Glaciares en Chile
Optimización del Cálculo de áreas.
VE-CC
DataIntelligence27-05-2021
Abstract
El primer tema que nos ha ocupado ha sido el establecer un cronograma que se entrega en las primeras lÃneas, se expone un ejemplo completo de cómo debe ser ejecutado el RF, el CART y las SVM adjuntando una matriz de confusión en cada una. LA SVM revienta la MdeC aunque las dos restantes tienen un nivel de precisión bárbaro.
Por otro lado, nos hemos encontrado con múltiples errores en las instalaciones de R, Rgee y Rstudio en diferentes computadores, siendo que en los propios han funcionado ok. Instalar aplicaciones complejas es delicado y se sospecha la corrupción de muchos elementos en las computadoras donde ni se instalan ni corren bien. Realizaremos el siguiente ejercicio: utilizaremos una de las computadoras de las propias, donde todo funcione ok. Borraremos todo R, RStudio y Python, volviendo a quedar todo ok luego de ser reinstalado. Suponemos que el problema es el software windows. ¿Será solución reparar las máquinas volviéndolas a sus estados de fábrica?
La última parte la aborda Victor y nos ofrece un metodo eficiente para ir obteniendo las áreas glaciares estáticas, para llegar rápido a las dinámicas.
library(rgee)
ee_Initialize("tarredwall@gmail.com", drive = TRUE)## -- rgee 1.0.9 --------------------------------------- earthengine-api 0.1.259 --
## v email: tarredwall@gmail.com
## v Google Drive credentials:
v Google Drive credentials: FOUND
## v Initializing Google Earth Engine:
v Initializing Google Earth Engine: DONE!
##
v Earth Engine user: users/tarredwall
## --------------------------------------------------------------------------------
Drag Racing
Drag Racing
R
A partir del 15 de febrero de 2018, se debe utilizar la siguiente denominación:
Yo
I Región de Arica y Parinacota. Lunes 24 II Región de Tarapacá. Martes 25 III Región de Antofagasta. Miercoles 26 IV Región de Atacama. Jueves 27 V Región de Coquimbo. Viernes 28 VI Región de ValparaÃso. Lunes 31
Victor
VII Región Metropolitana de Santiago. Lunes 24 VIII Región del Libertador General Bernardo O’Higgins. Martes 25 IX Región del Maule. Miercoles 26 X Región del Ñuble. Jueves 27 XI Región del BiobÃo. Viernes 28 XII Región de La AraucanÃa. Lunes 31
Abner
XIII Región de Los RÃos. Miercoles 26 XIV Región de Los Lagos. Jueves 27 XV Región de Aysén del General Carlos Ibáñez del Campo. Viernes 28 XVI Región de Magallanes y la Antártica Chilena. Lunes 31
1. Cargamos el shp de la primera region y lo filterboundiamos con la imagen satelital sentinel. Cargamos el shp de glaciares, y obtenemos las muestras interiores de los limites administrativos de color rojo que significaran glaciares que tendren numero 0. Obtenemos muestras para los colores verdes fuera de los limites administrativos. Unimos ambas muestras y categorizamos nuevamente la imagen satelital sentinel.
I Región de Arica y Parinacota.
1 Lo primero es dibujar el shp de la Región de Arica y parinacota
region_arica <- st_read("Regiones_separadas/Region_15.shp",
quiet = TRUE) %>%
sf_as_ee()## Registered S3 method overwritten by 'geojsonsf':
## method from
## print.geojson geojsonarica <- region_arica$geometry()
Map$setCenter(lon = -69.7522, lat = -18.65665, zoom = 7)
Map$addLayer(arica)2 lo segundo es asociar una imagen satelital Sentinel filtrafda por fecha, nubosidad y vinculada a la region de estudio. Y establecemos que el parametro sea la media.
start <- ee$Date("2019-01-01")
finish <- ee$Date("2019-04-01")
cc <- 20
sentinel1 = ee$ImageCollection('COPERNICUS/S2')$filterDate(start, finish)$filterBounds(arica)$
filter(ee$Filter$lt("CLOUDY_PIXEL_PERCENTAGE", cc))
first <- sentinel1$median()
vizParams <- list(
bands = c("B8","B5" , "B3"),
# bands = c("B2", "B3"),
min = 100,
max = 1000,
gamma = 2
)
Map$setCenter(lon = -69.7522, lat = -18.65665, zoom = 7)
Map$addLayer(first, vizParams, "Landsat 8 image")3 Lo tercero es construir un raster con NDGI mayor a 0 siempre vinculada a la primera imagen satelital que hemos extraÃdo.
Construimos una imagen raster con una columna llamada nd con dos categorÃas para cada pixel: 0 si es rojo, 1 si es verde
getNDGI <- function(image)
{
image$normalizedDifference(c("B4", "B3")) > 0
}
ndgi <- getNDGI(first)## Warning: Ops.ee.image.Image will be deprecated in rgee v.1.1.0. Please install
## rgeeExtra (https://github.com/r-earthengine/rgeeExtra). Deeply sorry for the
## inconveniences.ndgiParams <- list(palette = c(
"#d73027", "#f46d43", "#fdae61",
"#fee08b", "#d9ef8b", "#a6d96a",
"#66bd63", "#1a9850"
))
Map$setCenter(lon = -69.7522, lat = -18.65665, zoom = 7)
Map$addLayer(ndgi, ndgiParams, "NDGI")4 obtenemos los shp que nos indican la presencia de glaciares en la Region de Arica y Parinacota.
glaciar_arica <- st_read("GlaciaresxRegion/Simple_ARICA_Y_PARINACOTA.shp",
quiet = TRUE) %>%
sf_as_ee()
glaciar_arica <- glaciar_arica$geometry()
Map$setCenter(lon = -69.7522, lat = -18.65665, zoom = 7)
Map$addLayer(glaciar_arica, ndgiParams, "NDGI")5 Intersectamos las regiones glaciares de la region de Arica con nuestro ndvi, y de esta forma obtenemos un universo muestral al cual poder extraer nuestro primer set de muestras (pixeles rojos interiores).
