Science Team

Science Team Network

Explore the network below to learn more about the BioSCape Team. Use the "+" and "-" to zoom in and drag to pan. Then click on the projects and people to learn more. See each project's geographic focus here.

Science Team Leadership

Adam M. Wilson

PI & Terrestrial Science LeadUniversity at Buffalo

Erin Hestir

Marine Science LeadUC Merced

Jasper Slingsby

South African Lead University of Cape Town

Anabelle Cardoso

Science Team ManagerUniversity at BuffaloUniversity of Cape Town

Science Team

Impacts of Invasive Alien Species on Biodiversity and Ecosystem Functioning

Peter Adler/Utah State University

The Greater Cape Floristic Region (GCFR) is an ideal location to study the impacts of global change on biodiversity and ecosystem functioning. As a global biodiversity hotspot, the GCFR is under severe pressure from biological invasions and other global changes, and a large human population relies on the ecosystem services it provides. Within this unique context, the BioSCape project offers an opportunity to demonstrate how advances in remote sensing make it possible to not only map critical ecological variables, including biodiversity, ecosystem function, and biological invasions, but also to resolve long standing debates about the relationships among them. Our proposed research will work towards these goals by addressing two objectives. First, we will utilize finer spatial and spectral resolution hyperspectral imagery and data fusion to generate improved maps of a) alien tree invasions (based on AVIRIS-NG, LVIS and ALOS), b) structural and spectral diversity (based on LVIS and AVIRIS-NG) at both local (alpha) and landscape (beta) scales, and c) four important ecosystem functions: primary production and the temporal stability of primary production (based on MODIS and FluxSat), water-use efficiency (based on the ECOSTRESS algorithm and using AVIRIS-NG and HyTES), and fuel loading (based on LVIS). Second, using the maps produced under the first objective, we will test the following ecological hypotheses:

  1. Hypothesis 1: Higher structural and spectral alpha-diversity decrease invasibility, after accounting for confounding effects of landscape-scale covariates.

  2. Hypothesis 2: As alien tree invasions progress, they increase structural and spectral alpha-diversity, but decrease beta-diversity.

  3. Hypothesis 3: Ecosystem functioning (net primary productivity and its temporal stability, water use efficiency, and fuel load) depends on both invasion status (H3A) and biodiversity (H3B).

These hypotheses have traditionally been tested using statistics developed to analyze carefully controlled, small-scale field experiments where the effects of one or a few independent variables are isolated. These techniques are not appropriate for disentangling complex interactions among variables measured across large spatial extents. Fortunately, we can address this challenge by adapting statistical approaches for causal inference from econometrics and public health. The novelty of our proposed research, and contribution to the BioSCape Science Team, is the combination of remotely-sensed data with sophisticated statistical approaches to causal inference and hypothesis testing. Our team includes biodiversity scientists from Utah State University and the University of Colorado, and remote sensing experts from the University of Colorado and the NASA Goddard Space Flight Center. The team also includes South African remote sensing specialists, botanists and ecologists with extensive experience working in the GCFR. Our proposed work will advance a general understanding of the interactions among invasions, biodiversity, and ecosystem function and address critical, regional conservation and management challenges.

Plant Community Assembly and Trait Evolution in the South African Greater Cape Floristic Region

Jeannine Cavender-Bares/University of Minnesota

The Greater Cape Floristic Region (GCFR) is a hotspot of plant biodiversity with high endemism. It includes southern Hemisphere lineages known to exhibit phylogenetically distinct traits with unique evolutionary histories. The spectral properties of these species remain understudied and the evolutionary processes that explain them are largely unknown. Plant communities are hypothesized to assemble along topographic and other environmental gradients and as a consequence of dispersal limitation. Deciphering the processes of community assembly can be advanced by examining long-term evolutionary processes and environmental filtering from trait, spectral and phylogenetic information coupled with topographic and environmental data. We plan a field campaign to collect plant traits and spectral data (400 nm to 2500 nm) at the leaf level from species within plots to enhance previously published collections. These data will provide plant functional trait and spectral information that can be compared to AVIRIS NG data and used to help validate modeled functional variation from airborne data. The campaign also serves as a reconnaissance mission to understand the processes of community assembly, evolutionary ecology and biogeographic history of the GCFR to help inform the NASA BioSCAPE mission. It will also serve to connect NASA BioSCAPE and the South African research community with the NSF-funded Biology Integration Institute ASCEND (Advancing Spectral biology in Changing ENvironments to Understand Diversity).

