Remote Sensing Metrics Research Articles

Matching Spring Phenology Indicators in Ground Observations and Remote-Sensing Metrics

Accurate monitoring of leaf phenology, from individual trees to entire ecosystems, is vital for understanding and modeling forest carbon and water cycles, as well as assessing climate change impact. However, the accuracy of many remote-sensing phenological products remains difficult to directly corroborate using ground-based monitoring, owing to variations in the observed indicators and the scales used. This limitation hampers the practical implementation of remote-sensing phenological metrics. In our study, the start of growing season (SOS) from 2016 to 2021 was estimated for the continental USA using Sentinel-2 images. The results were then matched with several ground-based spring vegetation phenology metrics obtained by the USA National Phenology Network (USA-NPN). In this study, we focused on the relationships between the leaf-unfolding degree (LUD), the SOS, and the factors that drive these measures. Our results revealed that: (1) the ground-based leaves and increasing leaf size stages were significantly correlated with the SOS; (2) with the closest match being observed for a leaf spread of 13%; (2) the relationship between the SOS and LUD varied according to the species and ecoregion, and the pre-season cumulative radiation was found to be the main factor affecting the degree of matching between the ground observations and the metrics derived from the Sentinel-2 data. Our investigations provide a ground-based spring phenology metric that can be used to verify or evaluate remote-sensing spring phenology products and will help to improve the accuracy of remote-sensing phenology metrics.

Carbon Stock Prediction in Managed Forest Ecosystems Using Bayesian and Frequentist Geostatistical Techniques and New Generation Remote Sensing Metrics

The study compares the performance of a hierarchical Bayesian geostatistical methodology with a frequentist geostatistical approach, specifically, Kriging with External Drift (KED), for predicting C stock using prediction aides from the Landsat-8 and Sentinel-2 multispectral remote sensing platforms. The frequentist geostatistical approach’s reliance on the long-run frequency of repeated experiments for constructing confidence intervals is not always practical or feasible, as practitioners typically have access to a single dataset due to cost constraints on surveys and sampling. We evaluated two approaches for C stock prediction using two new generation multispectral remote sensing datasets because of the inherent uncertainty characterizing spatial prediction problems in the unsampled locations, as well as differences in how the Bayesian and frequentist geostatistical paradigms handle uncertainty. Information on C stock spectral prediction in the form of NDVI, SAVI, and EVI derived from multispectral remote sensing platforms, Landsat-8 and Sentinel-2, was used to build Bayesian and frequentist-based C stock predictive models in the sampled plantation forest ecosystem. Sentinel-2-based C stock predictive models outperform their Landsat-8 counterparts using both the Bayesian and frequentist inference approaches. However, the Bayesian-based Sentinel-2 C stock predictive model (RMSE = 0.17 MgCha−1) is more accurate than its frequentist-based Sentinel-2 (RMSE = 1.19 MgCha−1) C stock equivalent. The Sentinel-2 frequentist-based C stock predictive model gave the C stock prediction range of 1 ≤ MgCha−1 ≤ 290, whilst the Sentinel-2 Bayesian-based C stock predictive model resulted in the prediction range of 1 ≤ MgCha−1 ≤ 285. However, both the Bayesian and frequentist C stock predictive models built with the Landsat-8 sensor overpredicted the sampled C stock because the range of predicted values fell outside the range of the observed C stock values. As a result, we recommend and conclude that the Bayesian-based C stock prediction method, when it is combined with high-quality remote sensing data such as that of Sentinel-2, is an effective inferential statistical methodology for reporting C stock in managed plantation forest ecosystems.