ESTA ES LA IMAGEN SOBRE LA QUE EXTRAEREMOS EL PRIMER SET DE MUESTRAS
glaciares_adm_con_ngvi <- ndgi$clip(glaciar_arica)
Map$setCenter(lon = -69.80584, lat = -17.67837, zoom = 15)
Map$addLayer(glaciares_adm_con_ngvi, ndgiParams, "NDGI")digresion
digresion
6 Hay que hacer una diferencia (se le resta al shp de la región de Arica los shps de los glaciares) para intersectarla con nuestra imagen ngdi y obtener el segundo universo muestral.
6.1 Primero obtenemos nuestros shapes sin zonas administrativas glaciares:
# region_arica <- st_read("Regiones_separadas/Region_15.shp",
# quiet = TRUE) %>%
# sf_as_ee()
# arica <- region_arica$geometry()# glaciar_arica <- st_read("GlaciaresxRegion/Simple_ARICA_Y_PARINACOTA.shp",
# quiet = TRUE) %>%
# sf_as_ee()
# glaciar_arica <- glaciar_arica$geometry()arica_sin_glacia <- arica$difference(glaciar_arica, ee$ErrorMargin(1))
Map$setCenter(lon = -69.80584, lat = -17.67837, zoom = 7)
Map$addLayer(arica_sin_glacia, ndgiParams, "NDGI")6.2 Necesitamos intersectar nuestro ndvi con la zona de Arica y Parinacota que excluye los glaciares.
ndvi_arica_sin_glaciares <- ndgi$clip(arica_sin_glacia)
Map$setCenter(lon = -69.80584, lat = -17.67837, zoom = 7)
Map$addLayer(ndvi_arica_sin_glaciares, ndgiParams, "NDGI")7 Ahora tenemos dos regiones preparadas sobre las cuales podemos extraer nuestras muestras. A efectos de orden nombrémoslas:
7.1. pixeles rojos interiores
7.2. pixeles verdes exteriores
7.1. pixeles rojos interiores
Necesitamos intersectar las regiones administrativas glaciares con nuestro raster ndvi y de ahi extraer las muestras de color rojo que simbolizaran los glaciares.
limites_glaciares_con_rasters <- ndgi$clip(glaciar_arica)
Map$setCenter(lon = -69.80584, lat = -17.67837, zoom = 15)
Map$addLayer(limites_glaciares_con_rasters , ndgiParams, "NDGI")1 Muestreo
# region_arica <- st_read("Regiones_separadas/Region_15.shp", quiet = TRUE)
# region_aricaLos rojos interiores son glaciares, 0 = rojo y son llamados muestras_in_gee_rojas.
muestras_in <- ndgi$sampleRegions(
collection = ee$Feature(glaciar_arica),
scale = 100,
tileScale = 16,
geometries = TRUE
)
muestras_in_gee_rojas = muestras_in$filter(ee$Filter$eq('nd', 0))
Map$setCenter(lon = -69.80584, lat = -17.67837, zoom = 7)
Map$addLayer(
eeObject = muestras_in_gee_rojas,
visParams = {},
name = "puntos glaciares"
)7.2. Pixeles verdes exteriores
ndvi_arica_sin_glaciares <- ndgi$clip(arica_sin_glacia)
Map$setCenter(lon = -69.80584, lat = -17.67837, zoom = 7)
Map$addLayer(ndvi_arica_sin_glaciares, ndgiParams, "NDGI")Los verdes exteriores no son glaciares 1 = verde
muestras_in <- ndgi$sampleRegions(
collection = ee$Feature(ndvi_arica_sin_glaciares),
scale = 1000,
tileScale = 16,
geometries = TRUE
)
muestras_out_gee_verdes = muestras_in$filter(ee$Filter$eq('nd', 1))
Map$setCenter(lon = -69.80584, lat = -17.67837, zoom = 7)
Map$addLayer(
eeObject = muestras_out_gee_verdes,
visParams = {},
name = "puntos no glaciares"
)III debemos unir muestras y generar el random forest
union_total <- muestras_out_gee_verdes$merge(muestras_in_gee_rojas)3 Aplicamos Random Forest
region_arica <- st_read("Regiones_separadas/Region_15.shp",
quiet = TRUE) %>%
sf_as_ee()
arica <- region_arica$geometry()
start <- ee$Date("2019-01-01")
finish <- ee$Date("2019-04-01")
cc <- 20
sentinel1 = ee$ImageCollection('COPERNICUS/S2')$filterDate(start, finish)$filterBounds(arica)$
filter(ee$Filter$lt("CLOUDY_PIXEL_PERCENTAGE", cc))
first <- sentinel1$median()
# Use these bands for prediction.
bands <- c("B2", "B3", "B4", "B8")
# This property of the table stores the land cover labels.
label <- "nd"
# Overlay the points on the imagery to get training.
training <- first$select(bands)$sampleRegions(
collection = union_total,
properties = list(label),
scale = 10
)
# Train a CART classifier with default parameters.
trained <- ee$Classifier$smileRandomForest(10)$train(training, label, bands)
# Classify the image with the same bands used for training.
classified1 <- first$select(bands)$classify(trained)