Intrinsic Dimensionality and Data Fusion to Monitor Cape Biodiversity

Kerry Cawse-Nicholson/Jet Propulsion Laboratory

An international team of biodiversity and remote sensing experts, with extensive experience in the Cape Floristic Region (CFR), will tackle two of the BioSCape main objectives: 1) to understand the distribution and abundance of biodiversity, and 2) understand the role of biodiversity in ecosystem function. This work will address both of those goals by producing maps of taxonomic alpha, beta, and gamma diversity over the CFR using intrinsic dimensionality (derived from AVIRIS-NG reflectance and HyTES emissivity) and the generalized dissimilarity model, and relating the spatial patterns of diversity to an aspect of ecosystem function using evapotranspiration (from HyTES land surface temperature and ECOSTRESS) as a metric. This work will evaluate spatial diversity patterns by focusing on regions recovering from drought and/or fire. This work is highly relevant to the goals outlined in the BioSCape solicitation, and is also important for the NASA Biological Diversity Program, which focuses on understanding the composition of life on Earth by documenting and identifying factors that determine the distribution, abundance, physiology of organisms on Earth. The approach proposed here does not require training data, and is able to detect sub-pixel vegetation, which will also be important in the design of upcoming missions such as SBG, which aims to understand the structure, function, and biodiversity of Earth's ecosystems, and how and why are they changing in time and space. In the 2017 Decadal Survey, biodiversity was identified as a Most Important" science question, but current techniques are not mature enough for global applicability. Our international team combines interdisciplinary science with local expertise on the Fynbos: PI Cawse-Nicholson pioneered the Random Matrix Theory technique for calculating the intrinsic spectral dimensionality or number of sub-pixel components in a hyperspectral image, and the developer of the Generalized Dissimilarity Model as an unfunded collaborator to provide guidance and oversight in algorithm implementation and interpretation. The PI is also the deputy Science Lead for ECOSTRESS, and routinely produces evapotranspiration products that are publicly available. This team also contains a large South African contingent almost all former colleagues of the PI, with a good working relationship. The South African scientists have proposed a complimentary effort to the South African NEOFrontiers solicitation, which contains a significant fieldwork component in the CFR. If successful, they will be active participants in the computation and evaluation of biodiversity of the CFR. They will provide invaluable incountry assistance for logistics and provide insight into interpretation given their extensive expertise in the area.

BioSoundSCape: Connecting Acoustics and Remote Sensing to Study Habitat-Animal Diversity Across Environmental Gradients

Matthew Clark/Sonoma State University

The planet is experiencing a rapid decline in biodiversity from increasing anthropogenic pressure and the onset of climate change. Measuring biodiversity loss at relevant spatial and temporal scales needed for effective conservation policies is challenging. New costefficient and scalable approaches are needed to better track and understand changes in biodiversity. In the biodiversity hotspot Greater Cape Floristic Region (GCFR) of South Africa, we propose to measure ground-based animal diversity using low-cost autonomous recording units (ARUs) and scale these measurements using remotely-sensed indicators of habitat variation. Our approach combines acoustic diversity with plant spectral (i.e., chemical) and structural diversity to address the central questions: 1) What is the relationship among measures of acoustic, spectral and structural diversity and how do those relationships change across spatial scales and vegetation types?; and, 2) How does anthropogenic and natural disturbance affect acoustic diversity and habitat quality? ARUs collect large amounts of sound data at specific locations for days or weeks, are easy to use, and can be rapidly deployed to survey many locations with relatively low costs. The growing field of bioacoustics has made great progress in estimating animal diversity from acoustic indices and species detection with machine learning, yet techniques can be time consuming and often lack transferability across ecosystems. In contrast to previous techniques, we propose a generalizable and species agnostic approach to estimate animal diversity directly from acoustic diversity, giving it the advantage of being readily transferable to other ecosystems. During two field campaigns, we will deploy ARUs in 1,200 locations throughout the GCFR to record sounds (the soundscape"). At these locations, we will also quantify bird and frog species richness using traditional in situ observation techniques. Selected components of the soundscapes are processed to measure acoustic diversity, and we test the hypothesis that bird and frog species richness is positively related to acoustic diversity. Next, coincident imaging spectroscopy and LiDAR data from the NASA BioSCape airborne campaign is used to characterize plant spectral and structural habitat diversity, respectively. Our approach then integrates measures of acoustic, spectral, and structural diversity to investigate the strength and shape of the animal-habitat diversity relationship across vegetation types and gradients of natural and anthropogenic disturbance. We will study the change of the relationship across gradients of anthropogenic (distance to main roads) and fire (time since last fire, number of times burned) disturbance. Fire is an integral part of maintaining GCFR biodiversity at a regional scale, and thus we plan to analyze how soundscapes vary with fire history.

