skip to main content

Explore Research Products in the NSF-PAR

  • NSF Public Access
  • Search Results
  • Cutting out the middleman: calibrating and validating a dynamic vegetation model (ED2-PROSPECT5) using remotely sensed surface reflectance
Title: Cutting out the middleman: calibrating and validating a dynamic vegetation model (ED2-PROSPECT5) using remotely sensed surface reflectance
Abstract. Canopy radiative transfer is the primary mechanism by which models relate vegetation composition and state to the surface energy balance, which is important to light- and temperature-sensitive plant processes as well as understanding land–atmosphere feedbacks.In addition, certain parameters (e.g., specific leaf area, SLA) that have an outsized influence on vegetation model behavior can be constrained by observations of shortwave reflectance, thus reducing model predictive uncertainty.Importantly, calibrating against radiative transfer outputs allows models to directly use remote sensing reflectance products without relying on highly derived products (such as MODIS leaf area index) whose assumptions may be incompatible with the target vegetation model and whose uncertainties are usually not well quantified.Here, we created the EDR model by coupling the two-stream representation of canopy radiative transfer in the Ecosystem Demography model version 2 (ED2) with a leaf radiative transfer model (PROSPECT-5) and a simple soil reflectance model to predict full-range, high-spectral-resolution surface reflectance that is dependent on the underlying ED2 model state.We then calibrated this model against estimates of hemispherical reflectance (corrected for directional effects) from the NASA Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) and survey data from 54 temperate forest plots in the northeastern United States.The calibration significantly reduced uncertainty in model parameters related to leaf biochemistry and morphology and canopy structure for five plant functional types.Using a single common set of parameters across all sites, the calibrated model was able to accurately reproduce surface reflectance for sites with highly varied forest composition and structure.However, the calibrated model's predictions of leaf area index (LAI) were less robust, capturing only 46 % of the variability in the observations.Comparing the ED2 radiative transfer model with another two-stream soil–leaf–canopy radiative transfer model commonly used in remote sensing studies (PRO4SAIL) illustrated structural errors in the ED2 representation of direct radiation backscatter that resulted in systematic underestimation of reflectance.In addition, we also highlight that, to directly compare with a two-stream radiative transfer model like EDR, we had to perform an additional processing step to convert the directional reflectance estimates of AVIRIS to hemispherical reflectance (also known as “albedo”).In future work, we recommend that vegetation models add the capability to predict directional reflectance, to allow them to more directly assimilate a wide range of airborne and satellite reflectance products.We ultimately conclude that despite these challenges, using dynamic vegetation models to predict surface reflectance is a promising avenue for model calibration and validation using remote sensing data.  more » « less
Award ID(s):
1655095
NSF-PAR ID:
10229660
Author(s) / Creator(s):
; ; ; ;
Date Published:
Journal Name:
Geoscientific Model Development
Volume:
14
Issue:
5
ISSN:
1991-9603
Page Range / eLocation ID:
2603 to 2633
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; ; ( , Geoscientific Model Development)
    null (Ed.)
    Abstract. At the leaf level, stomata control the exchange of water and carbon across the air–leaf interface. Stomatal conductance is typically modeledempirically, based on environmental conditions at the leaf surface. Recently developed stomatal optimization models show great skills at predictingcarbon and water fluxes at both the leaf and tree levels. However, how well the optimization models perform atlarger scales has not been extensively evaluated. Furthermore, stomatal models are often used with simple single-leaf representations of canopy radiative transfer (RT), such asbig-leaf models. Nevertheless, the single-leaf canopy RT schemes do not have the capability to model optical properties of the leaves nor the entirecanopy. As a result, they are unable to directly link canopy optical properties with light distribution within the canopy to remote sensing dataobserved from afar. Here, we incorporated one optimization-based and two empirical stomatal models with a comprehensive RT model in the landcomponent of a new Earth system model within CliMA, the Climate Modelling Alliance. The model allowed us to simultaneously simulate carbon and waterfluxes as well as leaf and canopy reflectance and fluorescence spectra. We tested our model by comparing our modeled carbon and water fluxes andsolar-induced chlorophyll fluorescence (SIF) to two flux tower observations (a gymnosperm forest and an angiosperm forest) and satellite SIFretrievals, respectively. All three stomatal models quantitatively predicted the carbon and water fluxes for both forests. The optimization model,in particular, showed increased skill in predicting the water flux given the lower error (ca. 14.2 % and 21.8 % improvement for thegymnosperm and angiosperm forests, respectively) and better 1:1 comparison (slope increases from ca. 0.34 to 0.91 for the gymnosperm forest andfrom ca. 0.38 to 0.62 for the angiosperm forest). Our model also predicted the SIF yield, quantitatively reproducing seasonal cycles for bothforests. We found that using stomatal optimization with a comprehensive RT model showed high accuracy in simulating land surface processes. Theever-increasing number of regional and global datasets of terrestrial plants, such as leaf area index and chlorophyll contents, will helpparameterize the land model and improve future Earth system modeling in general. 
    more » « less
  2. ; ; ; ; ; ; ; ; ; ( , Ecology)
    Elizabeth Borer (Ed.)
    Understanding spatial and temporal variation in plant traits is needed to accurately predict how communities and ecosystems will respond to global change. The National Ecological Observatory Network’s (NEON’s) Airborne Observation Platform (AOP) provides hyperspectral images and associated data products at numerous field sites at 1 m spatial resolution, potentially allowing high-resolution trait mapping. We tested the accuracy of readily available data products of NEON’s AOP, such as Leaf Area Index (LAI), Total Biomass, Ecosystem Structure (Canopy height model [CHM]), and Canopy Nitrogen, by comparing them to spatially extensive field measurements from a mesic tallgrass prairie. Correlations with AOP data products exhibited generally weak or no relationships with corresponding field measurements. The strongest relationships were between AOP LAI and ground-measured LAI (r = 0.32) and AOP Total Biomass and ground-measured biomass (r = 0.23). We also examined how well the full reflectance spectra (380–2,500 nm), as opposed to derived products, could predict vegetation traits using partial least-squares regression (PLSR) models. Among all the eight traits examined, only Nitrogen had a validation of more than 0.25. For all vegetation traits, validation ranged from 0.08 to 0.29 and the range of the root mean square error of prediction (RMSEP) was 14–64%. Our results suggest that currently available AOP-derived data products should not be used without extensive ground-based validation. Relationships using the full reflectance spectra may be more promising, although careful consideration of field and AOP data mismatches in space and/or time, biases in field-based measurements or AOP algorithms, and model uncertainty are needed. Finally, grassland sites may be especially challenging for airborne spectroscopy because of their high species diversity within a small area, mixed functional types of plant communities, and heterogeneous mosaics of disturbance and resource availability. Remote sensing observations are one of the most promising approaches to understanding ecological patterns across space and time. But the opportunity to engage a diverse community of NEON data users will depend on establishing rigorous links with in-situ field measurements across a diversity of sites. 
    more » « less
  3. ; ; ; ; ; ; ; ; ; ( , Ecology)
    Abstract