# Viz parameters.
# viz_img <- list(bands = c("B4", "B3", "B2"), max = 0.4)
vizParams <- list(
bands = c("B8","B5" , "B3"),
# bands = c("B2", "B3"),
min = 100,
max = 1000,
gamma = 2
)
viz_class <- list(palette = c("red", "green", "blue"), min = 0, max = 2)
Map$addLayer(first, vizParams, name = "image") +
Map$addLayer(classified1, viz_class, name = "classification")18:29 18:30
Le aplicamos la matriz de confusion:
trainAccuracy = trained$confusionMatrix()
trainAccuracy$getInfo()## [[1]]
## [1] 22 3
##
## [[2]]
## [1] 1 1775418:40 18:43
4 Aplicamos CART
region_arica <- st_read("Regiones_separadas/Region_15.shp",
quiet = TRUE) %>%
sf_as_ee()
arica <- region_arica$geometry()
start <- ee$Date("2019-01-01")
finish <- ee$Date("2019-04-01")
cc <- 20
sentinel1 = ee$ImageCollection('COPERNICUS/S2')$filterDate(start, finish)$filterBounds(arica)$
filter(ee$Filter$lt("CLOUDY_PIXEL_PERCENTAGE", cc))
first <- sentinel1$median()
# Use these bands for prediction.
bands <- c("B2", "B3", "B4", "B8")
# This property of the table stores the land cover labels.
label <- "nd"
# Overlay the points on the imagery to get training.
training <- first$select(bands)$sampleRegions(
collection = union_total,
properties = list(label),
scale = 10
)
# Train a CART classifier with default parameters.
trained <- ee$Classifier$smileCart(10)$train(training, label, bands)
# Classify the image with the same bands used for training.
classified <- first$select(bands)$classify(trained)
# Viz parameters.
# viz_img <- list(bands = c("B4", "B3", "B2"), max = 0.4)
vizParams <- list(
bands = c("B8","B5" , "B3"),
# bands = c("B2", "B3"),
min = 100,
max = 1000,
gamma = 2
)
viz_class <- list(palette = c("red", "green", "blue"), min = 0, max = 2)
Map$addLayer(first, vizParams, name = "image") +
Map$addLayer(classified, viz_class, name = "classification")18:47 18:50
trainAccuracy = trained$confusionMatrix()
trainAccuracy$getInfo()## [[1]]
## [1] 16 9
##
## [[2]]
## [1] 2 1775318:51 18 53
4 Aplicamos SVM
region_arica <- st_read("Regiones_separadas/Region_15.shp",
quiet = TRUE) %>%
sf_as_ee()
arica <- region_arica$geometry()
start <- ee$Date("2019-01-01")
finish <- ee$Date("2019-04-01")
cc <- 20
sentinel1 = ee$ImageCollection('COPERNICUS/S2')$filterDate(start, finish)$filterBounds(arica)$
filter(ee$Filter$lt("CLOUDY_PIXEL_PERCENTAGE", cc))
first <- sentinel1$median()
# Use these bands for prediction.
bands <- c("B2", "B3", "B4", "B8")
# This property of the table stores the land cover labels.
label <- "nd"
# Overlay the points on the imagery to get training.
training <- first$select(bands)$sampleRegions(
collection = union_total,
properties = list(label),
scale = 10
)
classifier = ee$Classifier$libsvm(
kernelType = "RBF",
gamma = 0.5,
cost = 10
)
# Train a CART classifier with default parameters.
trained = classifier$train(training, label, bands)
# Classify the image with the same bands used for training.
classified <- first$select(bands)$classify(trained)
# Viz parameters.
# viz_img <- list(bands = c("B4", "B3", "B2"), max = 0.4)
vizParams <- list(
bands = c("B8","B5" , "B3"),
# bands = c("B2", "B3"),
min = 100,
max = 1000,
gamma = 2
)
viz_class <- list(palette = c("red", "green", "blue"), min = 0, max = 2)
Map$addLayer(first, vizParams, name = "image") +
Map$addLayer(classified, viz_class, name = "classification")No se puede obtener la matriz de confusion para libsvm pues excede limites de memoria
# trainAccuracy = trained$confusionMatrix()
# trainAccuracy$getInfo()2. Una vez obtenido nuestro raster random forest interceptamos por limites glaciares, Y EN CADA UNO DE ELLOS determinamos la cantidad de pixeles rojos
mask_0 <- read_sf("GlaciaresxRegion/Simple_ARICA_Y_PARINACOTA.shp")reg_15_2 <- mask_0 %>% dplyr::select(COD_GLA)
varia <- reg_15_2 %>%
sf_as_ee()
region_0 <- varia$geometry()
sale_4 <- classified1$clip(region_0)
puntos_r <- sale_4$select('classification')$eq(0)
areas_rojas = puntos_r$reduceRegions(
reducer= ee$Reducer$sum(),
collection= varia,
scale= 30
)
rojas <- areas_rojas$getInfo()
rojas1 <- as.data.frame(rojas)puntos_v <- sale_4$select('classification')$eq(1)
areas_verdes = puntos_v$reduceRegions(
reducer= ee$Reducer$sum(),
collection= varia,
scale= 30
)
verdes <- areas_verdes$getInfo()
verdes1 <- as.data.frame(verdes)rojas1$features.properties.COD_GLA.50## [1] "CL101021034" "CL101021034"rojas1$features.properties.sum.50## [1] 0 0r_v <- rbind(colnames(rojas1),rojas1,verdes1)
r_v <- r_v[1:4,]
r_v <- r_v[-3,]
r_v <- t(r_v)
r_v <- as.data.frame(r_v)
names(r_v)[1] <- "col1"
names(r_v)[2] <- "col2"
names(r_v)[3] <- "col3"
r_v$col4 <- r_v$col3data1 <- filter(r_v, grepl("features.properties.COD_GLA",col1))
data1 <- as.data.frame(data1[,2])
data2 <- filter(r_v, grepl("features.properties.sum",col1))
data2 <- as.data.frame(data2[,c(2,3)])
