Spatial-temporal mapping of forest vegetation cover changes along highways in Brunei using deep learning techniques and Sentinel-2 images

Kalinaki, Kassim; Malik, Owais Ahmed; Lai, Daphne Teck Ching; Sukri, Rahayu Sukmaria; Wahab, Rodzay Bin Haji Abdul

doi:10.1016/j.ecoinf.2023.102193

Cited by 12 publications

(3 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reeds of the Gramineae family, tamarisk shrub, and poplar tree were the dominant species (Figure 2), and minor vegetation types included camel thorn and bitter pea [38,44]. [45,46] (https://step.esa.int/main/snap-supported-plugins/sen2cor/sen2cor-v2-11/ (accessed on 15 December 2023)) in the SNAP Toolbox version 2.11.0 was used for atmospheric corrections, radiometric calibrations, and to obtain the Sentinel-2L2A image data. The NDVI was calculated to determine the confidence interval and the raster calculator tool was used to calculate the FVC, which was cropped to obtain the vegetation cover of the study area.…”

Section: Study Areamentioning

confidence: 99%

“…In this study, we used images of eight sample plots taken by UAV aerial photography at a size of 1000 m × 300 m, as shown in Figure 1, to calculate the vegetation cover using the image-element dichotomous model for the NDVI of Sentinel-2 in the same period and to verify the accuracy. [45,46] (https: //step.esa.int/main/snap-supported-plugins/sen2cor/sen2cor-v2-11/ (accessed on 15 December 2023)) in the SNAP Toolbox version 2.11.0 was used for atmospheric corrections, radiometric calibrations, and to obtain the Sentinel-2L2A image data. The NDVI was calculated to determine the confidence interval and the raster calculator tool was used to calculate the FVC, which was cropped to obtain the vegetation cover of the study area.…”

Section: 22 Drone Datamentioning

confidence: 99%

See 1 more Smart Citation

The Impact of Groundwater Burial Depth on the Vegetation of the Dariyabui Oasis in the Central Desert

Bai,

Guo,

Wang

et al. 2023

Sustainability

View full text Add to dashboard Cite

Vegetation and groundwater are important components of the ecological environment of oases in desert hinterlands and their relationship is crucial to ecosystem stability. In this study, Sentinel-2 data for 2016–2022 and measured groundwater burial depths were analysed for the Dariyabui Oasis in the hinterland of the Taklamakan Desert. The spatial and temporal changes in vegetation and groundwater burial depth from 2019 to 2022 were analysed based on the image–element dichotomous model of the normalised difference vegetation index, utilising the inverse distance weight interpolation method, cubic curve regression, image–element difference, slope trend analysis, and the Markov transfer matrix for determining the temporal and spatial response law between the two. Finally, the threshold value of groundwater burial depth for different vegetation cover types was clarified. The fractional vegetation cover of the Dariyabui Oasis showed a slight increase from 2016 to 2022. Vegetation in the northwest and southeast of the oasis increased, whereas vegetation decreased in the mid-north and northeast regions; 5.14% of the total area experienced increased coverage, whereas 3.35% experienced decreased coverage. The depth of groundwater in the oasis showed a pattern of gradual increase from the entrance to the end of the oasis, that is, south to north. The depth of groundwater in the oasis from 2019 to 2022 was stable, with a 4-year average depth of 4.1069 m and a maximum fluctuation of 0.4560 m. The interannual changes in the groundwater level showed an increasing trend in January–April, while groundwater levels showed a decreasing trend in May–July and August–October and remained constant in June–July and October–December. Oasis vegetation cover showed a negative correlation with groundwater depth, with a depth interval for the highest low-cover vegetation distribution of 3–6 m, and an ultimate depth threshold of 7 m. The depth interval with the highest medium-cover vegetation distribution was 3–4 m, that with the highest high-cover distribution was 2–4 m, and the ultimate depth threshold was 6 m. The depth of the oasis ranged from 3 to 6 m and the ultimate depth threshold was 7 m.

show abstract

Section: Study Areamentioning

confidence: 99%

Section: 22 Drone Datamentioning

confidence: 99%

The Impact of Groundwater Burial Depth on the Vegetation of the Dariyabui Oasis in the Central Desert

Bai,

Guo,

Wang

et al. 2023

Sustainability

View full text Add to dashboard Cite

show abstract

“…Almost 91% of Australia is occupied by vegetation [ 1 , 2 ], with forest ecosystems covering about 17% of the continent [ 3 ] concentrated in higher rainfall coastal areas. Understanding the spatial and temporal occurrence of vegetation is important to scientists and natural resource managers to support nature conservation, ecosystem science and resource inventories, sustainable land use, forestry, and agriculture [ 4 , 5 , 6 ]. Modelling the distribution of species based on bioclimatic variables using a variety of algorithms has a long history in ecology, e.g., [ 7 , 8 , 9 , 10 ].…”

Section: Introductionmentioning

confidence: 99%

U-Net Convolutional Neural Network for Mapping Natural Vegetation and Forest Types from Landsat Imagery in Southeastern Australia

Boston,

Van Dijk,

Thackway

2024

J. Imaging

View full text Add to dashboard Cite

Accurate and comparable annual mapping is critical to understanding changing vegetation distribution and informing land use planning and management. A U-Net convolutional neural network (CNN) model was used to map natural vegetation and forest types based on annual Landsat geomedian reflectance composite images for a 500 km × 500 km study area in southeastern Australia. The CNN was developed using 2018 imagery. Label data were a ten-class natural vegetation and forest classification (i.e., Acacia, Callitris, Casuarina, Eucalyptus, Grassland, Mangrove, Melaleuca, Plantation, Rainforest and Non-Forest) derived by combining current best-available regional-scale maps of Australian forest types, natural vegetation and land use. The best CNN generated using six Landsat geomedian bands as input produced better results than a pixel-based random forest algorithm, with higher overall accuracy (OA) and weighted mean F1 score for all vegetation classes (93 vs. 87% in both cases) and a higher Kappa score (86 vs. 74%). The trained CNN was used to generate annual vegetation maps for 2000–2019 and evaluated for an independent test area of 100 km × 100 km using statistics describing accuracy regarding the label data and temporal stability. Seventy-six percent of pixels did not change over the 20 years (2000–2019), and year-on-year results were highly correlated (94–97% OA). The accuracy of the CNN model was further verified for the study area using 3456 independent vegetation survey plots where the species of interest had ≥ 50% crown cover. The CNN showed an 81% OA compared with the plot data. The model accuracy was also higher than the label data (76%), which suggests that imperfect training data may not be a major obstacle to CNN-based mapping. Applying the CNN to other regions would help to test the spatial transferability of these techniques and whether they can support the automated production of accurate and comparable annual maps of natural vegetation and forest types required for national reporting.

show abstract

Deep learning approaches for estimating forest vegetation cover and exploring influential ecosystem factors

Habeeb,

Mustafa

2024

Earth Sci Inform

View full text Add to dashboard Cite

Spatial-temporal mapping of forest vegetation cover changes along highways in Brunei using deep learning techniques and Sentinel-2 images

Cited by 12 publications

References 45 publications

The Impact of Groundwater Burial Depth on the Vegetation of the Dariyabui Oasis in the Central Desert

The Impact of Groundwater Burial Depth on the Vegetation of the Dariyabui Oasis in the Central Desert

U-Net Convolutional Neural Network for Mapping Natural Vegetation and Forest Types from Landsat Imagery in Southeastern Australia

Deep learning approaches for estimating forest vegetation cover and exploring influential ecosystem factors

Contact Info

Product

Resources

About