Variation In Canopy Reflectance Research Articles

About the link between biodiversity and spectral variation

AbstractAimThe spectral variability hypothesis (SVH) suggests a link between spectral variation and plant biodiversity. The underlying assumptions are that higher spectral variation in canopy reflectance (depending on scale) is caused by either (1) variation in habitats or linked vegetation types or plant communities with their specific optical community traits or (2) variation in the species themselves and their specific optical traits.MethodsThe SVH was examined in several empirical remote‐sensing case studies, which often report some correlation between spectral variation and biodiversity‐related variables (mostly plant species counts); however, the strength of the observed correlations varies between studies. In contrast, studies focussing on understanding the causal relationship between (plant) species counts and spectral variation remain scarce. Here, we discuss these causal relationships and support our perspectives through simulations and experimental data.ResultsWe reveal that in many situations the spectral variation caused by species or functional traits is subtle in comparison to other factors such as seasonality and physiological status. Moreover, the degree of contrast in reflectance has little to do with the number but rather with the identity of the species or communities involved. Hence, spectral variability should not be expressed based on contrast but rather based on metrics expressing manifoldness. While we describe cases where a certain link between spectral variation and plant species diversity can be expected, we believe that as a scientific hypothesis (which suggests a general validity of this assumed relationship) the SVH is flawed and requires refinement.ConclusionsTo this end we call for more research examining the drivers of spectral variation in vegetation canopies and their link to plant species diversity and biodiversity in general. Such research will allow critically assessing under which conditions spectral variation is a useful indicator for biodiversity monitoring and how it could be integrated into monitoring networks.

Radiative transfer modelling reveals why canopy reflectance follows function

Optical remote sensing is potentially highly informative to track Earth’s plant functional diversity. Yet, causal explanations of how and why plant functioning is expressed in canopy reflectance remain limited. Variation in canopy reflectance can be described by radiative transfer models (here PROSAIL) that incorporate plant traits affecting light transmission in canopies. To establish causal links between canopy reflectance and plant functioning, we investigate how two plant functional schemes, i.e. the Leaf Economic Spectrum (LES) and CSR plant strategies, are related to traits with relevance to reflectance. These traits indeed related to both functional schemes, whereas only traits describing leaf properties correlated with the LES. In contrast, traits related to canopy structure showed no correlation to the LES, but to CSR strategies, as the latter integrates both plant economics and size traits, rather than solely leaf economics. Multiple optically relevant traits featured comparable or higher correspondence to the CSR space than those traits originally used to allocate CSR scores. This evidences that plant functions and strategies are directly expressed in reflectance and entails that canopy ‘reflectance follows function’. This opens up new possibilities to understand differences in plant functioning and to harness optical remote sensing data for monitoring Earth´s functional diversity.

Differentiating plant functional types using reflectance: which traits make the difference?

AbstractAbiotic ecosystem properties together with plant species interaction create differences in structural and physiological traits among plant species. Certain plant traits cause a spatial and temporal variation in canopy reflectance that enables the differentiation of plant functional types, using earth observation data. However, it often remains unclear which traits drive the differences in reflectance between plant functional types, since the spectral regions in which electromagnetic radiation is influenced by certain plant traits are often overlapping. The present study aims to assess the relative (statistical) contributions of plant traits to the separability of plant functional groups using their reflectance. We apply the radiative transfer model PROSAIL to simulate optical canopy reflectance of 38 herbaceous plant species based on field‐measured traits such as leaf area index, leaf inclination distribution, chlorophyll content, carotenoid content, water and dry matter content. These traits of the selected grassland species were measured in an outdoor plant experiment. The 38 species differed in growth form and strategy types according to Grime′s CSR model and hence represented a broad range of plant functioning. We determined the relative (statistical) contribution of each plant trait for separating plant functional groups via reflectance. Therein we used a separation into growth forms, that is graminoids and herbs, and into CSR strategy types. Our results show that the relative contribution of plant traits to differentiate between the examined plant functional types (PFT) groups using canopy reflectance depends on the PFT scheme applied. Plant traits describing the canopy structure were more important in this regard than leaf traits. Accordingly, leaf area index (LAI) and leaf inclination showed consistently high importance for separating the examined PFT groups. This indicates that the role of canopy structure for spectrally differentiating PFT might have been underestimated.