data3 <- cbind(data1,data2)
kbl(data3) %>%
kable_styling(bootstrap_options = c("striped", "hover")) %>%
kable_paper() %>%
scroll_box(width = "100%", height = "300px")| data1[, 2] | col2 | col3 | |
|---|---|---|---|
| features.properties.sum | CL101001010 | 14.7098039215686 | 88.3176470588235 |
| features.properties.sum.1 | CL101010001 | 68.6235294117647 | 1146.45882352941 |
| features.properties.sum.2 | CL101010006 | 8.89803921568627 | 174.352941176471 |
| features.properties.sum.3 | CL101010009 | 4.64705882352941 | 213.039215686275 |
| features.properties.sum.4 | CL101010012 | 117.701960784314 | 162.419607843137 |
| features.properties.sum.5 | CL101010016 | 3.63137254901961 | 21.0156862745098 |
| features.properties.sum.6 | CL101010002 | 8.65098039215686 | 16.6823529411765 |
| features.properties.sum.7 | CL101001031 | 0.972549019607843 | 351.627450980392 |
| features.properties.sum.8 | CL101001030 | 2.77254901960784 | 201.36862745098 |
| features.properties.sum.9 | CL101001011 | 5.39607843137255 | 60.8 |
| features.properties.sum.10 | CL101020003 | 1 | 50.2078431372549 |
| features.properties.sum.11 | CL101001032 | 53.1450980392157 | 179.686274509804 |
| features.properties.sum.12 | CL101030019 | 8.02745098039216 | 118.164705882353 |
| features.properties.sum.13 | CL101010018 | 0.156862745098039 | 13.756862745098 |
| features.properties.sum.14 | CL101010017 | 14.7882352941176 | 105.01568627451 |
| features.properties.sum.15 | CL101010011 | 46.8862745098039 | 105.537254901961 |
| features.properties.sum.16 | CL101001029 | 21.5843137254902 | 90.4470588235294 |
| features.properties.sum.17 | CL101030020 | 0 | 16.6823529411765 |
| features.properties.sum.18 | CL101000019 | 0 | 47.5098039215686 |
| features.properties.sum.19 | CL101501019 | 0 | 8.86274509803921 |
| features.properties.sum.20 | CL101501018 | 0 | 13.8470588235294 |
| features.properties.sum.21 | CL101501017 | 0 | 406.901960784314 |
| features.properties.sum.22 | CL101501016 | 0 | 302.623529411765 |
| features.properties.sum.23 | CL101501014 | 0 | 121.58431372549 |
| features.properties.sum.24 | CL101501015 | 0 | 23.2039215686274 |
| features.properties.sum.25 | CL101501012 | 0 | 80.6274509803922 |
| features.properties.sum.26 | CL101501013 | 0 | 13.9058823529412 |
| features.properties.sum.27 | CL101501011 | 0 | 225.81568627451 |
| features.properties.sum.28 | CL101501010 | 0 | 24.1098039215686 |
| features.properties.sum.29 | CL101501009 | 0 | 300.964705882353 |
| features.properties.sum.30 | CL101501008 | 0 | 91.6941176470588 |
| features.properties.sum.31 | CL101501007 | 0 | 10.0588235294118 |
| features.properties.sum.32 | CL101501006 | 0 | 10.6392156862745 |
| features.properties.sum.33 | CL101501005 | 0 | 112.294117647059 |
| features.properties.sum.34 | CL101501004 | 0 | 262.254901960784 |
| features.properties.sum.35 | CL101501003 | 0 | 16.6078431372549 |
| features.properties.sum.36 | CL101501002 | 0 | 147.466666666667 |
| features.properties.sum.37 | CL101501001 | 0 | 505.835294117647 |
| features.properties.sum.38 | CL101500007 | 0 | 41.5607843137255 |
| features.properties.sum.39 | CL101500006 | 0 | 580.36862745098 |
| features.properties.sum.40 | CL101500005 | 0 | 138.066666666667 |
| features.properties.sum.41 | CL101500004 | 0 | 135.772549019608 |
| features.properties.sum.42 | CL101500003 | 0 | 30.5294117647059 |
| features.properties.sum.43 | CL101500002 | 0 | 22.9647058823529 |
| features.properties.sum.44 | CL101500001 | 0 | 24.2666666666667 |
| features.properties.sum.45 | CL101410005 | 0 | 47.2235294117647 |
| features.properties.sum.46 | CL101410004 | 0 | 23.7176470588235 |
| features.properties.sum.47 | CL101410003 | 0 | 77.7098039215687 |
| features.properties.sum.48 | CL101030026 | 0 | 167.282352941176 |
| features.properties.sum.49 | CL101021035 | 0 | 11.4117647058824 |
| features.properties.sum.50 | CL101021034 | 0 | 7.15686274509804 |
| features.properties.sum.51 | CL101021036 | 0 | 11.0078431372549 |
| features.properties.sum.52 | CL101410002 | 0 | 30.3254901960784 |
| features.properties.sum.53 | CL101410001 | 0 | 220.6 |
| features.properties.sum.54 | CL101300047 | 0 | 24.7529411764706 |
| features.properties.sum.55 | CL101300046 | 0 | 31.4352941176471 |
| features.properties.sum.56 | CL101300045 | 0 | 33.5843137254902 |
| features.properties.sum.57 | CL101300039 | 0 | 19.7960784313725 |
| features.properties.sum.58 | CL101300044 | 0 | 5.83529411764706 |
| features.properties.sum.59 | CL101300040 | 0 | 26.5372549019608 |
| features.properties.sum.60 | CL101300043 | 0 | 19.0274509803922 |
| features.properties.sum.61 | CL101300042 | 0 | 88.9647058823529 |
| features.properties.sum.62 | CL101300041 | 0 | 10.2745098039216 |
| features.properties.sum.63 | CL101300038 | 0 | 146.788235294118 |
