Mapping Tree Canopy Cover and Aboveground Biomass  in Sudano-Sahelian Woodlands Using Landsat 8 and  Random Forest

Karlson, Martin; Ostwald, Madelene; Reese, Heather; Sanou, Josias; Tankoano, Boalidioa; Mattsson, Eskil

doi:10.3390/rs70810017

Cited by 249 publications

(162 citation statements)

References 91 publications

(141 reference statements)

Supporting

Mentioning

149

Contrasting

Unclassified

Order By: Relevance

“…RF is a classification and regression tree approach that is often efficient in the predictive model [59], with each tree randomly selected from the subsets of predictor variables. The number of trees (ntree) is yielded by the original data, based on a bootstrap sample, which was determined from the relationship between N and the error.…”

Section: Statistical Analysis and Modelingmentioning

confidence: 99%

Integrating Airborne LiDAR and Optical Data to Estimate Forest Aboveground Biomass in Arid and Semi-Arid Regions of China

Luodan

Pan

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Forest Aboveground Biomass (AGB) is a key parameter for assessing forest productivity and global carbon content. In previous studies, AGB has been estimated using various prediction methods and types of remote sensing data. Increasingly, there is a trend towards integrating various data sources such as Light Detection and Ranging (LiDAR) and optical data. In this study, we constructed and compared the accuracies of five models for estimating AGB of forests in the upper Heihe River Basin in Northwest China. The five models were constructed using field and remotely-sensed data (optical and LiDAR) and algorithms including Random Forest (RF), Support Vector Machines (SVM), Back Propagation Neural Networks (BPNN), K-Nearest Neighbor (KNN) and the Generalized Linear Mixed Model (GLMM). Models based on the RF algorithm emerged as being the best among the five algorithms irrespective of the datasets used. The Random Forest AGB model, using only LiDAR data (R 2 = 0.899, RMSE = 14.0 t/ha) as the input data, was more effective than the one using optical data (R 2 = 0.835, RMSE = 22.724 t/ha). Compared to LiDAR or optical data alone, the AGB model (R 2 = 0.913, RMSE = 13.352 t/ha) that used the RF algorithm and integrated LiDAR and optical data was found to be optimal. Incorporation of terrain variables with optical data resulted in only slight improvements in accuracy. The models developed in this study could be useful for using integrated airborne LiDAR and passive optical data to accurately estimate forest biomass.

show abstract

Section: Statistical Analysis and Modelingmentioning

confidence: 99%

Integrating Airborne LiDAR and Optical Data to Estimate Forest Aboveground Biomass in Arid and Semi-Arid Regions of China

Luodan

Pan

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…RF is relatively insensitive to the number and multi-collinearity of input data, and makes no assumptions about distributions. Furthermore, it has been shown to provide reliable and stable classification results, outperforming other classifiers [56][57][58][59]. In our classification, we used 1000 trees for the RF model, and the number of selected features was set as the square root of all features.…”

Section: Land Cover and Land Use Classification And Change Analysismentioning

confidence: 99%

Agricultural Expansion and Intensification in the Foothills of Mount Kenya: A Landscape Perspective