Natural outdoor environment, neighbourhood social cohesion and mental health: Using multilevel structural equation modelling, streetscape and remote-sensing metrics

Although a growing body of research has explored the relationship between neighbourhood natural outdoor environments and mental health, most studies have measured neighbourhood natural outdoor environments from a bird’s-eye perspective, rather than measuring the visual experience of green and blue space that individuals have at the ground level. In addition, few studies have investigated how different dimensions of neighbourhood social cohesion mediates the relationship between the natural outdoor environment and mental health. To bridge these gaps, we examined the relationship between neighbourhood natural outdoor environments and individuals’ mental health in Guangzhou, China, using a combination of questionnaire survey data, streetscape and remote-sensing metrics, and multilevel structural equation modelling. More particularly, this study explored the mediating effects of three dimensions of neighbourhood social cohesion (i.e. neighbourhood attachment, neighbourly interaction, and community participation). The results indicate that neighbourhood green space and blue space are both positively associated with individuals’ mental health. Neighbourhood street greenery exerts beneficial effects on mental health, directly by its visual effect, and indirectly by improving neighbourhood attachment and community participation. Neighbourhood street-view blue space and surrounding green (blue) space also has a positive influence on mental health, but it does so in a direct manner only. This study contributes to our knowledge by estimating the mediating impacts of three dimensions of neighbourhood social cohesion and applying both streetscape and remote-sensing metrics of visible green space and blue space within neighbourhoods.

Potential of UAV photogrammetry for characterization of forest canopy structure in uneven-aged mixed conifer–broadleaf forests

ABSTRACT Forest canopy structure is an important parameter in multipurpose forest management. An understanding of forest structure plays a particularly important role in the management of uneven-aged forests. The identification of vertical and horizontal variations in forest canopy structure using a ground-based survey is resource intensive, hence often demands for alternative data sources. In this study, one of the advanced remote sensing (RS) techniques, i.e. digital aerial photogrammetry was used to characterize forest canopy structure in a mixed conifer–broadleaf forest. We used aerial imagery acquired with a fixed-wing unmanned aerial vehicle (UAV) platform to produce RS metrics that could be used to classify and map forest structure types at landscape scale. Our results demonstrated that few structural and spectral metrics derived from UAV photogrammetric data, e.g. mean height, standard deviation of height, canopy cover, and percentage broadleaf vegetation cover, could characterize the forest structure across landscapes, particularly at the forest management compartment level, in a limited amount of time. We used cluster analysis for classification of forest structure types and identified five forest structure classes with varying levels of forest canopy structural complexity: (1) short, open-canopy, conifer-dominated structure; (2) short, dense-canopy, broadleaf-dominated structure; (3) tall, closed-canopy, broadleaf-dominated structure; (4) very tall, closed-canopy, conifer-dominated structure with a relatively high degree of variation in canopy height; and (5) very tall, closed-canopy, conifer-dominated structure with a relatively low degree of variation in canopy height. These classes showed relationships with forest management activities (e.g. selection harvesting) and natural disturbances (e.g. typhoon damage). Spatial distribution of forest canopy structural complexity that was revealed in this study is capable of providing important information for forest management planning and habitat modelling. Further, the simple, and flexible data-driven method used in this study to characterize forest structure has the potential to be applied with necessary changes over larger landscapes and different forest types for characterizing and mapping forest structural complexity.

Linking phytoplankton pigment composition and optical properties: A framework for developing remote-sensing metrics for monitoring cyanobacteria

This study has been performed in the framework of a research program aiming to develop a low-cost aerial sensor for the monitoring of cyanobacteria in freshwater ecosystems that could be used for early detection. Several empirical and mechanistic remote-sensing tools have been already developed and tested at large scales and have proven useful in monitoring cyanobacterial blooms. However, the effectiveness of these tools for early detection is hard to assess because such work requires the detection of low concentrations of characteristic pigments amid complex ecosystems exhibiting several confounding factors (turbidity, blooms of other species, etc.). We developed a framework for performing high-throughput measurements of the absorbance and reflectance of small volumes (∼ = 20 mL) of controlled mixtures of phytoplankton species and studied the potential of this framework to validate remote-sensing proxies of cyanobacteria concentration. The absorption and reflectance spectra of single and multiple cultures carried a specific signal that allowed for the quantitative analysis of culture mixes. This specific signal was shown to be related to known pigment absorbance spectra. The concentrations of chlorophyll-a and -b, phycocyanin and phycoerythrin could be obtained from direct absorbance measurements and were correlated with the concentration obtained after pigment extraction (R2 ≥ 0.96 for all pigments). A systematic test of every possible two-band and three-band normalized difference between optical indices was then performed, and the coincidental correlation with chlorophyll-b (absent in cyanobacteria) was used as an indicator of non-specificity. Two-band indices were shown to suffer from non-specificity issues and could not yield strong and specific relationships with phycocyanin or phycoerythrin (maximum R2 < 0.5). On the other hand, the three-band modified normalized difference indices yielded strong specific relationships (R2 > 0.8).