| features.properties.sum.64 | CL101300037 | 0 | 12.6823529411765 |
| features.properties.sum.65 | CL101300036 | 0 | 153.941176470588 |
| features.properties.sum.66 | CL101300035 | 0 | 9.86274509803921 |
| features.properties.sum.67 | CL101300034 | 0 | 38.643137254902 |
| features.properties.sum.68 | CL101300032 | 0 | 35.6235294117647 |
| features.properties.sum.69 | CL101300031 | 0 | 5.71764705882353 |
| features.properties.sum.70 | CL101300030 | 0 | 26.043137254902 |
| features.properties.sum.71 | CL101300029 | 0 | 19.5529411764706 |
| features.properties.sum.72 | CL101300028 | 0 | 49.7411764705882 |
| features.properties.sum.73 | CL101300026 | 0 | 82.5137254901961 |
| features.properties.sum.74 | CL101300027 | 0 | 8.84705882352941 |
| features.properties.sum.75 | CL101300025 | 0 | 142.960784313726 |
| features.properties.sum.76 | CL101300024 | 0 | 80.6705882352941 |
| features.properties.sum.77 | CL101300023 | 0 | 77.5882352941176 |
| features.properties.sum.78 | CL101300022 | 0 | 26.3058823529412 |
| features.properties.sum.79 | CL101300021 | 0 | 178.309803921569 |
| features.properties.sum.80 | CL101300020 | 0 | 6.17647058823529 |
| features.properties.sum.81 | CL101300019 | 0 | 9.00392156862745 |
| features.properties.sum.82 | CL101300018 | 0 | 81.1647058823529 |
| features.properties.sum.83 | CL101300017 | 0 | 21.9686274509804 |
| features.properties.sum.84 | CL101300016 | 0 | 39.1333333333333 |
| features.properties.sum.85 | CL101300015 | 0 | 7.69803921568628 |
| features.properties.sum.86 | CL101020006 | 0 | 40.0313725490196 |
| features.properties.sum.87 | CL101020007 | 0 | 32.0549019607843 |
| features.properties.sum.88 | CL101300014 | 0 | 103.380392156863 |
| features.properties.sum.89 | CL101300013 | 0 | 182.917647058824 |
| features.properties.sum.90 | CL101300011 | 0 | 53.3686274509804 |
| features.properties.sum.91 | CL101300012 | 0 | 27.2392156862745 |
| features.properties.sum.92 | CL101300010 | 0 | 26.9098039215686 |
| features.properties.sum.93 | CL101300009 | 0 | 49.1882352941176 |
| features.properties.sum.94 | CL101300008 | 0 | 47.5764705882353 |
| features.properties.sum.95 | CL101300007 | 0 | 40.564705882353 |
| features.properties.sum.96 | CL101300006 | 0 | 11.6549019607843 |
| features.properties.sum.97 | CL101300005 | 0 | 36.1137254901961 |
| features.properties.sum.98 | CL101300004 | 0 | 15.4078431372549 |
| features.properties.sum.99 | CL101300003 | 0 | 39.9764705882353 |
| features.properties.sum.100 | CL101300002 | 0 | 6.03921568627451 |
| features.properties.sum.101 | CL101300001 | 0 | 41.2352941176471 |
| features.properties.sum.102 | CL101202037 | 0 | 16.7137254901961 |
| features.properties.sum.103 | CL101202007 | 0 | 96.4196078431372 |
| features.properties.sum.104 | CL101202016 | 0 | 78.721568627451 |
| features.properties.sum.105 | CL101202015 | 0 | 13.3803921568627 |
| features.properties.sum.106 | CL101202014 | 0 | 10.1058823529412 |
| features.properties.sum.107 | CL101201024 | 0 | 25.5960784313726 |
| features.properties.sum.108 | CL101202003 | 0 | 72.2117647058824 |
| features.properties.sum.109 | CL101202009 | 0 | 22.6745098039216 |
| features.properties.sum.110 | CL101202013 | 0 | 5.94117647058823 |
| features.properties.sum.111 | CL101202031 | 0.188235294117647 | 139.403921568627 |
| features.properties.sum.112 | CL101202030 | 0 | 14.2509803921569 |
| features.properties.sum.113 | CL101202034 | 0 | 119.58431372549 |
| features.properties.sum.114 | CL101202033 | 0 | 22.9882352941177 |
| features.properties.sum.115 | CL101202035 | 0 | 390.549019607843 |
| features.properties.sum.116 | CL101202036 | 0 | 101.462745098039 |
| features.properties.sum.117 | CL101020001 | 0 | 47.9960784313726 |
| features.properties.sum.118 | CL101020002 | 0 | 20.5921568627451 |
| features.properties.sum.119 | CL101001034 | 0 | 40.3137254901961 |
| features.properties.sum.120 | CL101001035 | 0 | 14.9647058823529 |
| features.properties.sum.121 | CL101202032 | 0 | 58.3098039215686 |
| features.properties.sum.122 | CL101201023 | 0 | 39.0509803921569 |
| features.properties.sum.123 | CL101202001 | 0 | 58.8509803921568 |
| features.properties.sum.124 | CL101202026 | 0 | 51.3843137254902 |
| features.properties.sum.125 | CL101202027 | 0 | 11.4156862745098 |
| features.properties.sum.126 | CL101202029 | 0 | 8.53333333333333 |
| features.properties.sum.127 | CL101202028 | 0 | 8.34509803921569 |
| features.properties.sum.128 | CL101202022 | 0 | 11.5803921568627 |
| features.properties.sum.129 | CL101200025 | 0 | 17.3098039215686 |
| features.properties.sum.130 | CL101200032 | 0 | 25.9960784313726 |
| features.properties.sum.131 | CL101200021 | 0 | 121.474509803922 |
| features.properties.sum.132 | CL101200017 | 0 | 31.6823529411765 |
