Mapping Agricultural Intensification in the Brazilian Savanna: A Machine Learning Approach Using Harmonized Data from Landsat Sentinel-2
Agricultural intensification practices have been adopted in the Brazilian savanna (Cerrado), mainly in the transition between Cerrado and the Amazon Forest, to increase productivity while reducing pressure for new land clearing. Due to the growing demand for more sustainable practices, more accurate...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/12/7/263 |
| _version_ | 1797436007675592704 |
|---|---|
| author | Édson Luis Bolfe Taya Cristo Parreiras Lucas Augusto Pereira da Silva Edson Eyji Sano Giovana Maranhão Bettiol Daniel de Castro Victoria Ieda Del’Arco Sanches Luiz Eduardo Vicente |
| author_facet | Édson Luis Bolfe Taya Cristo Parreiras Lucas Augusto Pereira da Silva Edson Eyji Sano Giovana Maranhão Bettiol Daniel de Castro Victoria Ieda Del’Arco Sanches Luiz Eduardo Vicente |
| author_sort | Édson Luis Bolfe |
| collection | DOAJ |
| description | Agricultural intensification practices have been adopted in the Brazilian savanna (Cerrado), mainly in the transition between Cerrado and the Amazon Forest, to increase productivity while reducing pressure for new land clearing. Due to the growing demand for more sustainable practices, more accurate information on geospatial monitoring is required. Remote sensing products and artificial intelligence models for pixel-by-pixel classification have great potential. Therefore, we developed a methodological framework with spectral indices (Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and Soil-Adjusted Vegetation Index (SAVI)) derived from the Harmonized Landsat Sentinel-2 (HLS) and machine learning algorithms (Random Forest (RF), Artificial Neural Networks (ANNs), and Extreme Gradient Boosting (XGBoost)) to map agricultural intensification considering three hierarchical levels, i.e., temporary crops (level 1), the number of crop cycles (level 2), and the crop types from the second season in double-crop systems (level 3) in the 2021–2022 crop growing season in the municipality of Sorriso, Mato Grosso State, Brazil. All models were statistically similar, with an overall accuracy between 85 and 99%. The NDVI was the most suitable index for discriminating cultures at all hierarchical levels. The RF-NDVI combination mapped best at level 1, while at levels 2 and 3, the best model was XGBoost-NDVI. Our results indicate the great potential of combining HLS data and machine learning to provide accurate geospatial information for decision-makers in monitoring agricultural intensification, with an aim toward the sustainable development of agriculture. |
| first_indexed | 2024-03-09T10:55:29Z |
| format | Article |
| id | doaj.art-9da7a7dfb93b4b048659f08e0f9f0a39 |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-03-09T10:55:29Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-9da7a7dfb93b4b048659f08e0f9f0a392023-12-01T01:39:23ZengMDPI AGISPRS International Journal of Geo-Information2220-99642023-07-0112726310.3390/ijgi12070263Mapping Agricultural Intensification in the Brazilian Savanna: A Machine Learning Approach Using Harmonized Data from Landsat Sentinel-2Édson Luis Bolfe0Taya Cristo Parreiras1Lucas Augusto Pereira da Silva2Edson Eyji Sano3Giovana Maranhão Bettiol4Daniel de Castro Victoria5Ieda Del’Arco Sanches6Luiz Eduardo Vicente7Brazilian Agricultural Research Corporation (Embrapa Agricultura Digital), Campinas 13083-886, BrazilInstitute of Geosciences, State University of Campinas (Unicamp), Campinas 13083-855, BrazilInstitute of Geography, Federal University of Uberlândia (UFU), Uberlândia 38408-100, BrazilBrazilian Agricultural Research Corporation (Embrapa Cerrados), Planaltina 73301-970, BrazilBrazilian Agricultural Research Corporation (Embrapa Cerrados), Planaltina 73301-970, BrazilBrazilian Agricultural Research Corporation (Embrapa Agricultura Digital), Campinas 13083-886, BrazilNational Institute for Space Research (INPE), São José dos Campos 12227-010, BrazilBrazilian Agricultural Research Corporation (Embrapa Meio Ambiente), Jaguariúna 13820-000, BrazilAgricultural intensification practices have been adopted in the Brazilian savanna (Cerrado), mainly in the transition between Cerrado and the Amazon Forest, to