Preface: Remote Sensing of Biodiversity

Abstract

Since the 1992 Earth Summit in Rio de Janeiro, the importance of biological diversity in supporting and maintaining ecosystem functions and processes has become increasingly understood [1]. Biodiversity "connects the web of life," that is, biodiversity represents the diversity of species in an ecosystem, landscape, region and globe. It is their combined interactions, with each other and their environment that alters biogeochemical cycles and the climate system. In recent years, biodiversity has come to broadly include diversity in terms of taxonomic, systematic and genetic attributes, morphological and structural attributes, and their ecological and functional traits. Human actions are driving climate change and land use changes throughout the globe, causing major losses of biodiversity [2,3]. These actions impact the climate, influencing magnitude and the timing of disturbance events, e.g., drought and wildfires, create pollution and contamination in water, air, and soil, and cause many other impacts that accelerate species losses, altering ecosystem functions and their services. The loss of biological diversity impacts ecological processes at scales comparable to other drivers of global environmental change [4] and likely interacts synergistically with climate change [5]. Biodiversity is recognized as a key factor in the maintenance of healthy ecosystems and for the sustainability of conservation efforts. Losses of biodiversity reduce the stability and resilience of ecosystems through the loss of functional traits associated with resource capture and decomposition [1]. The rapid pace of global change requires increased knowledge about species composition, numbers of species, and the states of health and the conditions for global ecosystems in order to respond effectively. Tittensor et al. (2014) [6] show that many of the 20 Aichi indicator targets from the Convention on Biological Diversity are unlikely to be met by 2020. The development of large plant trait databases like TRY [7,8] that have data on thousands of vascular plant species (46,085 species) still remain significantly under-sampled, particularly outside the northern temperate zone [9]. Remote sensing provides the only feasible way to measure and monitor biodiversity changes at the scales necessary. Nonetheless, until recently, there has been little success in monitoring ecologically meaningful aspects of diversity (e.g., alpha, beta, and gamma diversity). Today, there are an increasing number of remote sensing satellites and aircraft instruments that can provide a wide range of observational capability, in terms of spatial, temporal, and spectral resolutions, especially when combined with "big data" computational capacity and in situ monitoring systems. Similarly, significant progress in image processing algorithms has increased the potential for the successful characterization of biodiversity at various scales.We are pleased to present this special issue of 18 state-of-the-science papers [10-27] covering a wide range biological diversity issues ass...

A study on the impact of the spatial and spectral resolution on plant species richness in Mediterranean regions using optical remote sensing data

“…The optimal number of spectral species was estimated per RS data by testing the range of values ∈ [3,30] and selecting the one that minimizes the Davies-Bouldin Index. Among the considered images, the AVIRIS-NG image exhibited the highest number of spectral species (19), followed by the synthetic 16m hyperspectral image (18), synthetic 30m hyperspectral image (16), synthetic 4m multispectral image (9) and the Landsat-8 multispectral image (8). As expected, fewer spectral species can be identified when considering the Landsat-8 image due to its lower spatial and spectral resolution compared to the other data.…”

“…Indeed, RS data can continuously provide information at a global scale, with shorter revisit times and lower costs compared to standard field survey. 8 Different RS-based approaches have been developed to estimate Plant Species Richness (PSR), 9,10 which is typically used as a proxy for biodiversity due to its correlation with ecosystem functioning. 11 Among them, the Spectral Diversity Hypothesis (SDH) approach has proven to be very effective in estimating PSR using high spatial resolution airborne hyperspectral and multiespectral data.…”

“…On the other hand, freely obtainable multispectral data such as Sentinel-2, Landsat-8, and Moderate Resolution Imaging Spectroradiometer (MODIS), provide continuous acquisition of satellite images at high temporal frequency but at lower spatial and spectral resolution, making them less suitable for species discrimination than hyperspectral data. 8 In this context, there is an urgent need to evaluate and compare the limitations and advantages of optical RS data with different spatial and spectral resolutions in order to effectively assess biodiversity. Despite some studies that have compared different optical sensors for estimating PSR, 18 little has been done to disentangle the impact of spatial and spectral resolutions on biodiversity assessment using the SDH approach.…”

Vegetation and inundation characteristics of waterbird breeding sites in the Murray–Darling Basin, Australia

“…Relying solely on static data layers is not sufficient. Although satellite imagery for environmental monitoring is well-established (Ustin 2004;Pickens and King 2014), many of the Australian vegetation datasets lack consistent wetland vegetation mapping at a scale suitable for a Basin-wide assessment. The vegetation layers currently available, NVIS and ANAE, are created from collated vegetation maps with varying mapping methods, resolution and quality.…”

Sentinel-2 time series analysis for monitoring multi-taxon biodiversity in mountain beech forests

“…Indeed, coarse spatial and spectral resolution imagery has been used for a long time, hindering local-scale interpretation of biodiversity (Anderson, 2018). However, with increasingly detailed information, continued advances in satellite or airborne sensors provide objective, time-efficient, and reliable proxies for biological diversity (Mura et al, 2015;Hauser et al, 2021) at both local and large scales (Ustin, 2016;Udali et al, 2021). These approaches can be used for wall-to-wall mapping of biodiversity (Mura et al, 2016;Mao et al, 2018) and provide essential guidance for monitoring biodiversity in forest ecosystems, supporting forest planning and adaptive management (Groves, 2003;Hoekstra et al, 2005).…”