| features.properties.sum.133 | CL101200012 | 0 | 25.9333333333333 |
| features.properties.sum.134 | CL101200015 | 0 | 104.909803921569 |
| features.properties.sum.135 | CL101200014 | 0 | 54.7803921568628 |
| features.properties.sum.136 | CL101200013 | 0 | 208.623529411765 |
| features.properties.sum.137 | CL101200008 | 0 | 16.1803921568627 |
| features.properties.sum.138 | CL101200007 | 0 | 117.780392156863 |
| features.properties.sum.139 | CL101200006 | 0 | 21.4117647058823 |
| features.properties.sum.140 | CL101200005 | 0 | 32.2039215686274 |
| features.properties.sum.141 | CL101200004 | 0 | 78.7490196078432 |
| features.properties.sum.142 | CL101201003 | 0 | 75.6666666666667 |
| features.properties.sum.143 | CL101201001 | 0 | 47.9960784313726 |
| features.properties.sum.144 | CL101201002 | 0 | 12.6980392156863 |
| features.properties.sum.145 | CL101200033 | 0 | 60.556862745098 |
| features.properties.sum.146 | CL101200034 | 0 | 72.8509803921569 |
| features.properties.sum.147 | CL101201011 | 0 | 38.4941176470588 |
| features.properties.sum.148 | CL101201009 | 0 | 77.4039215686274 |
| features.properties.sum.149 | CL101201012 | 0 | 110.317647058824 |
| features.properties.sum.150 | CL101000007 | 0 | 52.7803921568627 |
| features.properties.sum.151 | CL101000012 | 0 | 56.878431372549 |
| features.properties.sum.152 | CL101000008 | 0 | 6.55294117647059 |
| features.properties.sum.153 | CL101000003 | 0 | 49.2745098039215 |
| features.properties.sum.154 | CL101000004 | 0 | 15.4705882352941 |
| features.properties.sum.155 | CL101000005 | 0 | 6.65490196078431 |
| features.properties.sum.156 | CL101000002 | 0 | 36.3098039215686 |
| features.properties.sum.157 | CL101201016 | 0 | 34.4078431372549 |
| features.properties.sum.158 | CL101201015 | 0 | 25.6941176470588 |
| features.properties.sum.159 | CL101201018 | 0 | 57.443137254902 |
| features.properties.sum.160 | CL101201017 | 0 | 4.85490196078431 |
| features.properties.sum.161 | CL101201019 | 0 | 90.1333333333333 |
| features.properties.sum.162 | CL101201020 | 0 | 44.2588235294118 |
| features.properties.sum.163 | CL101201022 | 0 | 66.9764705882353 |
| features.properties.sum.164 | CL101201021 | 0 | 31.4705882352941 |
| features.properties.sum.165 | CL101201014 | 0 | 130.129411764706 |
| features.properties.sum.166 | CL101201013 | 0 | 316.247058823529 |
| features.properties.sum.167 | CL101000009 | 0 | 8.20392156862745 |
| features.properties.sum.168 | CL101000010 | 0 | 11.9803921568627 |
| features.properties.sum.169 | CL101000011 | 0 | 23.0627450980392 |
| features.properties.sum.170 | CL101000014 | 0 | 46.478431372549 |
| features.properties.sum.171 | CL101000013 | 0 | 38.9921568627451 |
| features.properties.sum.172 | CL101000015 | 0 | 28.6235294117647 |
| features.properties.sum.173 | CL101000020 | 0 | 747.180392156863 |
| features.properties.sum.174 | CL101000021 | 0 | 40.2627450980392 |
| features.properties.sum.175 | CL101000022 | 0 | 21.2 |
| features.properties.sum.176 | CL101201004 | 0 | 83.4823529411765 |
| features.properties.sum.177 | CL101201005 | 0.992156862745098 | 29.0352941176471 |
| features.properties.sum.178 | CL101000018 | 0 | 49.1803921568628 |
| features.properties.sum.179 | CL101201010 | 0 | 7.7921568627451 |
| features.properties.sum.180 | CL101000016 | 0 | 5.16470588235294 |
| features.properties.sum.181 | CL101000017 | 0 | 12.8509803921569 |
| features.properties.sum.182 | CL101200022 | 0 | 17.5647058823529 |
| features.properties.sum.183 | CL101200024 | 0 | 70.7764705882353 |
| features.properties.sum.184 | CL101200023 | 0 | 32.2392156862745 |
| features.properties.sum.185 | CL101200020 | 0 | 127.945098039216 |
| features.properties.sum.186 | CL101200016 | 0 | 73.0862745098039 |
| features.properties.sum.187 | CL101001005 | 0.72156862745098 | 47.2705882352941 |
| features.properties.sum.188 | CL101001003 | 0 | 128.670588235294 |
| features.properties.sum.189 | CL101001002 | 0 | 46.3294117647059 |
| features.properties.sum.190 | CL101001001 | 0 | 22.9882352941176 |
| features.properties.sum.191 | CL101001004 | 0 | 21.4705882352941 |
| features.properties.sum.192 | CL101001009 | 0 | 63.0901960784314 |
| features.properties.sum.193 | CL101001008 | 0 | 4.76078431372549 |
| features.properties.sum.194 | CL101001007 | 0 | 5.56862745098039 |
| features.properties.sum.195 | CL101001019 | 0 | 40.2 |
| features.properties.sum.196 | CL101001018 | 0 | 14.1647058823529 |
| features.properties.sum.197 | CL101001020 | 0 | 49.643137254902 |
| features.properties.sum.198 | CL101001022 | 0 | 42.9803921568628 |
| features.properties.sum.199 | CL101001021 | 0 | 129.254901960784 |
| features.properties.sum.200 | CL101010003 | 0 | 53.7529411764706 |
| features.properties.sum.201 | CL101010004 | 0 | 37.0627450980392 |
