Qi Chen | University of Hawaii (original) (raw)
Papers by Qi Chen
Estimating tree aboveground biomass (AGB) and carbon (C) stocks using remote sensing is a critica... more Estimating tree aboveground biomass (AGB) and carbon (C) stocks using remote sensing is a critical
component for understanding the global C cycle and mitigating climate change. However, the importance
of allometry for remote sensing of AGB has not been recognized until recently. The overarching goals of
this study are to understand the differences and relationships among three national-scale allometric
methods (CRM, Jenkins, and the regional models) of the Forest Inventory and Analysis (FIA) program
in the U.S. and to examine the impacts of using alternative allometry on the fitting statistics of remote
sensing-based woody AGB models. Airborne lidar data from three study sites in the Pacific Northwest,
USA were used to predict woody AGB estimated from the different allometric methods. It was found that
the CRM and Jenkins estimates of woody AGB are related via the CRM adjustment factor. In terms of
lidar-biomass modeling, CRM had the smallest model errors, while the Jenkins method had the largest
ones and the regional method was between. The best model fitting from CRM is attributed to its inclusion
of tree height in calculating merchantable stem volume and the strong dependence of non-merchantable
stem biomass on merchantable stem biomass. This study also argues that it is important to characterize
the allometric model errors for gaining a complete understanding of the remotely-sensed AGB prediction
errors.
PLoS Biology, 2013
Ongoing greenhouse gas emissions can modify climate processes and induce shifts in ocean temperat... more Ongoing greenhouse gas emissions can modify climate processes and induce shifts in ocean temperature, pH, oxygen concentration, and productivity, which in turn could alter biological and social systems. Here, we provide a synoptic global assessment of the simultaneous changes in future ocean biogeochemical variables over marine biota and their broader implications for people. We analyzed modern Earth System Models forced by greenhouse gas concentration pathways until 2100 and showed that the entire world's ocean surface will be simultaneously impacted by varying intensities of ocean warming, acidification, oxygen depletion, or shortfalls in productivity. In contrast, only a small fraction of the world's ocean surface, mostly in polar regions, will experience increased oxygenation and productivity, while almost nowhere will there be ocean cooling or pH elevation. We compiled the global distribution of 32 marine habitats and biodiversity hotspots and found that they would all experience simultaneous exposure to changes in multiple biogeochemical variables. This superposition highlights the high risk for synergistic ecosystem responses, the suite of physiological adaptations needed to cope with future climate change, and the potential for reorganization of global biodiversity patterns. If co-occurring biogeochemical changes influence the delivery of ocean goods and services, then they could also have a considerable effect on human welfare. Approximately 470 to 870 million of the poorest people in the world rely heavily on the ocean for food, jobs, and revenues and live in countries that will be most affected by simultaneous changes in ocean biogeochemistry. These results highlight the high risk of degradation of marine ecosystems and associated human hardship expected in a future following current trends in anthropogenic greenhouse gas emissions.
The relationship between lidar-derived metrics and biomass could vary across different vegetation... more The relationship between lidar-derived metrics and biomass could vary across different vegetation types.
However, in many studies, there are usually a limited number of field plots associated with each vegetation
type, making it difficult to fit reliable statistical models for each vegetation type. To address this problem, this
study used mixed-effects modeling to integrate airborne lidar data and vegetation types derived from aerial
photographs for biomass mapping over a forest site in the Sierra Nevada mountain range in California, USA. It
was found that the incorporation of vegetation types via mixed-effects models can improve biomass estimation
from sparse samples. Compared to the use of lidar data alone in multiplicative models, the mixed-effectsmodels
could increase the R2 from 0.77 to 0.83 with RMSE (root mean square error) reduced by 10% (from 80.8 to
72.2 Mg/ha) when the lidar metrics derived from all returns were used. It was also found that the SAF (Society
of American Forest) cover types are as powerful as the NVC (National Vegetation Classification) alliance-level
vegetation types in themixed-effectsmodeling of biomass, implying that the future mapping of vegetation classes
could focus on dominant species. This research can be extended to investigate the synergistic use of high spatial
resolution satellite imagery, digital image classification, and airborne lidar data for more automatic mapping of
vegetation types, biomass, and carbon.