Predicting old-growth tropical forest attributes from very high resolution (VHR)-derived surface metrics

ABSTRACTOld-growth tropical forests are increasingly vanishing worldwide. Although the accurate quantification of tropical old-growth forests attributes is essential to understand, manage, and conserve their high diversity and biomass, conducting this task over large areas and at fine detail is not only expensive and time consuming, but also often practically impossible. This calls for the search for more efficient alternatives, particularly those based on remote sensing. In this study, we evaluate the potential of several surface metrics (tone and texture) extracted from very high resolution (VHR) satellite imagery to model the structural and diversity attributes of a tropical dry forest (TDF) in southern Mexico. We constructed simple linear models that used each forest attribute as dependent variables, and the tone and texture metrics extracted from several bands, the panchromatic (resolution = 0.5 m), red (R), infrared, and two vegetation indices (normalized difference vegetation index (NDVI), enhanced vegetation index (EVI); resolution = 2 m), of a VHR image (GeoEye-1) as predictive variables. The significance of the models including one, two, two and its interaction, and three image metrics was evaluated by comparing them with null models. The structural characteristics of the TDF (basal area (BA), mean height, stem density) showed the highest modelling potential, with the goodness-of-fit (R2) values ranging from 0.58 to 0.66. Conversely, no significant models were obtained for total crown area (TCA) and all diversity attributes. Our results show that remote-sensing metrics detect the spatial variation in the structural attributes of this old-growth TDF better than they detect the variation in its diversity. Our ability to model forest attributes at large scales at fine detail (sampling plots <0.2 ha) can be much improved by combining the use of VHR imagery with an array as wide as possible of the image surface metrics, including both tone and texture.

Predicting spatial variations of tree species richness in tropical forests from high-resolution remote sensing.

There is an increasing interest in identifying theories, empirical data sets, and remote-sensing metrics that can quantify tropical forest alpha diversity at a landscape scale. Quantifying patterns of tree species richness in the field is time consuming, especially in regions with over 100 tree species/ha. We examine species richness in a 50-ha plot in Barro Colorado Island in Panama and test if biophysical measurements of canopy reflectance from high-resolution satellite imagery and detailed vertical forest structure and topography from light detection and ranging (lidar) are associated with species richness across four tree size classes (>1, 1-10, >10, and >20 cm dbh) and three spatial scales (1, 0.25, and 0.04 ha). We use the 2010 tree inventory, including 204,757 individuals belonging to 301 species of freestanding woody plants or 166 ± 1.5 species/ha (mean ± SE), to compare with remote-sensing data. All remote-sensing metrics became less correlated with species richness as spatial resolution decreased from 1.0 ha to 0.04 ha and tree size increased from 1 cm to 20 cm dbh. When all stems with dbh > 1 cm in 1-ha plots were compared to remote-sensing metrics, standard deviation in canopy reflectance explained 13% of the variance in species richness. The standard deviations of canopy height and the topographic wetness index (TWI) derived from lidar were the best metrics to explain the spatial variance in species richness (15% and 24%, respectively). Using multiple regression models, we made predictions of species richness across Barro Colorado Island (BCI) at the 1-ha spatial scale for different tree size classes. We predicted variation in tree species richness among all plants (adjusted r² = 0.35) and trees with dbh > 10 cm (adjusted r² = 0.25). However, the best model results were for understory trees and shrubs (dbh 1-10 cm) (adjusted r² = 0.52) that comprise the majority of species richness in tropical forests. Our results indicate that high-resolution remote sensing can predict a large percentage of variance in species richness and potentially provide a framework to map and predict alpha diversity among trees in diverse tropical forests.