| features.properties.sum.202 | CL101010005 | 0 | 39.3921568627451 |
| features.properties.sum.203 | CL101010008 | 0 | 124.792156862745 |
| features.properties.sum.204 | CL101010015 | 0 | 169.090196078431 |
| features.properties.sum.205 | CL101010020 | 0 | 63.8235294117647 |
| features.properties.sum.206 | CL101010021 | 0 | 128.164705882353 |
| features.properties.sum.207 | CL101021013 | 0 | 265.352941176471 |
| features.properties.sum.208 | CL101021014 | 0 | 212.972549019608 |
| features.properties.sum.209 | CL101021011 | 0 | 145.439215686274 |
| features.properties.sum.210 | CL101021020 | 0 | 767.474509803922 |
| features.properties.sum.211 | CL101021012 | 0 | 36.0274509803922 |
| features.properties.sum.212 | CL101021005 | 0 | 96.9686274509804 |
| features.properties.sum.213 | CL101021021 | 0 | 169.454901960784 |
| features.properties.sum.214 | CL101021022 | 0 | 74.5607843137255 |
| features.properties.sum.215 | CL101021023 | 0 | 220.745098039216 |
| features.properties.sum.216 | CL101021024 | 0 | 16.3019607843137 |
| features.properties.sum.217 | CL101021025 | 0 | 11.8901960784314 |
| features.properties.sum.218 | CL101021026 | 0 | 11.5529411764706 |
| features.properties.sum.219 | CL101030012 | 0 | 98.6039215686274 |
| features.properties.sum.220 | CL101030014 | 0 | 109.333333333333 |
| features.properties.sum.221 | CL101030017 | 0 | 8.92549019607843 |
| features.properties.sum.222 | CL101021030 | 0 | 70.7843137254902 |
| features.properties.sum.223 | CL101021031 | 0 | 164.796078431373 |
| features.properties.sum.224 | CL101021033 | 0 | 10.9686274509804 |
| features.properties.sum.225 | CL101021032 | 0 | 310.21568627451 |
| features.properties.sum.226 | CL101030011 | 0 | 104.909803921569 |
| features.properties.sum.227 | CL101030009 | 0 | 135.6 |
| features.properties.sum.228 | CL101030010 | 0 | 484.788235294117 |
| features.properties.sum.229 | CL101030007 | 0 | 93.2392156862745 |
| features.properties.sum.230 | CL101030006 | 0 | 23.5411764705882 |
| features.properties.sum.231 | CL101030004 | 0 | 6.06666666666667 |
| features.properties.sum.232 | CL101030001 | 0 | 161.494117647059 |
| features.properties.sum.233 | CL101030002 | 0 | 184.992156862745 |
| features.properties.sum.234 | CL101030003 | 0 | 19.8980392156863 |
| features.properties.sum.235 | CL101021027 | 0 | 19.6196078431373 |
| features.properties.sum.236 | CL101030005 | 0 | 6.46666666666667 |
| features.properties.sum.237 | CL101030025 | 0 | 167.596078431373 |
| features.properties.sum.238 | CL101030024 | 0 | 27.8 |
| features.properties.sum.239 | CL101030023 | 0 | 24.0117647058824 |
| features.properties.sum.240 | CL101030022 | 0 | 57.3529411764706 |
| features.properties.sum.241 | CL101030021 | 0 | 39.2470588235294 |
| features.properties.sum.242 | CL101040010 | 0 | 5.65098039215686 |
| features.properties.sum.243 | CL101410006 | 0 | 7.51764705882353 |
| features.properties.sum.244 | CL101300033 | 0 | 18.2117647058823 |
| features.properties.sum.245 | CL101021019 | 0 | 460.937254901961 |
| features.properties.sum.246 | CL101001023 | 13.121568627451 | 630.4 |
| features.properties.sum.247 | CL101010007 | 18.1372549019608 | 140.662745098039 |
| features.properties.sum.248 | CL101010013 | 102.537254901961 | 557.658823529412 |
| features.properties.sum.249 | CL101010014 | 16.6705882352941 | 411.498039215686 |
| features.properties.sum.250 | CL101021010 | 9.16470588235294 | 89.8274509803921 |
| features.properties.sum.251 | CL101021004 | 26.0117647058824 | 716.745098039216 |
| features.properties.sum.252 | CL101010026 | 40.0039215686274 | 555.819607843138 |
| features.properties.sum.253 | CL101021007 | 8.75294117647059 | 551.666666666667 |
| features.properties.sum.254 | CL101010022 | 7.38823529411765 | 577.482352941177 |
| features.properties.sum.255 | CL101021008 | 3.10980392156863 | 342.776470588235 |
| features.properties.sum.256 | CL101010024 | 0 | 271.811764705882 |
| features.properties.sum.257 | CL101010025 | 0 | 259.737254901961 |
| features.properties.sum.258 | CL101030008 | 25.5294117647059 | 98.9686274509804 |
| features.properties.sum.259 | CL101021029 | 27.8235294117647 | 169.835294117647 |
| features.properties.sum.260 | CL101030013 | 6 | 85.7803921568628 |
| features.properties.sum.261 | CL101030016 | 0 | 9.71372549019608 |
| features.properties.sum.262 | CL101030018 | 0 | 17.721568627451 |
| features.properties.sum.263 | CL101030015 | 2.70588235294118 | 290.086274509804 |
| features.properties.sum.264 | CL101202018 | 0.00392156862745098 | 36.956862745098 |
| features.properties.sum.265 | CL101202002 | 0 | 43.4549019607843 |
| features.properties.sum.266 | CL101202025 | 0 | 31.3607843137255 |
| features.properties.sum.267 | CL101202006 | 7.47450980392157 | 35.8823529411765 |