Regional-scale drivers of forest structure and function in northwestern Amazonia.

Field studies in Amazonia have found a relationship at continental scales between soil fertility and broad trends in forest structure and function. Little is known at regional scales, however, about how discrete patterns in forest structure or functional attributes map onto underlying edaphic or geological patterns. We collected airborne LiDAR (Light Detection and Ranging) data and VSWIR (Visible to Shortwave Infrared) imaging spectroscopy measurements over 600 km2 of northwestern Amazonian lowland forests. We also established 83 inventories of plant species composition and soil properties, distributed between two widespread geological formations. Using these data, we mapped forest structure and canopy reflectance, and compared them to patterns in plant species composition, soils, and underlying geology. We found that variations in soils and species composition explained up to 70% of variation in canopy height, and corresponded to profound changes in forest vertical profiles. We further found that soils and plant species composition explained more than 90% of the variation in canopy reflectance as measured by imaging spectroscopy, indicating edaphic and compositional control of canopy chemical properties. We last found that soils explained between 30% and 70% of the variation in gap frequency in these forests, depending on the height threshold used to define gaps. Our findings indicate that a relatively small number of edaphic and compositional variables, corresponding to underlying geology, may be responsible for variations in canopy structure and chemistry over large expanses of Amazonian forest.

Seasonal variation in canopy reflectance and its application to determine the water status and water use by citrus trees in the Western Cape, South Africa

This study describes the diurnal and seasonal dynamics of the canopy reflectance, water use and water status of Midknight Valencia citrus trees under semi-arid conditions. Hyperspectral canopy reflectance data was collected on 30 trees at monthly intervals over a period of 16 months in a commercial orchard in South Africa. The mean canopy reflectance in the wavelength range 350–2500 nm followed a clear seasonal trend influenced by environmental conditions and tree phenology. Mean monthly reflectance peaked in summer (∼22%) while the lowest value (∼15%) was reached in winter with the seasonal changes in the sun's position accounting for a significant proportion of the variations. A sensitivity analysis of a Penman–Monteith transpiration model showed that water use by individual trees changed by up to 13% when the canopy reflectance was varied over the seasonal range of measured values. This suggested that the seasonal changes in tree water use influenced the seasonal trend of the canopy reflectance. Thus monitoring the canopy reflectance of citrus trees could offer information on the tree water status. To test this, sap flow data of water uptake and loss by the trees were compared with the canopy spectra. Sap flow data showed a heavy reliance by the citrus trees on the internally stored water with up to 25% of the daily total transpiration withdrawn from the trees’ internal water storage pools when soil water was limited. This depletion of internally stored water, and hence the change in tree water status, was detected using spectral indices based on the first order derivatives of the canopy reflectance centered at two and, at most, four spectral bands. We conclude that even if citrus trees are evergreen, their canopy reflectance changes significantly throughout the year with a considerable impact on tree energy balance and water use. In addition, the contribution of the internally stored water to daily transpiration is a possible indicator of drought stress for citrus trees detectable from changes in canopy reflectance and it has potential applications in irrigation scheduling using canopy level spectral information.

Using canopy reflectance models and spectral angles to assess potential of remote sensing to detect invasive weeds

One of the goals of applied remote sensing is to map locations of invasive weeds. However, differences in plant cover and leaf area index (LAI) alter canopy reflectance, making detection of a single species difficult. Variation in canopy reflectance may be simulated using the Scattering by Arbitrarily Inclined Leaves (SAIL) model. Simulated reflectances are used to calculate spectral angles to determine the separability of an invasive weed from co-occurring vegetation. Leafy spurge is a noxious invasive weed with yellow-green flower-bracts. Spectral angles from SAIL model simulations show that flowering leafy spurge may be detected when LAI is greater than 1.0 and flower-bract cover is greater than 10%. A threshold of 3.5 deg (0.061 radians) was determined to provide the best separation between leafy spurge and co-occurring vegetation. To test this prediction, the Spectral Angle Mapper was used to classify leafy spurge using AVIRIS, Landsat ETM+ and SPOT data. Classification accuracy was inversely related to simulated spectral angles from the SAIL model analyses. Using canopy reflectance models and spectral angles may help identify those invasive species that are potentially detectable by remote sensing, and may indicate the conditions where detection will be problematic based on variation of LAI, cover and other variables.