Where’s woolly? An integrative use of remote sensing to improve predictions of the spatial distribution of an invasive forest pest the Hemlock Woolly Adelgid

Non-native pests and pathogens present serious challenges to the management of forested ecosystems around the world. Early detection of pest and pathogen invasions may allow timely control and prevention methods to be implemented. Species distribution models (SDMs) and remote sensing (RS) methods have both been used effectively to determine locations of pest and pathogen damage. However, previous work integrating these two methods has rarely used RS metrics that have biological meaning. We use RS difference indices that show changes in forest cover from defoliation in order to map Hemlock Woolly Adelgid (HWA), Adelges tsugae, locations using MaxEnt in the Delaware Water Gap National Recreation Area (DWGNRA). Brightness, greenness, wetness, and Normalized Difference Vegetation Index (NDVI) were calculated from Landsat Thematic Mapper (TM) images for December 1982 and 2010. A difference for each index was created by subtracting the 1982 value from the 2010 value. We compared two models, one using difference indices and the other using 2010 indices along with other ancillary data layers, to determine if the more complicated but more biologically relevant difference indices were necessary for improved model performance. Variables with low importance were removed from both models, leaving NDVI, Wetness, soil, and elevation in the two final models. The difference model had an improvement in accuracy of three percent, across a number of threshold values. Despite this small difference in accuracy, however, the infected area predicted by the difference model (5.1% of total area) was approximately ½ of that predicted by the single year model (9.6% of total area). These results suggest that using remote sensing difference indices improves model accuracy slightly in terms of errors of omission, but also decreases predicted area of forest infestation by about 50%, suggesting that errors of commission would be substantially reduced using the difference approach. This method can provide forest managers more accurate information on the best locations to sample and treat.

Season-dependence of remote sensing indicators of tree species diversity

During recent years, many studies have been undertaken to investigate how spectral characteristics of forests can provide information on spatial patterns of tree species diversity (TSD). Important advances have been made, and significant relationships between TSD and remotely sensed indicators of net primary productivity and environmental heterogeneity have been reported. However, the season-dependence of these relationships has not yet been fully investigated, and the influence of phenology remains poorly understood. In this study, we aim to assess how the relationships between remote sensing indicators and TSD depend on the season of the year. TSD measures, including species richness, Shannon’s diversity and Simpson’s diversity, were determined for 82 field plots in the Afromontane cloud forests of Taita Hills, Kenya. A time series of 15 Landsat images were used to calculate a set of spectral and heterogeneity metrics. The relationship between remote-sensing metrics and TSD measures was analysed by simple and multivariate regression analysis. We conclude that the relationships between remote-sensing metrics and TSD are season-dependent. Hence, it is demonstrated the date of image acquisition is an important aspect to be considered in biodiversity studies. Given that the dependence of the relationships is closely linked to climate seasonality defining vegetation phenology, the relationships may also vary according to geographical conditions.

Drought stress and carbon uptake in an Amazon forest measured with spaceborne imaging spectroscopy.

Amazônia contains vast stores of carbon in high-diversity ecosystems, yet this region undergoes major changes in precipitation affecting land use, carbon dynamics, and climate. The extent and structural complexity of Amazon forests impedes ground studies of ecosystem functions such as net primary production (NPP), water cycling, and carbon sequestration. Traditional modeling and remote-sensing approaches are not well suited to tropical forest studies, because (i) biophysical mechanisms determining drought effects on canopy water and carbon dynamics are poorly known, and (ii) remote-sensing metrics of canopy greenness may be insensitive to small changes in leaf area accompanying drought. New spaceborne imaging spectroscopy may detect drought stress in tropical forests, helping to monitor forest physiology and constrain carbon models. We combined a forest drought experiment in Amazônia with spaceborne imaging spectrometer measurements of this area. With field data on rainfall, soil water, and leaf and canopy responses, we tested whether spaceborne hyperspectral observations quantify differences in canopy water and NPP resulting from drought stress. We found that hyperspectral metrics of canopy water content and light-use efficiency are highly sensitive to drought. Using these observations, forest NPP was estimated with greater sensitivity to drought conditions than with traditional combinations of modeling, remote-sensing, and field measurements. Spaceborne imaging spectroscopy will increase the accuracy of ecological studies in humid tropical forests.