| features.properties.sum.268 | CL101202010 | 150.070588235294 | 210.423529411765 |
| features.properties.sum.269 | CL101202017 | 5.33725490196078 | 45.8823529411765 |
| features.properties.sum.270 | CL101202021 | 0 | 30.6666666666667 |
| features.properties.sum.271 | CL101202023 | 0 | 15.7843137254902 |
| features.properties.sum.272 | CL101000001 | 0 | 26.9921568627451 |
| features.properties.sum.273 | CL101000006 | 49.4156862745098 | 141.356862745098 |
| features.properties.sum.274 | CL101200001 | 60.1333333333333 | 65.6980392156863 |
| features.properties.sum.275 | CL101021028 | 3.55686274509804 | 14.6039215686275 |
| features.properties.sum.276 | CL101021002 | 1.04313725490196 | 132.333333333333 |
| features.properties.sum.277 | CL101021017 | 2.01176470588235 | 83.5764705882353 |
| features.properties.sum.278 | CL101001025 | 3.7921568627451 | 233.278431372549 |
| features.properties.sum.279 | CL101001015 | 4.8156862745098 | 162.078431372549 |
| features.properties.sum.280 | CL101001016 | 0.333333333333333 | 86.4 |
| features.properties.sum.281 | CL101001013 | 1.07450980392157 | 17.4823529411765 |
| features.properties.sum.282 | CL101001012 | 9.05098039215686 | 85.9529411764706 |
| features.properties.sum.283 | CL101001014 | 7.94117647058823 | 134.713725490196 |
| features.properties.sum.284 | CL101001006 | 0 | 37.878431372549 |
| features.properties.sum.285 | CL101202011 | 0 | 13.7529411764706 |
| features.properties.sum.286 | CL101202008 | 1.18823529411765 | 21.1686274509804 |
| features.properties.sum.287 | CL101202020 | 1 | 21.2156862745098 |
| features.properties.sum.288 | CL101202024 | 0 | 33.6862745098039 |
| features.properties.sum.289 | CL101201006 | 37.843137254902 | 118.733333333333 |
| features.properties.sum.290 | CL101201007 | 40.6549019607843 | 62.8156862745098 |
| features.properties.sum.291 | CL101201008 | 15.1019607843137 | 72.2196078431372 |
| features.properties.sum.292 | CL101200003 | 4.22352941176471 | 16.6941176470588 |
| features.properties.sum.293 | CL101200002 | 34.9843137254902 | 52.8901960784314 |
| features.properties.sum.294 | CL101200018 | 7.08627450980392 | 109.541176470588 |
| features.properties.sum.295 | CL101200011 | 11.8588235294118 | 38.3921568627451 |
| features.properties.sum.296 | CL101200019 | 7.43529411764706 | 27.4509803921569 |
| features.properties.sum.297 | CL101200026 | 1.87450980392157 | 21.3176470588235 |
| features.properties.sum.298 | CL101200028 | 14.7450980392157 | 30.0941176470588 |
| features.properties.sum.299 | CL101200027 | 8.86666666666667 | 11.9921568627451 |
| features.properties.sum.300 | CL101200029 | 249.8 | 303.517647058824 |
| features.properties.sum.301 | CL101202005 | 7.6156862745098 | 38.4901960784314 |
| features.properties.sum.302 | CL101202019 | 0.0509803921568627 | 19.8901960784314 |
| features.properties.sum.303 | CL101021003 | 0 | 13.8039215686274 |
| features.properties.sum.304 | CL101021001 | 0 | 105.776470588235 |
| features.properties.sum.305 | CL101010023 | 0.309803921568627 | 89.7254901960784 |
| features.properties.sum.306 | CL101021015 | 4.15686274509804 | 150.352941176471 |
| features.properties.sum.307 | CL101010019 | 9.64313725490196 | 84.5568627450981 |
| features.properties.sum.308 | CL101021016 | 10.3411764705882 | 35.7529411764706 |
| features.properties.sum.309 | CL101001028 | 4.00392156862745 | 41.7098039215686 |
| features.properties.sum.310 | CL101001026 | 6.65098039215686 | 134.305882352941 |
| features.properties.sum.311 | CL101001027 | 6.48235294117647 | 68.1137254901961 |
| features.properties.sum.312 | CL101001024 | 0 | 55.6627450980392 |
| features.properties.sum.313 | CL101001033 | 65.3176470588235 | 188.521568627451 |
| features.properties.sum.314 | CL101020004 | 30.2549019607843 | 52.6745098039216 |
| features.properties.sum.315 | CL101020005 | 13.321568627451 | 50.156862745098 |
| features.properties.sum.316 | CL101021009 | 18.443137254902 | 138.247058823529 |
| features.properties.sum.317 | CL101021006 | 10.0509803921569 | 351.890196078431 |
| features.properties.sum.318 | CL101200031 | 8.85098039215686 | 57.9019607843137 |
| features.properties.sum.319 | CL101200030 | 12.1647058823529 | 409.529411764706 |
| features.properties.sum.320 | CL101200010 | 34.043137254902 | 53.0156862745098 |
| features.properties.sum.321 | CL101200009 | 36.5921568627451 | 203.905882352941 |
| features.properties.sum.322 | CL101202004 | 1.13725490196078 | 12.7450980392157 |
| features.properties.sum.323 | CL101202012 | 0.662745098039216 | 20.8470588235294 |
| features.properties.sum.324 | CL101001017 | 5.22745098039216 | 217.407843137255 |
| features.properties.sum.325 | CL101010010 | 19.2509803921569 | 174.894117647059 |
| features.properties.sum.326 | CL101021018 | 13.3019607843137 | 116.592156862745 |