Our proposed research intends to address two of the general objectives of the BioScape solicitation. First, we will research the distribution and abundance of biodiversity in the GCFR at multiple spatial scales. Second, we will evaluate the response of biodiversity to global change by studying how natural processes (changing fire regime) and human pressure affect the animal-habitat relationship at a regional scale. This research is critical to understand how and why the diversity of life on Earth is changing, and will help us to scale biodiversity surveys from local to global scales using airborne and spaceborne sensors.

Integrating Remote Sensing and Biodiversity Observations to Map and Monitor Plant Taxonomic, Phylogenetic, and Functional Beta-Diversity in the Greater Cape Floristic Region

Matthew Fitzpatrick/University of Maryland

Scientists have long anticipated the potential for remote sensing to transform our ability to map and monitor global biodiversity. We are now entering an exciting period where technology and data have matured to the point that remote sensing of biodiversity at fine spatial and temporal resolution for most of the Earth's surface appears within reach. Given ongoing biodiversity losses and threats from global change, the alignment of these advances could not have come at a more crucial time. Yet substantial challenges remain - from questions regarding relationships between remote sensing observations and different components of biodiversity, to technical hurdles in the statistical integration of biodiversity observations with high-dimensional remote sensing data to extrapolate patterns across unsurveyed regions.

The goal of this project is to develop, evaluate, and demonstrate the applicability of a flexible biodiversity mapping pipeline that produces standard outputs to populate community composition essential biodiversity variables (EBVs). For this purpose, we will harness two community-level modeling algorithms - Gradient Forest and Generalized Dissimilarity Modeling - both of which have shown great promise for estimating turnover in community composition directly from analysis and modeling of high-dimensional remote sensing data and in-situ biodiversity observations. We will systematically evaluate these methods in our spatial modeling pipeline to quantify and map plant biodiversity in the Greater Cape Floristic Region (GCFR) of southern Africa. Our mapping pipeline combines several innovations to: (1) accommodate high dimensional optical and thermal imaging spectroscopy, multispectral, and LiDAR remote sensing observations, (2) perform nonlinear rescaling of these data such that they best represent community-level biodiversity and (3) predict taxonomic, phylogenetic, and functional beta-diversity EBVs. We also will test hypotheses regarding the relationships between remote sensing information and different dimensions of biodiversity. To demonstrate the applicability of our basic science objectives, we will work with partners in South Africa to map woody plant invasions and groundwater-dependent vegetation and assess outcomes of different management actions, including fire versus clearing for restoration.

We will collect new in-situ biodiversity data in the form of plot-level plant survey data and DNA for plant phylogenetics, which we will associate with plant functional trait data from existing sources.We will use data from AVIRIS, HyTES, and LVIS to quantify

vegetation attributes and Landsat/Sentinel timeseries to measure phenological signals indicative of groundwater-dependent vegetation. We will use GF and GDM in tandem to integrate these data and predict continuous spatial patterns of community composition. We will comprehensively evaluate our mapping pipeline using multiple approaches and will quantify, document, and report errors and uncertainties in our analyses.

A critical research need is improving understanding of our ability to quantify biodiversity using different remote sensing platforms and how this varies when considering taxonomic, phylogenetic, or functional dimensions. Beta-diversity represents an especially critical component of the GCFR's biodiversity, which has some of the highest levels of species turnover globally. In addition, despite calls for the development of frameworks to produce EBVs, no such systems have been fully implemented. By filling these needs, our project will address several of BioSCapes primary research objectives as well as several priorities from the Decadal Survey and NASA's Earth Science Strategy. In addition, this project will fill key data and research needs that we have identified with our team of South African collaborators.