Physical influences of landscape on a large-extent ecological disturbance: the northeastern North American ice storm of 1998

The 1998 ice storm was a large-extent ecological disturbance that severely affected the eastern Adirondack forests of northern New York. Ice damage produced widespread breakage of limbs and trunks in susceptible trees. Although ice storms are common within northeastern North American forests, the magnitude and extent of the 1998 storm far exceeded damage caused by typical ice storms in the recent past. While plot and stand-scale ecological impacts of ice storms have received attention insofar as tree species vulnerability, stand age susceptibility, and microhabitat alterations, larger-extent damage patterns have not been previously evaluated. The normalized difference vegetation index (NDVI) was employed to assess forest vigor and canopy density in atmospherically corrected Landsat Thematic Mapper (TM) satellite imagery of the Adirondacks. Digital change analysis of the baseline forest condition (1990 NDVI data), and the condition encountered in a post-storm image (1998 NDVI data) was conducted. Forest damage was separated from natural variations in canopy reflectance by employing a generalized linear model that incorporated in situ measurements. A robust empirical variogram analysis revealed that locations of tree damage were significantly correlated for distances up to 300 meters. Intensely-damaged forest exhibited greater spatial dependence, but over a smaller distance. Canopy damage was not greater proximate to stream and forest boundaries, and did not follow our hypothesis of decreasing damage with distance from the boundary. Overall, we show that local topography (elevation and aspect), forest composition (deciduous or coniferous), and the meteorological characteristics of the disturbance event acted together to determine the spatial extent of ice storm damage.

Remote sensing of relative moisture status in old growth Douglas fir

One of the limiting factors affecting the growth of trees is the presence or absence of sufficient moisture. In locations where seasonal moisture deficits are frequent this can lead to substantial variations in the magnitude of tree growth. In these situations the detection of variations of canopy moisture through remote sensing techniques can improve forest mapping and management. This Letter reports on a study examining the potential of utilizing optical remotely sensed data to detect variations in canopy reflectance at a number of growth-limited sites located on southern Vancouver Island, Canada. Topographic variations coupled with rapidly drained soils and precipitation induced moisture deficits promote spatial variations in growth rates of the dominant tree species, coastal Douglas fir. Optical remotely sensed data were collected using the AVIRIS sensor and comparison of annual growth rates with reflectance data made.

Variability of boreal forest reflectances as measured from a helicopter platform

Helicopter and surface‐based radiometric measurements of forested sites acquired during the Boreal Ecosystem‐Atmosphere Study (BOREAS) were analyzed to assess reflectance variability and to address the relationships of reflectances and vegetation indices (VIs) with overstory leaf area index (LAI). Statistical measures of central tendency and linear regression analysis, including a multivariate regression which introduced the effects of understory reflectance, were used. The sensors were deployed during all three intensive field campaigns (IFCs) of 1994 and consisted of (1) a helicopter‐based eight‐channel Barnes modular multiband radiometer (MMR) and a ground‐based Spectron Engineering SE‐590 field spectroradiometer to measure spectral reflectance factors, (2) a ground‐based Sun photometer used for atmospheric correction, and (3) a LI‐COR LAI‐2000 plant canopy analyzer and the tracing radiation and architecture of canopies (TRAC) sensor for retrieval of canopy LAI. Of the eight MMR bands, those most strongly correlated to variations in surface LAI were the second and third middle infrared (IR) (1.57–1.80 μm and 2.08–2.37 μm, respectively) with r2=0.45 in both cases, and the red (0.63–0.68 μm) with r2=0.34. In general, linear regression analyses between the canopy LAI measurements and both the helicopter reflectances and VIs resulted in low r2 values commonly less than 0.3. When stratified by season and species, only among sites that were vegetated by aspen (Populus tremuloides) was a stronger relationship observed. In the aspen stands, LAI displayed a traditional trend that maximizes the utility of the VIs: a negative relationship with red reflectance (r2=0.62) and a positive relationship with the near IR (NIR) of r2 = 0.21. The absence of a positive LAI‐NIR reflectance trend degraded the information content of the VIs for the conifers. A multivariate regression confirmed the influence of the understory vegetation on the sparse canopies of the boreal forest and suggested that a more complex model‐based approach is necessary to better characterize variations in canopy reflectance. Atmospheric contamination of the radiometric signal by smoke from forest fires may also have contributed to the initial unsatisfactory results.

