Mapping Mediterranean Forest Plant Associations and Habitats with Functional Principal Component Analysis Using Landsat 8 NDVI Time Series

Abstract

The classification of plant associations and their mapping play a key role in defining habitat biodiversity management, monitoring, and conservation strategies. In this work we present a methodological framework to map Mediterranean forest plant associations and habitats that relies on the application of the Functional Principal Component Analysis (FPCA) to the remotely sensed Normalized Difference Vegetation Index (NDVI) time series. FPCA, considering the chronological order of the data, reduced the NDVI time series data complexity and provided (as FPCA scores) the main seasonal NDVI phenological variations of the forests. We performed a supervised classification of the FPCA scores combined with topographic and lithological features of the study area to map the forest plant associations. The supervised mapping achieved an overall accuracy of 87.5%. The FPCA scores contributed to the global accuracy of the map much more than the topographic and lithological features. The results showed that (i) the main seasonal phenological variations (FPCA scores) are effective spatial predictors to obtain accurate plant associations and habitat maps; (ii) the FPCA is a suitable solution to simultaneously express the relationships between remotely sensed and ecological field data, since it allows us to integrate these two different perspectives about plant associations in a single graph. The proposed approach based on the FPCA is useful for forest habitat monitoring, as it can contribute to produce periodically detailed vegetation-based habitat maps that reflect the “current” status of vegetation and habitats, also supporting the study of plant associations.

Recognition and Characterization of Forest Plant Communities through Remote-Sensing NDVI Time Series

“…Recently, the use of remotely sensed NDVI time series combined with functional data analysis opened new ways to understand phenology, supporting phytosociologists. It is possible to integrate phytosociological data with remotely sensed data to find a relationship with syntaxonomic units of high detail that are plant associations and habitats, as in our previous work [10]. In this work, we show that plant communities are distinguished by floristic composition and phenological behavior, and NDVI time series processed with a functional approach are helpful to improve habitat mapping.…”

“…The representation of clusters on reduced phenological space ordination (forest plant communities), as shown in Figure 4c, facilitates the interpretation of forest plant community ecology and phenology (similarly to Brooks et al [61]), integrating sensor-based and field monitoring data (e.g., [10,50]). A comprehensive method that translates remote-sensing data into vegetation units that are defined a priori and based on species composition, such as phytosociological units, is desired [11].…”

“…FPCA scores are a core aspect of this study, as they could be used in subsequent analysis (e.g., correlation analysis, functional data clustering [49]). In an ordered reduced space (here, a phenological space), the functions (here, the NDVI time series) could be plotted supporting experts in ecological interpretation [10,50].…”

“…The main aim was to compare two different areas (coastal and Apennine) that share similar forest plant communities. The first subset, namely, Mount Conero, was derived from our previous study [10]. This area was previously also studied in detail from a phytosociological point of view [35,52].…”

“…With the development of satellite technology, remote vegetation detection has represented a methodological evolution of great application interest. However, the integration between phytosociology and remote sensing still remains difficult [9], so much so that the "combined" approach that involves the use of remote-sensing technologies associated with field surveys still remains a target to be satisfied [4,6,10]. The greatest difficulties encountered are to identify a correspondence between remote-sensing products and plant communities that sometimes differ from each other by a few species (diagnostic species) that are difficult to be captured by sensors [11].…”