Cyanobacteria and Surface Aquatic Vegetation of the Cape Freshwater Systems (CyanoSCape): A Hyperspectral Data Campaign and Analysis

Liane Guild/Ames Research Center

In Southern Africa, the impacts of anthropogenic activities on biodiversity and ecosystem services are exacerbated by the climate crisis. Rapid land use change and the lack of emphasis on environmentally sustainable agricultural practices has hindered hydrological processes and compromised riverine and aquatic ecosystems. This poses obvious risks to natural/indigenous aquatic biodiversity and long-term ecosystem sustainability.

Phytoplankton serve as the foundation of the freshwater food web with zooplankton as consumers, which feed fish, invertebrates, and so on up the food chain that comprises the biodiversity of the freshwater system that serves as habitat for biodiversity as well.

The diversity of phytoplankton includes photosynthesizing bacteria (cyanobacteria), plant-like diatoms, dinoflagellates, green algae, and coccolithophores. Eutrophication and toxic cyanobacteria blooms (cyanoHABs) in the inland waters of the Greater Cape Floristic Region (GCFR) incur significant effects on the biodiversity of the overall phytoplankton assemblage and provide a favorable environment for the overgrowth of floating aquatic vegetation (FAV), which is often invasive and associated with reduced aquatic biodiversity.

The algal biodiversity of the GCFR's freshwater systems is not well characterized. Hyperspectral optical observations are expected to facilitate the improvement of current phytoplankton functional type retrievals significantly, as the sensitivity is sufficient that the distinctive, fine spectral features of different phytoplankton groups can be detected. This will enable testing emerging algorithms and inform the development of new algorithms for use with upcoming hyperspectral satellite missions in this decade.

Innovations in optical sensor sensitivity and next generation machine learning capabilities considerably enhance the potential for accurate and rapid detection of phytoplankton, namely the presence, extent, and diversity of cyanobacteria present in cyanoHABs and additionally, invasive FAV. Upcoming hyperspectral satellite missions such as NASA's Surface Biology and Geology (SBG), Plankton, Aerosol, Cloud, ocean Ecosystem (PACE), and the European Space Agency's Copernicus Hyperspectral Imaging Mission (CHIME) will provide imagery with unprecedented spectral and spatial resolution which will further enable the discovery of linkages between the seasonality and dynamics of HABs and FAV.

The overarching goal of this project is to utilize hyperspectral data, with recently developed and next-generation algorithms, to determine the biodiversity of freshwater systems phytoplankton assemblage with emphasis on genus level distinction, as well as monitor the prevalence and diversity of FAV.

The proposed objectives are: 1. Characterize phytoplankton community composition of example freshwater systems of the GCFR through aligned field spectroscopy, water sample collection, and subsequent microscopy and HPLC analysis; 2. Collect field spectroradiometer hyperspectral data and surface bio-optical data coincident with airborne hyperspectral imagery for vicarious calibration and assessment of radiometric integrity of derived atmospherically corrected surface reflectance over productive waters; 3. Apply and test the capability of published and next-generation algorithms for hyperspectral delineation of the phytoplankton assemblage (biodiversity) using field and airborne measured surface reflectance and assess product uncertainty introduced through the processing of airborne data.; and 4. Discriminate FAV biodiversity to the highest taxonomic level possible using combined airborne hyperspectral and field spectroscopy (FAV spectral library) data. Expected results include the ability to distinguish phytoplankton assemblage and improved detection of the biodiversity of cyanobacteria genera and FAV as well as this variability in freshwater sites.

Biodiversity Across Scales: Mapping Taxonomic, Phylogenetic, and Functional Diversity with eDNA, Field Surveys, and Remote Sensing Data

Erica Stavros/University of Colorado Boulder

Biodiversity sustains life on Earth and is rapidly changing under the sixth mass extinction. While conservation focuses on species-specific management, there is a need for bioindicators to study, monitor, and assess our success at curbing biodiversity loss globally. Remote sensing can make consistent measurements globally, but work is needed to relate it to the scales of biodiversity that are managed (taxon and phylogenies). Imaging spectroscopy and thermal imaging can map key ecosystem functions that relate to species habitat. As such, the 2017-2027 Earth Science Decadal Survey recognized mapping biodiversity as a very important" observational need using imaging spectroscopy and thermal imaging. Thus, we critically need an objective method for calibrating and validating our algorithms from one location to another.

