@Article{CarreirasJoneLucaShim:2017:MaMaLa,
               author = "Carreiras, Jo{\~a}o M. B. and Jones, Joshua and Lucas, Richard M. 
                         and Shimabukuro, Yosio Edemir",
          affiliation = "{University of Sheffield} and {Aberystwyth University} and 
                         {University of New South Wales} and {Instituto Nacional de 
                         Pesquisas Espaciais (INPE)}",
                title = "Mapping major land cover types and retrieving the age of secondary 
                         forests in the Brazilian Amazon by combining single-date optical 
                         and radar remote sensing data",
              journal = "Remote Sensing of Environment",
                 year = "2017",
               volume = "194",
                pages = "16--32",
                month = "June",
             keywords = "Age of secondary forests, ALOS PALSAR, Amazon, Landsat TM, Random 
                         forests, Tropical secondary forests.",
             abstract = "Secondary forests play an important role in restoring carbon and 
                         biodiversity lost previously through deforestation and degradation 
                         and yet there is little information available on the extent of 
                         different successional stages. Such knowledge is particularly 
                         needed in tropical regions where past and current disturbance 
                         rates have been high but regeneration is rapid. Focusing on three 
                         areas in the Brazilian Amazon (Manaus, Santar{\'e}m, Machadinho 
                         d'Oeste), this study aimed to evaluate the use of single-date 
                         Landsat Thematic Mapper (TM) and Advanced Land Observing Satellite 
                         (ALOS) Phased Arrayed L-band Synthetic Aperture Radar (PALSAR) 
                         data in the 20072010 period for i) discriminating mature forest, 
                         non-forest and secondary forest, and ii) retrieving the age of 
                         secondary forests (ASF), with 100 m × 100 m training areas 
                         obtained by the analysis of an extensive time-series of Landsat 
                         sensor data over the three sites. A machine learning algorithm 
                         (random forests) was used in combination with ALOS PALSAR 
                         backscatter intensity at HH and HV polarizations and Landsat 5 TM 
                         surface reflectance in the visible, near-infrared and shortwave 
                         infrared spectral regions. Overall accuracy when discriminating 
                         mature forest, non-forest and secondary forest is high (9596%), 
                         with the highest errors in the secondary forest class (omission 
                         and commission errors in the range 46% and 1220% respectively) 
                         because of misclassification as mature forest. Root mean square 
                         error (RMSE) and bias when retrieving ASF ranged between 4.34.7 
                         years (relative RMSE = 25.532.0%) and 0.040.08 years respectively. 
                         On average, unbiased ASF estimates can be obtained using the 
                         method proposed here (Wilcoxon test, p-value > 0.05). However, the 
                         bias decomposition by 5-year interval ASF classes showed that most 
                         age estimates are biased, with consistent overestimation in 
                         secondary forests up to 1015 years of age and underestimation in 
                         secondary forests of at least 20 years of age. Comparison with the 
                         classification results obtained from the analysis of extensive 
                         time-series of Landsat sensor data showed a good agreement, with 
                         Pearson's coefficient of correlation (R) of the proportion of 
                         mature forest, non-forest and secondary forest at 1-km grid cells 
                         ranging between 0.970.98, 0.960.98 and 0.840.90 in the 20072010 
                         period, respectively. The agreement was lower (R = 0.820.85) when 
                         using the same dataset to compare the ability of ALOS PALSAR and 
                         Landsat 5 TM data to retrieve ASF. This was also dependent on the 
                         study area, especially when considering mapping secondary forest 
                         and retrieving ASF, with Manaus displaying better agreement when 
                         compared to the results at Santar{\'e}m and Machadinho d'Oeste.",
                  doi = "10.1016/j.rse.2017.03.016",
                  url = "http://dx.doi.org/10.1016/j.rse.2017.03.016",
                 issn = "0034-4257",
             language = "en",
           targetfile = "carreiras_mapping.pdf",
        urlaccessdate = "27 abr. 2024"
}