Aula Prática 02 - Curso SITS / Equipe Prodes
1. Instalação do pacote SITS
# ("run" se ainda não estiver instalado na sua máquina)
# install.packages("sits", dependencies = TRUE)2. Carregar pacotes e bibliotecas
library(sits)## SITS - satellite image time series analysis.## Loaded sits v1.5.4.
## See ?sits for help, citation("sits") for use in publication.
## Documentation avaliable in https://e-sensing.github.io/sitsbook/# https://github.com/e-sensing/sits
library(tibble)
# https://tibble.tidyverse.org/
# pacote para manipulação de tabelas no formato "tibble";
# oferece melhor visualização, indexação mais clara e integração com pacotes do tidyverse.trtrtrrtrtrt
library(dplyr)##
## Anexando pacote: 'dplyr'## Os seguintes objetos são mascarados por 'package:stats':
##
## filter, lag## Os seguintes objetos são mascarados por 'package:base':
##
## intersect, setdiff, setequal, union# https://dplyr.tidyverse.org/
# pacote para manipulação eficiente de dados tabulares;
# essencial para análise de dados, especialmente em fluxos de trabalho do tidyverse.3. Criar diretórios
# criar diretórios:
treinamento_dir <- "~/treinamento_sits_prodes"
dir.create(treinamento_dir, recursive = TRUE)## Warning in dir.create(treinamento_dir, recursive = TRUE):
## 'C:\Users\User\Documents\treinamento_sits_prodes' já existe# diretório de dados
data_dir <- "~/treinamento_sits_prodes/data_dir"
temp_dir <- "~/treinamento_sits_prodes/data_dir/temp_dir"
dir.create(temp_dir, recursive = TRUE)## Warning in dir.create(temp_dir, recursive = TRUE):
## 'C:\Users\User\Documents\treinamento_sits_prodes\data_dir\temp_dir' já existe_____________________________________________________________________________4_
4. Cubos
4.1. Cubo não-regular
# usar a função sits_cube para criar cubo de dados.
cube <- sits_cube(
source = "MPC",
# (https://planetarycomputer.microsoft.com/dataset/sentinel-2-l2a)
collection = "SENTINEL-2-L2A",
bands = c("B11", "B8A", "B04"),
tiles = "20LMR",
start_date = '2022-08-23',
end_date = '2022-12-23'
)## | | | 0% | |======================================================================| 100%4.1.1 Visualizar timeline
# timeline do cubo irregular
sits_timeline(cube)## [1] "2022-08-23" "2022-08-26" "2022-08-28" "2022-08-31" "2022-09-02"
## [6] "2022-09-05" "2022-09-07" "2022-09-10" "2022-09-12" "2022-09-15"
## [11] "2022-09-17" "2022-09-20" "2022-09-22" "2022-09-25" "2022-09-27"
## [16] "2022-09-30" "2022-10-02" "2022-10-05" "2022-10-07" "2022-10-10"
## [21] "2022-10-12" "2022-10-15" "2022-10-17" "2022-10-20" "2022-10-22"
## [26] "2022-10-25" "2022-10-27" "2022-10-30" "2022-11-01" "2022-11-04"
## [31] "2022-11-06" "2022-11-09" "2022-11-11" "2022-11-14" "2022-11-16"
## [36] "2022-11-19" "2022-11-21" "2022-11-24" "2022-11-26" "2022-11-29"
## [41] "2022-12-01" "2022-12-04" "2022-12-06" "2022-12-09" "2022-12-11"
## [46] "2022-12-14" "2022-12-16" "2022-12-19" "2022-12-21"4.1.2 Plotar uma imagem do cubo
# plot cubo - não regularizado
plot(cube, red = "B11", green = "B8A", blue = "B04", date = "2022-11-24")## Warning in CPL_gdalinfo(if (missing(source)) character(0) else source, options,