increase productivity while reducing pressure for new land clearing. Due to the growing demand for more sustainable practices, more accurate information on geospatial monitoring is required. Remote sensing products and artificial intelligence models for pixel-by-pixel classification have great potential. Therefore, we developed a methodological framework with spectral indices (Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and Soil-Adjusted Vegetation Index (SAVI)) derived from the Harmonized Landsat Sentinel-2 (HLS) and machine learning algorithms (Random Forest (RF), Artificial Neural Networks (ANNs), and Extreme Gradient Boosting (XGBoost)) to map agricultural intensification considering three hierarchical levels, i.e., temporary crops (level 1), the number of crop cycles (level 2), and the crop types from the second season in double-crop systems (level 3) in the 2021–2022 crop growing season in the municipality of Sorriso, Mato Grosso State, Brazil. All models were statistically similar, with an overall accuracy between 85 and 99%. The NDVI was the most suitable index for discriminating cultures at all hierarchical levels. The RF-NDVI combination mapped best at level 1, while at levels 2 and 3, the best model was XGBoost-NDVI. Our results indicate the great potential of combining HLS data and machine learning to provide accurate geospatial information for decision-makers in monitoring agricultural intensification, with an aim toward the sustainable development of agriculture.https://www.mdpi.com/2220-9964/12/7/263multisensorHLSagricultureartificial intelligenceremote sensingCerrado |
| spellingShingle | Édson Luis Bolfe Taya Cristo Parreiras Lucas Augusto Pereira da Silva Edson Eyji Sano Giovana Maranhão Bettiol Daniel de Castro Victoria Ieda Del’Arco Sanches Luiz Eduardo Vicente Mapping Agricultural Intensification in the Brazilian Savanna: A Machine Learning Approach Using Harmonized Data from Landsat Sentinel-2 ISPRS International Journal of Geo-Information multisensor HLS agriculture artificial intelligence remote sensing Cerrado |
| title | Mapping Agricultural Intensification in the Brazilian Savanna: A Machine Learning Approach Using Harmonized Data from Landsat Sentinel-2 |
| title_full | Mapping Agricultural Intensification in the Brazilian Savanna: A Machine Learning Approach Using Harmonized Data from Landsat Sentinel-2 |
| title_fullStr | Mapping Agricultural Intensification in the Brazilian Savanna: A Machine Learning Approach Using Harmonized Data from Landsat Sentinel-2 |
| title_full_unstemmed | Mapping Agricultural Intensification in the Brazilian Savanna: A Machine Learning Approach Using Harmonized Data from Landsat Sentinel-2 |
| title_short | Mapping Agricultural Intensification in the Brazilian Savanna: A Machine Learning Approach Using Harmonized Data from Landsat Sentinel-2 |
| title_sort | mapping agricultural intensification in the brazilian savanna a machine learning approach using harmonized data from landsat sentinel 2 |
| topic | multisensor HLS agriculture artificial intelligence remote sensing Cerrado |
| url | https://www.mdpi.com/2220-9964/12/7/263 |
| work_keys_str_mv | AT edsonluisbolfe mappingagriculturalintensificationinthebraziliansavannaamachinelearningapproachusingharmonizeddatafromlandsatsentinel2 AT tayacristoparreiras mappingagriculturalintensificationinthebraziliansavannaamachinelearningapproachusingharmonizeddatafromlandsatsentinel2 AT lucasaugustopereiradasilva mappingagriculturalintensificationinthebraziliansavannaamachinelearningapproachusingharmonizeddatafromlandsatsentinel2 AT edsoneyjisano mappingagriculturalintensificationinthebraziliansavannaamachinelearningapproachusingharmonizeddatafromlandsatsentinel2 AT giovanamaranhaobettiol mappingagriculturalintensificationinthebraziliansavannaamachinelearningapproachusingharmonizeddatafromlandsatsentinel2 AT danieldecastrovictoria mappingagriculturalintensificationinthebraziliansavannaamachinelearningapproachusingharmonizeddatafromlandsatsentinel2 AT iedadelarcosanches mappingagriculturalintensificationinthebraziliansavannaamachinelearningapproachusingharmonizeddatafromlandsatsentinel2 AT luizeduardovicente mappingagriculturalintensificationinthebraziliansavannaamachinelearningapproachusingharmonizeddatafromlandsatsentinel2 |