Current research has correlated remote sensing to field survey observations of keystone species as indicators representing broader diversity across kingdoms. These observations, however, are limited in the taxonomic breadth they can efficiently capture, are labor intensive, and have certain observer biases. In contrast, environmental DNA (eDNA) can be collected by citizen scientists and provides a unified bioinventory across organismal scales (bacteria, fungi, plants, invertebrates and vertebrates) that can explain point-to-watershed scale biodiversity patterns. The challenge with eDNA is that it is still a relatively new methodology and little is known about how transient the signal is and how it correlates with observations from above (i.e., remote sensing and traditional field metrics).

Because NASA will be launching a global imaging spectrometer and thermal imager as part of the Surface Biology and Geology (SBG) mission (expected launch in the late 2020s), there is a pressing need to understand how relatively affordable, and consistent observations of biodiversity like that of eDNA, relate to such remote sensing measurements. Specifically, there is a need to understand how taxonomic, phylogenetic, and functional diversity relate (BioSCape Objective 1b). Our cross-disciplinary, international team will collect eDNA, invertebrate, and vegetation surveys in the dry season in conjunction with NASA AVIRIS-ng and HyTES (precursors to SBG) and after rain (no flights). We ask: how do eDNA, traditional field, and remote sensing observations of biodiversity relate across spatial scales within the Berg and Eerste River watersheds? We test three hypotheses on: 1) the relationships between phylogenetic, taxonomic, and functional biodiversity, 2) how biodiversity organizes in a watershed, and 3) how signals relate to hydrometeorological processes. We will explore relationships between phylogenetic (eDNA), taxonomic (eDNA, vegetation), and functional (AVIRIS-ng and HyTES) biodiversity and create maps. We will apply unsupervised classification and correspondence analysis with hydrologic units (e.g. drainages) to assess scales of organization. Finally, we will conduct hysteresis analysis using stream gauge data to understand the temporal lags based on water surface processes (e.g., runoff vs. leaching). We will support future investigations and broader participation by delivering an open-source software package creating functional diversity metrics from trait data, building on previously published algorithms (Hytools and EcoSML), that feature our analyses as examples in user tutorials. Our women-led, citizen science approach fosters an inclusive international team by creating a pathway for mentorship to the NASA SERVIR Women's Global Develop Partnership, which can increase the success of female students who see themselves in leadership roles. Outcomes will have significant implications for global biodiversity mapping by testing applicability of joint eDNA-remote sensing while also advancing our understanding of the organizational units of biodiversity across scales.

CapeTraits: Patterns of Functional Trait Variation and Diversity Across the Greater Cape Floristic Region and Comparison with Other Mediterranean Ecosystems

Philip Townsend/University of Wisconsin-Madison

Mediterranean ecosystems globally exhibit characteristic flora, vegetation structure and function resulting from their mild, wet winters and warm, dry summers. Among Mediterranean ecosystems, the Greater Cape Floristic Region (GCFR) of South Africa contains an unusually high level of endemism and some of the most diverse vegetation assemblages in the world due to the region's long-term geologic and climatic stability, nutrient-poor soils and complex geography. Here, we ask whether the phylogenetically diverse and distinct flora of the GCFR exhibits a commensurate level of functional diversity-or functional distinctiveness-across scales within the region or in comparison to other Mediterranean areas. Theory predicts that vegetation functional traits (and trait variation) would be comparable to other Mediterranean regions due to convergent trait evolution and habitat selection by lineages that have evolved in and adapted to similar climates. Yet the Cape flora may exhibit distinct trait variation as a consequence of its unique biogeographic and evolutionary history.

We propose to:

1. Quantify functional trait composition and diversity and model their environmental drivers and the macroevolutionary processes that underlie them in the GSFC using NASA airborne AVIRIS-NG imagery, leaf-level spectra and phylogenetic information; and

2. Test the extent to which functional trait composition and diversity in the Cape mirrors that in other Mediterranean ecosystems with similar airborne imagery.