Vegetation canopy PAR absorptance and NDVI: An assessment for ten tree species with the SAIL model

The relation between the normalized difference vegetation index (NDVI) and the fraction of absorved photosynthetically active radiation (f APAR) was examined for ten different forest types by using the scattering-from-arbitrarily-inclined-leaves (SAIL) radiative transfer model. Leaf reflectance and transmittance, twigreflectance, and background reflectance data were collected as part of field experiments whose sites contain species whose ranges cover a significant part of western and northern North America. This provides a sense of various that occur at continental scales. Actual backgrounds of forests include litter and mosses; these materials did not fall along a soil line in red-near infrared reflectance space. the simulations indicated that, at low values of the leaf area index (LAI), the background reflectance had a significant effect on the canopy reflectance, although little effect on photosynthetically active radiation (PAR) absorption. At higher values of LAI, leaf optical properties were the factors that dominated canopy reflectance and NDVI. Variations in canopy reflectance due to leaf optical properties were large, but most species had similar reflectance patterns. Green leaf optical properties, among the species studied, had little effect on f APAR. The presence of twigs in the canopy had a noticeable effect on canopy reflectance and absorption of PAR, but these effects were secondary to the effects of background and leaf optical properties.

Assessing impacts of off-nadir observation on remote sensing of vegetation: use of the Suits model

The use of Suits' (1972a, b) digital radiative transfer model to simulate the effect of nonLambertian canopy reflectance on off-nadir observations of vegetation is discussed. Canopy reflectances of cord grass are calculated using the radiative transfer model, field radiometric measurements, and airborne multispectral scanner data. The effects of varying view angles on canopy reflectance are analyzed and compared. The comparison reveals that the model is effective in simulating the sense and magnitude of reflectance change due to variable angles of observations; however, the model does not reproduce the observed dependence of nadir canopy reflectance on solar zenith angle. It is concluded that the radiative transfer model is applicable for predicting the variation in canopy reflectance due to changing view zenith angles.

Effects of Nitrogen Nutrition on the Growth, Yield, and Reflectance Characteristics of Corn Canopies1

AbstractRemote sensing offers the potential to determine rapidly the physiological condition of crops over large areas. To realize this potential, information is needed on the seasonal reflectance patterns of healthy and stressed crop canopies. A field experiment was conducted in which spectral and agronomic measurements were collected from corn (Zea mays L.) canopies grown under four N treatment levels (0, 67, 134, and 202 kg/ha) on a Raub silt loam (Aquic Argiudoll) soil. Data were collected at the Purdue Agronomy Farm, West Lafayette, Ind., on 11 dates during 1978 and 12 dates during 1979. Reflected radiation of 0.4 to 2.4 µm wavelength was measured with an Exotech 20C spectroradiometer and used to compute a reflectance factor. Agronomic data collected coincidently included biomass, leaf area index, plant height, crop development stage, percent soil cover, and surface soil moisture percent. Data were analyzed to determine: 1) the relationship between the spectral and agronomic characteristics of the canopies and 2) the spectral differences and separability of the four treatments.Nitrogen treatment effects were seen across the entire wavelength interval measured. Red reflectance was increased and near infrared reflectance was decreased from N‐deprived canopies. Spectral differences between treatments were seen throughout each growing season. The near infraredred reflectance ratio (0.76 to 0.90 µm)/(0.63 to 0.69 pm) differed more between treatments than single band reflectance measures. Variations in canopy reflectance due to wetting and drying of soils were reduced by the use of the near infrared/red reflectance ratio.The spectra of the four N treatments were significantly different (Duncan's multiple range test, P = 0.05) on August dates. Early and late in the season only two spectral classes were resolved. Of the spectral variables examined, the near infraredhed reflectance ratio most effectively separated the treatments. Discrimination results for both years were similar in spite of larger treatment differences in 1979 than in 1978. Differences in spectral response between treatments were attributable to varying soil cover, leaf area index, and leaf pigmentation values, all of which changed with N treatment. The results further confirm the potential of remote sensing for monitoring the growth and development of crops.