Turning the Tide on Mapping Marginal Mangroves with Multi-Dimensional Space–Time Remote Sensing

Abstract

Mangroves are a globally important ecosystem experiencing significant anthropogenic and climate impacts. Two subtypes of mangrove are particularly vulnerable to climate-induced impacts (1): tidally submerged forests and (2) those that occur in arid and semi-arid regions. These mangroves are either susceptible to sea level rise or occur in conditions close to their physiological limits of temperature and freshwater availability. The spatial extent and impacts on these mangroves are poorly documented, because they have structural and environmental characteristics that affect their ability to be detected with remote sensing models. For example, tidally submerged mangroves occur in areas with large tidal ranges, which limits their visibility at high tide, and arid mangroves have sparse canopy cover and a shorter stature that occur in fringing and narrow stands parallel to the coastline. This study introduced the multi-dimensional space–time randomForest method (MSTRF) that increases the detectability of these mangroves and applies this on the North-west Australian coastline where both mangrove types are prevalent. MSTRF identified an optimal four-year period that produced the most accurate model (Accuracy of 80%, Kappa value 0.61). This model was able to detect an additional 32% (76,048 hectares) of mangroves that were previously undocumented in other datasets. We detected more mangrove cover using this timeseries combination of annual median composite Landsat images derived from scenes across the whole tidal cycle but also over climatic cycles such as EÑSO. The median composite images displayed less spectral differences in mangroves in the intertidal and arid zones compared to individual scenes where water was present during the tidal cycle or where the chlorophyll reflectance was low during hot and dry periods. We found that the MNDWI (Modified Normalised Water Index) and GCVI (Green Chlorophyll Vegetation Index) were the best predictors for deriving the mangrove layer using randomForest.

Species richness, extent and potential threats to mangroves of Sarangani Bay Protected Seascape, Philippines

“…Some structural and environmental constraints affect the detectability of mangroves with remote sensing models. For example, the sparse canopy and short stature of mangroves relative to other trees cause their limited visibility ( Hickey and Radford 2022 ). The mangroves in SBPS form narrow fringes and small patches of stands (Fig.…”

Spatio-temporal dynamics of mangrove extent in coastal Bekasi Regency, Indonesia

“…The impact of topography was reduced, and the spectral differences across bands were enhanced using the band ratio [15]. While GCVI was employed, several studies showed that mapping mangrove chlorophyll and GCVI had an excellent correlation [16]. This formula is expressed in equation 1-7.…”

Studying How Machine Learning Maps Mangroves in Moderate-Resolution Satellite Images

“…Spectral indices provide valuable information about the environment. In mangroves, key indices are those that offer information about green vegetation (Red and near-infrared) and water reflectance (Green and middle infrared) [16]. Integrating characteristics of water and vegetation can help differentiate between land vegetation and mangrove vegetation [20].…”

“…Integrating characteristics of water and vegetation can help differentiate between land vegetation and mangrove vegetation [20]. Hickey and Radford (2022) discovered that the Modified Normalized Water Index (MNDWI) and Green Chlorophyll Vegetation Index (GCVI) in Landsat-8 imagery are the best predictors for mangroves [16], Meanwhile, in Sentinel-2 imagery, it was discovered that MNDWI and Mangrove Discrimination Index (MDI) significantly enhance mangrove extraction performance [20]. Other relevant spectral indices for mangrove classification include the Normalized Difference Moisture Index (NDMI) [13], Normalized Difference Soil Index (NDSI) [17], and Mangrove Vegetation Index (MVI) [31].…”

“…It is calculated from the confusion matrix by comparing the values achieved in the classification process and the ground truth values (Equation 2). Models with Kappa > 0.8 have strong predictive power, models with Kappa between 0.6 and 0.8 are adequate, and models with Kappa < 0.5 lack discrimination capability [16].…”