We will implement remote sensing approaches to characterize the floristic functional trait composition and diversity in the GCFR at multiple spatial scales, from which we will model drivers of functional trait variation resulting from evolutionary and environmental filtering processes in relation to climate and environment. Our analyses will be conducted first within and across ecosystems in the GCFR, and subsequently in comparison to other Mediterranean regions. The proposed work makes use of published and new plot, species and trait data from our South African Co-I. It also leverages extensive existing workflows for trait mapping from airborne imaging spectroscopy in California, NEON and elsewhere. These complementary efforts will allow us to readily apply image corrections, implement trait models and distribute to the community all resulting data, including trait maps for >20 foliar traits related to photosynthetic metabolism (e.g., nitrogen, pigments, LMA), defense and stress (phenolics), decomposition (lignin) and resource allocation (nonstructural carbohydrates). Existing trait maps from California and NEON provide a foundation for the proposed comparative analyses. To address our objectives, we will test the extent to which remote sensing data-and imaging spectroscopy data in particular-capture both the differences in phylogenetic composition within the GCFR and between Mediterranean floras. Likewise, we will evaluate whether similarities in trait composition over large spatial extents can be explained by similar bioclimatic drivers of macroevolution and community assembly, as has long been hypothesized but never before tested at this scale. Maps of functional diversity (richness, divergence, evenness)

will be derived from probability density functions in the trait maps to assess variations in multivariate trait syndromes at different scales within the GCFR and between the GCFR and other Mediterranean systems, as well as to test whether the drivers of trait variation differ across these scales. This work will fill a significant gap in our knowledge of global functional trait variation by providing spatially explicit data of trait variation and functional diversity in the distinct Greater Cape Floristic Region. It will also inform forthcoming satellite missions such as Surface Biology and Geology (SBG) by testing the extent to which data-driven trait mapping approaches are transferable across physiognomically similar biomes.

RadSCape: Radiative Transfer Simulation and Validation of the Dynamic Structural and Spectral Properties of the Vegetation of the Cape

Jan van Aardt/Rochester Institute of Technology

The hyper-diverse Greater Cape Floristic Region (GCFR) encompasses two Global Biodiversity Hotspots, while also being a Global Extinction Hotspot, threatened by habitat loss and fragmentation, altered fire regimes, invasive species, and climate change, among others. Managing and mitigating these threats requires regularly-updated, spatially-explicit information for the entire region, which is currently only feasible using satellite remote sensing at relatively coarse scales. However, detecting the signal of change over and above the natural variability of ecosystems in the region, in both spectral and structural terms, remains a distinct challenge, especially given that the GCFR is dominated by open shrublands. This results in a high relative contribution of soils, stems, and other surfaces to observed reflectance, which is further exacerbated by the high variability of vegetation in space and time due to seasonality, natural disturbances such as fire or drought, and long-term post-disturbance recovery trajectories. Finally, and critical to this proposal, these interactions further increase in complexity due to the exceptionally high ecosystem diversity. This already-inherent diversity increases in complexity when one considers that the chemistry and physiology of foliage, and its interaction with structure, are key drivers of ecosystem functions. This has led to an interest in remote sensing of leaf traits as a means of leaf-to-ecosystem scaling. Although numerous studies - mostly forest focused - have demonstrated that leaf traits such as foliar nitrogen concentration (%N), leaf mass per unit area (LMA), and associated patterns of vegetation growth can be estimated from remotely sensed data, the mechanisms responsible remain unclear because, in whole plant canopies, leaf-level reflectance properties are confounded by the influence of structural variables at the leaf, stem, and whole-crown scales. This produces tremendous complexity in the number, size, orientation, and spatial arrangement of reflecting leaf surfaces and canopy gaps, and also affects the properties of light as it penetrates the canopy. All of these influences pose a parsimonious solution to the challenge of simulating land surface reflectance spectra in a region with >13 000 plant species and a similarly diverse abiotic environment. We thus propose to resolve this challenge by using a combination of in situ fynbos trait measurements and first-principles, physics-based radiative transfer modeling in a biophysically-robust simulation environment, validated with imaging spectroscopy and LiDAR remote sensing data from

NASA's BioSCape airborne campaigns. We will collect in situ spectroradiometer and terrestrial LiDAR data, to first develop and then annually refine, accurate virtual GCFR plot-level scenes, which will be used to i) assess the linkages between 3D (structural) and spectral diversity, ii) develop a mechanistic link between structure/spectra-to-traits, and iii) track biodiversity as a function of post-fire recovery, all of which will be iv) validated against the planned NASA airborne campaigns. An additional benefit of this work relates to a definitive assessment of what spatial and structural scales future air-/spaceborne missions should target in order to resolve the structural, species, and trait diversity in such complex ecosystems. The team boasts established imaging scientists and South African ecology collaborators, with both spectroscopy and LiDAR experience, as well as a vetted simulation tool (DIRSIG). We ultimately contend that achieving these project goals could significantly improve our understanding of interactions of photons with fynbos biophysical traits and inform innovative uses of remote sensing data for similar ecosystems at all scales, from airborne-to-satellite levels.