    Understanding spatial and temporal variation in plant traits is needed to accurately predict how communities and ecosystems will respond to global change. The National Ecological Observatory Network’s (NEON’s) Airborne Observation Platform (AOP) provides hyperspectral images and associated data products at numerous field sites at 1 m spatial resolution, potentially allowing high‐resolution trait mapping. We tested the accuracy of readily available data products of NEON’s AOP, such as Leaf Area Index (LAI), Total Biomass, Ecosystem Structure (Canopy height model [CHM]), and Canopy Nitrogen, by comparing them to spatially extensive field measurements from a mesic tallgrass prairie. Correlations with AOP data products exhibited generally weak or no relationships with corresponding field measurements. The strongest relationships were between AOP LAI and ground‐measured LAI (r = 0.32) and AOP Total Biomass and ground‐measured biomass (r = 0.23). We also examined how well the full reflectance spectra (380–2,500 nm), as opposed to derived products, could predict vegetation traits using partial least‐squares regression (PLSR) models. Among all the eight traits examined, only Nitrogen had a validation of more than 0.25. For all vegetation traits, validation ranged from 0.08 to 0.29 and the range of the root mean square error of prediction (RMSEP) was 14–64%. Our results suggest that currently available AOP‐derived data products should not be used without extensive ground‐based validation. Relationships using the full reflectance spectra may be more promising, although careful consideration of field and AOP data mismatches in space and/or time, biases in field‐based measurements or AOP algorithms, and model uncertainty are needed. Finally, grassland sites may be especially challenging for airborne spectroscopy because of their high species diversity within a small area, mixed functional types of plant communities, and heterogeneous mosaics of disturbance and resource availability. Remote sensing observations are one of the most promising approaches to understanding ecological patterns across space and time. But the opportunity to engage a diverse community of NEON data users will depend on establishing rigorous links with in‐situ field measurements across a diversity of sites.