## : GDAL Message 1: HTTP response code on
## https://sentinel2l2a01.blob.core.windows.net/sentinel2-l2/20/L/MR/2022/08/23/S2B_MSIL2A_20220823T142709_N0400_R053_T20LMR_20220824T143741.SAFE/GRANULE/L2A_T20LMR_A028536_20220823T143247/IMG_DATA/R10m/T20LMR_20220823T142709_B04_10m.tif?st=2026-02-05T03%3A05%3A43Z&se=2026-02-06T03%3A50%3A43Z&sp=rl&sv=2025-07-05&sr=c&skoid=9c8ff44a-6a2c-4dfb-b298-1c9212f64d9a&sktid=72f988bf-86f1-41af-91ab-2d7cd011db47&skt=2026-02-05T21%3A33%3A32Z&ske=2026-02-12T21%3A33%3A32Z&sks=b&skv=2025-07-05&sig=eNT4s5oscKxUVcVm5NP648TS2wpnIzjumrrc%2FTeXFhI%3D.vat.dbf:
## 403
4.2. Cubo regular
- os dados utilizados para construção desse cubo já passaram pelo processo de regularização (sits_regularize);
# pasta com imagens regularizadas
images <- paste0(data_dir,"/images")4.2.1 Carregar cubo regularizado
cube_reg <- sits_cube(
source = "MPC",
collection = "SENTINEL-2-L2A",
tiles = "20LMR",
bands = c("B11", "B8A","B08", "B04"),
start_date = '2022-08-23',
end_date = '2022-12-23',
data_dir = images
)## | | | 0% | |========= | 12% | |================== | 25% | |========================== | 38% | |=================================== | 50% | |============================================ | 62% | |==================================================== | 75% | |============================================================= | 88% | |======================================================================| 100%42.1 Visualizar timeline
# timeline do cubo regular (16 dias)
sits_timeline(cube_reg)## [1] "2022-09-02" "2022-09-18" "2022-10-04" "2022-10-20" "2022-11-05"
## [6] "2022-11-21" "2022-12-07" "2022-12-23"4.2.2 Plotar uma imagem do cubo
# plot cubo Regularizado
plot(cube_reg, red = "B11", green = "B8A", blue = "B04", date = "2022-11-21")
____________________________________________________________________________5__
5. Operações com cubos
5.1 Calculando de NDVI
- Uso da função sits_select()
# filtrar cube_reg
cube_ndvi<- sits_select(cube_reg,
bands = c('B11', 'B08', 'B04'),
start_date = '2022-09-23',
end_date = '2022-12-23') - Operação de bandas
# filtrar cubo
cube_ndvi <- sits_apply (cube_ndvi,
NDVI = (B08 - B04)/(B08 + B04),
output_dir = temp_dir)## | | | 0% | |============ | 17% | |======================= | 33% | |=================================== | 50% | |=============================================== | 67% | |========================================================== | 83% | |======================================================================| 100%# visulizar bandas
sits_bands(cube_ndvi)## [1] "B04" "B08" "B11" "NDVI"# timeline
sits_timeline(cube_ndvi)## [1] "2022-10-04" "2022-10-20" "2022-11-05" "2022-11-21" "2022-12-07"
## [6] "2022-12-23"- Plot NDVI
plot(cube_ndvi, band = "NDVI", date = "2022-11-05")
____________________________________________________________________________6__
6. Dataset Rondônia
6.1 Carregar dataset
# caminho para acessar o dataset
samples_path <- ("~/treinamento_sits_prodes/data_dir/samples/deforestation_samples_v18.rds")
#samples_path <- paste0(data_dir,"/samples/deforestation_samples_v18.rds")
# Carregando o arquivo dataset
deforestation_samples_v18 <- readRDS(samples_path)6.2 Exploração do dataset
6.2.1 Visualização da espacialização do dataset
sits_view(deforestation_samples_v18)## Registered S3 method overwritten by 'jsonify':
## method from
## print.json jsonlite6.2.2 Acessando dataset
- Balanceamento
summary(deforestation_samples_v18)## # A tibble: 9 × 3