et al. 2017

View full text Add to dashboard Cite

This study spatially assesses, quantifies, and visualizes the agricultural expansion and land use intensification in the northwestern foothills of Mount Kenya over the last 30 years: processes triggered by population growth, and, more recently, by large-scale commercial investments. We made use of Google Earth Engine to access the USGS Landsat data archive and to generate cloud-free seasonal composites. These enabled us to accurately differentiate between rainfed and irrigated cropland, which was important for assessing agricultural intensification. We developed three land cover and land use classifications using the random forest classifier, and assessed land cover and land use change by creating cross-tabulation matrices for the intervals from 1987 to 2002, 2002 to 2016, and 1987 to 2016 and calculating the net change. We then applied a landscape mosaic approach to each classification to identify landscape types categorized by land use intensity. We discuss the impacts of landscape changes on natural habitats, biodiversity, and water. Kappa accuracies for the three classifications lay between 78.3% and 82.1%. Our study confirms that rainfed and irrigated cropland expanded at the expense of natural habitats, including protected areas. Agricultural expansion took place mainly in the 1980s and 1990s, whereas agricultural intensification largely happened after 2000. Since then, not only large-scale producers, but also many smallholders have begun to practice irrigated farming. The spatial pattern of agricultural expansion and intensification in the study area is defined by water availability. Agricultural intensification and the expansion of horticulture agribusinesses increase pressure on water. Furthermore, the observed changes have heightened pressure on pasture and idle land due to the decrease in natural and agropastoral landscapes. Conflicts between pastoralists, smallholder farmers, large-scale ranches, and wildlife might further increase, particularly during the dry seasons and in years of extreme drought.

show abstract

“…A random forest is generated through the creation of a series of CARTs using bootstrapping, or resampling with replacement. Random forest is a flexible and powerful nonparametric technique that many mapping applications have recently implemented for a range of studies including mapping crops [44][45][46], wetlands [47,48], canopy height [49], algal blooms [50], urban sprawl [51], biomass [52], plantations [42], and many other thematic areas.…”

Section: Land Use Land Cover Mappingmentioning

confidence: 99%

Monitoring Rice Agriculture across Myanmar Using Time Series Sentinel-1 Assisted by Landsat-8 and PALSAR-2

et al. 2017

View full text Add to dashboard Cite

Assessment and monitoring of rice agriculture over large areas has been limited by cloud cover, optical sensor spatial and temporal resolutions, and lack of systematic or open access radar. Dense time series of open access Sentinel-1 C-band data at moderate spatial resolution offers new opportunities for monitoring agriculture. This is especially pertinent in South and Southeast Asia where rice is critical to food security and mostly grown during the rainy seasons when high cloud cover is present. In this research application, time series Sentinel-1A Interferometric Wide images (632) were utilized to map rice extent, crop calendar, inundation, and cropping intensity across Myanmar. An updated (2015) land use land cover map fusing Sentinel-1, Landsat-8 OLI, and PALSAR-2 were integrated and classified using a randomforest algorithm. Time series phenological analyses of the dense Sentinel-1 data were then executed to assess rice information across all of Myanmar. The broad land use land cover map identified 186,701 km 2 of cropland across Myanmar with mean out-of-sample kappa of over 90%. A phenological time series analysis refined the cropland class to create a rice mask by extrapolating unique indicators tied to the rice life cycle (dynamic range, inundation, growth stages) from the dense time series Sentinel-1 to map rice paddy characteristics in an automated approach. Analyses show that the harvested rice area was 6,652,111 ha with general (R 2 = 0.78) agreement with government census statistics. The outcomes show strong ability to assess and monitor rice production at moderate scales over a large cloud-prone region. In countries such as Myanmar with large populations and governments dependent upon rice production, more robust and transparent monitoring and assessment tools can help support better decision making. These results indicate that systematic and open access Synthetic Aperture Radar (SAR) can help scale information required by food security initiatives and Monitoring, Reporting, and Verification programs.

show abstract

Mapping Tree Canopy Cover and Aboveground Biomass in Sudano-Sahelian Woodlands Using Landsat 8 and Random Forest

Cited by 249 publications

References 91 publications

Integrating Airborne LiDAR and Optical Data to Estimate Forest Aboveground Biomass in Arid and Semi-Arid Regions of China

Integrating Airborne LiDAR and Optical Data to Estimate Forest Aboveground Biomass in Arid and Semi-Arid Regions of China

Agricultural Expansion and Intensification in the Foothills of Mount Kenya: A Landscape Perspective

Monitoring Rice Agriculture across Myanmar Using Time Series Sentinel-1 Assisted by Landsat-8 and PALSAR-2

Contact Info

Product

Resources

About