduan Mountains, Himalayas,
Nyainqentanglha Mountains, Karakoram, and Hindu Kush (see Figure2).
Variables such as WinTem, SumTem, and Pre
represent the linear trends of winter temperature, summer temperature,
and precipitation over the 2000–2020 period. Elev,
Aspect, and Slope are derived from NASADEM data and
correspond to elevation, aspect, and slope, respectively.
LakeArea indicates the area of glacial lakes, while
SurAlbedo, calculated from MODIS products, represents the
surface albedo of glaciers.
gtc = readr::read_csv(system.file("extdata/gtc.csv", package = "localsp"))
## Rows: 908 Columns: 11
## ── Column specification ────────────────────────────────────────────────────────────────────────────────
## Delimiter: ","
## dbl (11): X, Y, GTC, Slope, Elev, LakeArea, WinTem, SumTem, Pre, Aspect, SurAlbedo
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
gtc = sf::st_as_sf(gtc, coords = c("X","Y"), crs = 4326)
gtc
## Simple feature collection with 908 features and 9 fields
## Geometry type: POINT
## Dimension: XY
## Bounding box: xmin: 69.435 ymin: 27.4918 xmax: 102.75 ymax: 37.1104
## Geodetic CRS: WGS 84
## # A tibble: 908 × 10
## GTC Slope Elev LakeArea WinTem SumTem Pre Aspect SurAlbedo geometry
## * <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <POINT [°]>
## 1 -0.766 0.0369 139. 4750. 0.00245 0.318 14.3 0.000233 0.0220 (75.7125 34.5219)
## 2 -0.948 0.124 137. 4559. 0.00131 0.281 12.6 0.000379 0.0123 (76.0715 34.343)
## 3 -2.05 0.411 154. 4428. 0.000229 -0.192 7.70 0.000429 0.0121 (90.2298 28.1097)
## 4 -0.525 0.0101 296. 4272. 0.00215 0.234 17.1 0.00130 0.0332 (72.7349 35.7435)
## 5 -0.950 0.0597 229. 5845. 0.000472 0.0204 28.7 0.00174 0.0262 (88.6606 27.9974)
## 6 -1.88 0.991 137. 5371. 0.000472 -0.0492 16.3 0.00174 0.0262 (88.687 27.9951)
## 7 -2.55 1.28 126. 5477. 0.000472 -0.294 14.5 0.00174 0.0262 (88.7123 27.9911)
## 8 -0.909 0.0444 217. 4691. 0.00155 0.326 17.3 0.00175 0.0359 (75.3851 35.3621)
## 9 -3.94 0.406 161. 3912. -0.000868 0.0490 7.88 0.00177 0.00379 (96.5057 29.4569)
## 10 -0.819 0.0122 153. 4066. -0.000529 -0.0356 6.40 0.00178 0.0415 (71.9206 36.3066)
## # ℹ 898 more rows