## label count prop
## <chr> <int> <dbl>
## 1 Clear_Cut_Bare_Soil 944 0.157
## 2 Clear_Cut_Burned_Area 983 0.164
## 3 Clear_Cut_Vegetation 603 0.100
## 4 Forest 964 0.160
## 5 Mountainside_Forest 211 0.0351
## 6 Riparian_Forest 1247 0.208
## 7 Seasonally_Flooded 731 0.122
## 8 Water 109 0.0181
## 9 Wetland 215 0.0358- Manipulação
deforestation_samples_v18_CCBS <- dplyr::filter(deforestation_samples_v18, label %in% c("Clear_Cut_Bare_Soil"))
summary(deforestation_samples_v18_CCBS)## # A tibble: 1 × 3
## label count prop
## <chr> <int> <dbl>
## 1 Clear_Cut_Bare_Soil 944 1- Visualização das datas que compoem a série temporal
sits_timeline(deforestation_samples_v18)## [1] "2022-01-05" "2022-01-21" "2022-02-06" "2022-02-22" "2022-03-10"
## [6] "2022-03-26" "2022-04-11" "2022-04-27" "2022-05-13" "2022-05-29"
## [11] "2022-06-14" "2022-06-30" "2022-07-16" "2022-08-01" "2022-08-17"
## [16] "2022-09-02" "2022-09-18" "2022-10-04" "2022-10-20" "2022-11-05"
## [21] "2022-11-21" "2022-12-07" "2022-12-23"- Visualizaçao de séries temporais (original):
# séries temporais originais
deforestation_samples_v18 |> dplyr::filter(label %in% c("Clear_Cut_Bare_Soil"))|>
sits_select(bands = c("B11", "B08", "B04")) |>
plot()
- Visualizaçao suavizada dos padrões das séries: (modelo aditivo generativo)
# séries GAM
deforestation_samples_v18 |> dplyr::filter(label %in% c("Clear_Cut_Bare_Soil"))|>
sits_select(bands = c("B11", "B08", "B04")) |>
sits_patterns() |>
plot()
Visualizaçao individual de séries (instância)
deforestation_samples_v18 |> dplyr::filter(label %in% c("Clear_Cut_Bare_Soil"))|> slice(1)|>
plot()
___________________________________________________________________________7___
7. Extraindo séries temporais
7.1 Construção do cubo completo
cube_reg_completo <- sits_cube(
source = "MPC",
collection = "SENTINEL-2-L2A",
tiles = "20LMR",
bands = c("B11", "B8A","B08", "B04"),
start_date = '2022-01-05',
end_date = '2022-12-23',
data_dir = images
)## | | | 0% | |=== | 4% | |====== | 9% | |========= | 13% | |============ | 17% | |=============== | 22% | |================== | 26% | |===================== | 30% | |======================== | 35% | |=========================== | 39% | |============================== | 43% | |================================= | 48% | |===================================== | 52% | |======================================== | 57% | |=========================================== | 61% | |============================================== | 65% | |================================================= | 70% | |==================================================== | 74% | |======================================================= | 78% | |========================================================== | 83% | |============================================================= | 87% | |================================================================ | 91% | |=================================================================== | 96% | |======================================================================| 100%7.2 Preparação do dado
# criação de um novo dataset, sem a coluna "time_series"
deforestation_samples_v18_novo <- deforestation_samples_v18 |>
dplyr::select(-time_series) #apagando timeseries7.3 Extração de séries temporais
series_temporais_2022 <- sits_get_data(
cube = cube_reg_completo,
samples = deforestation_samples_v18_novo,
start_date = '2022-01-05',
end_date = '2022-12-23',
progress = FALSE
)series_temporais_2022 |> dplyr::filter(label %in% c("Clear_Cut_Bare_Soil"))|>
sits_select(bands = c("B11", "B08", "B04")) |>
sits_patterns() |>
plot()