     
    more » « less
  4. ; ; ; ; ; ; ; ; ; ; et al ( , New Phytologist)
    Summary

    Evergreen conifer forests are the most prevalent land cover type in North America. Seasonal changes in the color of evergreen forest canopies have been documented with near‐surface remote sensing, but the physiological mechanisms underlying these changes, and the implications for photosynthetic uptake, have not been fully elucidated.

    Here, we integrate on‐the‐ground phenological observations, leaf‐level physiological measurements, near surface hyperspectral remote sensing and digital camera imagery, tower‐based CO2flux measurements, and a predictive model to simulate seasonal canopy color dynamics.

    We show that seasonal changes in canopy color occur independently of new leaf production, but track changes in chlorophyll fluorescence, the photochemical reflectance index, and leaf pigmentation. We demonstrate that at winter‐dormant sites, seasonal changes in canopy color can be used to predict the onset of canopy‐level photosynthesis in spring, and its cessation in autumn. Finally, we parameterize a simple temperature‐based model to predict the seasonal cycle of canopy greenness, and we show that the model successfully simulates interannual variation in the timing of changes in canopy color.

    These results provide mechanistic insight into the factors driving seasonal changes in evergreen canopy color and provide opportunities to monitor and model seasonal variation in photosynthetic activity using color‐based vegetation indices.

     
    more » « less
  5. ; ; ; ; ; ; ; ( , Botany)
    Degree of canopy cover is linked to transpiration, carbon cycling and primary productivity of an ecosystem. In modern ecology, canopy structure is often quantified as Leaf Area Index (LAI), which is the amount of overstory leaf coverage relative to ground area. Although a key aspect of vegetation, the degree of canopy cover has proven difficult to reconstruct in deep time. One method, Reconstructed Leaf Area Index (rLAI), was developed to infer canopy structure using the relationship between non-grass leaf epidermal phytolith (plant biosilica) morphology, and leaf coverage in modern forests. This method leverages the observed correlation between epidermal phytolith size, shape (margin undulation), and light availability. When more light is available in a canopy, epidermal phytoliths tend to be smaller and less undulate, whereas less light availability is linked to larger and more undulate epidermal phytoliths. However, the calibration set used to develop this method was compiled from field sites and samples from localities in Costa Rica and it remains unclear how applicable it is to temperate North American fossil sites due to lack of data from relevant vegetation types and taxonomic differences between plant communities in the Neotropics vs. mid-latitude North America. For example, preliminary results measuring rLAI in phytolith assemblages from the Miocene of the North American Great Plains have yielded surprisingly high degrees of canopy density despite containing high relative abundances of open-habitat grasses. To test whether vegetational and taxonomic differences impact the calibration set, we constructed a new North American calibration using 24 quadrats from six sites, representing reasonable modern analogs for Miocene vegetation in eastern North America. Specifically, we sampled in Bennett Springs State Park in Lebanon, MO; Mark Twain National Forest in Rolla, MO; Tellico in Franklin, NC and Congaree National Park in Hopkins, SC. All sites include a range of canopy covers and vegetation types, from oak savannas and oak woodlands to mixed hardwood forests, pine savannas, and old growth bottomland forests. From each quadrat, we collected a soil sample and took hemispherical photos of the local canopy. From modern soil samples, biosilica was extracted in the lab, yielding phytolith assemblages which were scanned for epidermal phytoliths using a compound microscope. Recovered epidermal phytoliths size and margin undulation were measured and assemblage averages were used to predict measured LAI at each quadrat. Hemispherical photographs were processed using the software Gap Light Analyzer to obtain LAI values. We hypothesize there will be a linear relationship between actual LAI and LAI calculated from epidermal phytolith morphology, but its relationship will differ from that found in South America. Results will be used to reevaluate canopy coverage in sites within the Great Plains Miocene as well as applied to Pacific Northwest Miocene sites, both to understand changes to vegetation during global climatic events in their respective regions. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email