Mapping of Phytoplankton Functional Types from Space in Support of Coastal Resource Management and Decision Support Activities

Jinghui Wu/Columbia University

Background:

Sheltered bays along South Africa's (SAs) long coastline are few and far in between, and therefore, of high socio-economic importance. Most are highly productive, but ecologically distinct because of the differences in water masses that influence them. Bays along the west coast are influenced by the cold Benguela Current, whereas those along the east coast are exposed to the warm Agulhas Current. Southern bays are exposed to a blend of the aforementioned currents, and environmental conditions within them vary depending on local winds and large scale atmospheric and oceanographic phenomenon. On account of their high productivity and extraordinary taxonomic biodiversity, these bays can sustain thriving fisheries, commercial shellfish farming, tourism and recreational activities that are vital to the socio-economic well-being of SA's coastal communities. There is growing recognition that the health of these bay ecosystems, and the goods and services they provide, are coming under increased anthropogenic and climate-mediated threats, witnessed in the form of extreme temperature events, intensification of upwelling and rise in the frequency and intensity of Harmful Algal Blooms (HABs). Although it is known that these changes are detrimental to phytoplankton species richness and diversity, conventional means of monitoring these changes for use by stakeholders, including resource managers has been challenging.

Specific Objectives and Methods:

Our overarching goal is to develop a hyperspectral radiometric method to map the spatial distribution of phytoplankton functional types (PFT) across environmental gradients within three ecologically distinct but socio-economically vital bays, i.e. St Helena [Namaqua Bioregion], Walker [Agulhas Bioregion] and Algoa [Agulhas Bioregion]. The proposed method will detect specific PFTs, taking advantage of the unique hyperspectral remote sensing reflectance (Rrs) signals arising from phytoplankton pigment variability.

Our method of identifying PFTs will combine in-situ optical data with bio-optical, microscopic, phytoplankton pigment and e-DNA data collections, to develop algorithms that detect PFTs in particular HABs using the narrow-band hyperspectral datasets from AVIRIS-NG and PRISM that will be flown during the BIOSCape field campaign. Field datasets will be used to first disentangle the optical complexity of bay waters by separating Rrs signatures of phytoplankton communities from other seawater constituents such as mineral particles, colored dissolved and particulate organic matter. Residual hyperspectral Rrs signals attributable to phytoplankton will be used to detect different pigments essential for discriminating specific PFTs. Our team will work closely with the BIOSCape team to plan flight lines to ensure that the data over our study sites are minimally impacted by cloud cover, aerosols, sun-glint, etc. In addition, we will extend our algorithms to multi-spectral data from MODIS-Aqua, Suomi-VIIRS, VIIRS-20 and PACE to generate satellite maps of PFTs (with associated uncertainties), to address 4 hypotheses pertaining to two theme areas: (a) Distribution and abundance of biodiversity in the Greater Cape Floristic Region of South Africa (GCFR) and (c) Feedbacks between global environmental change, and ecosystem services in the GCFR. Engagement with stakeholders will take advantage of existing partnerships between our SA team and coastal industries.

Relevance to NASA activities:

Development of the capability to monitor PFTs from space opens its application to hyperspectral data from NASAs planned PACE and GLIMR missions. All QC'd field optical and bio-optical datasets, will be disseminated for public use via NASA SEABASS and NASA-BIOSCAPE and NASA-LDEO web portals. New satellite data products resulting from this study along with associated uncertainties, will be made publicly accessible via NASA-OBPG.

Institutional Partners

South African Environmental Observation Network (SAEON)


Glenn Moncrieff

South African Environmental Observation Network

South Africa National Biodiversity Institute (SANBI)


South Africa National Parks (SANParks)