The challenge to retrieve canopy height from large-footprint satellite lidar waveforms over mount... more The challenge to retrieve canopy height from large-footprint satellite lidar waveforms over mountainous
areas is formidable given the complex interaction of terrain and vegetation. This study explores the potential
of GLAS (Geoscience Laser Altimeter System) for retrieving maximum canopy height over mountainous areas
in the Pacific Coast region, including two conifers sites of tall and closed canopy and one broadleaf woodland
site of shorter and sparse canopy. Both direct methods and statistical models are developed and tested using
spatially extensive coincident airborne lidar data. The major findings include: 1) the direct methods tend to
overestimate the canopy height and are complicated by the identification of waveform signal start and
terrain ground elevation, 2) the exploratory data analysis indicates that the edge-extent linear regression
models have better generalizability than the edge-extent nonlinear models at the inter-site level, 3) the
inter-site level test with mixed-effects models reveals that the edge-extent linear models have statisticallyjustified
generalizability between the two conifer sites but not between the conifer and woodland sites,
4) the intra-site level test indicates that the edge-extent linear models have statistically-justified
generalizability across different vegetation community types within any given site; this, combined with
3), unveils that the statistical modeling of maximum canopy height over large areas with edge-extent linear
models only need to consider broad vegetation differences (such as woodlands versus conifer forests instead
of different vegetation communities within woodlands or conifer forests), and 5) the simulations indicate
that the errors and uncertainty in canopy height estimation can be significantly reduced by decreasing the
footprint size. It is recommended that the footprint size of the next-generation satellite lidar systems be at
least 10 m or so if we want to achieve meter-level accuracy of maximum canopy height estimation using
direct and statistical methods.
Gaussian decomposition has been used to extract terrain elevation from waveforms of the satellite... more Gaussian decomposition has been used to extract terrain elevation from waveforms of the satellite lidar GLAS (Geoscience Laser Altimeter System), on board ICESat (Ice, Cloud, and land Elevation Satellite). The common assumption is that one of the extracted Gaussian peaks, especially the lowest one, corresponds to the ground. However, Gaussian decomposition is usually complicated due to the broadened signals from both terrain and objects above over sloped areas. It is a critical and pressing research issue to quantify and understand the correspondence between Gaussian peaks and ground elevation. This study uses 2000 km2 airborne lidar data to assess the lowest two GLAS Gaussian peaks for terrain elevation estimation over mountainous forest areas in North Carolina. Airborne lidar data were used to extract not only ground elevation, but also terrain and canopy features such as slope and canopy height. Based on the analysis of a total of 500 GLAS shots, it was found that (1) the lowest peak tends to underestimate ground elevation; terrain steepness (slope) and canopy height have the highest correlation with the underestimation, (2) the second to the lowest peak is, on average, closer to the ground elevation over mountainous forest areas, and (3) the stronger peak among the lowest two is closest to the ground for both open terrain and mountainous forest areas. It is expected that this assessment will shed light on future algorithm improvements and/or better use of the GLAS products for terrain elevation estimation.
Filtering is a crucial step in lidar data processing. The Edge-based Morphological (EM) filtering... more Filtering is a crucial step in lidar data processing. The Edge-based Morphological (EM) filtering method proposed by Chen et al. (2007, Photogrammetric Engineering and Remote Sensing, 73, pp. 175–185) is fast and can be applied to different land use and land cover types. However, it requires a large number of
parameters. It is challenging for average users to tune these parameters without a good understanding of the algorithm. This study introduces a new method to identify buildings so that the total number of parameters to be tuned is reduced
from 7 to 2. Even with fewer parameters being tuned, it was found that the average filtering error slightly decreased compared to the original algorithm when tested with the benchmark dataset provided by the International Society
for Photogrammetry and Remote Sensing (ISPRS) Commission III/WG3. This is a useful contribution to the original algorithm given that it can achieve increased accuracy in a simpler way for users.
This study proposes a new metric called canopy geometric volume G, which is derived from small-fo... more This study proposes a new metric called canopy geometric
volume G, which is derived from small-footprint lidar data,
for estimating individual-tree basal area and stem volume.
Based on the plant allometry relationship, we found that
basal area B is exponentially related to G (B 1G3⁄4, where
1 is a constant) and stem volume V is proportional to
G (V 2G, where 2 is a constant). The models based on
these relationships were compared with a number of models
based on tree height and/or crown diameter. The models
were tested over individual trees in a deciduous oak woodland
in California in the case that individual tree crowns are
either correctly or incorrectly segmented. When trees are
incorrectly segmented, the theoretical model B 1G3⁄4 has
the best performance (adjusted R2, 0.78) and the model
V 2G has the second to the best performance ( 0.78).
When trees are correctly segmented, the theoretical models
are among the top three models for estimating basal area
( 0.77) and stem volume ( 0.79). Overall, these
theoretical models are the best when considering a number
of factors such as the performance, the model parsimony,
and the sensitivity to errors in tree crown segmentation.
Further research is needed to test these models over sites
with multiple species.
Lidar is changing the paradigm of terrain mapping and gaining popularity in many applications suc... more Lidar is changing the paradigm of terrain mapping and gaining
popularity in many applications such as floodplain mapping,
hydrology, geomorphology, forest inventory, urban planning, and landscape ecology. One of the major barriers for a wider application of lidar used to be the high cost of data acquisition. However, this problem has been greatly alleviated with the thrilling developments in hardware. The first commercial airborne lidar system was introduced just ten years ago (Flood, 2001). Now, the latest system is capable of transmitting 100,000 pulses per second from an altitude of up to 2km. The pulse repetition rate has reached a maximum of more than 150 kHz and has increased by about 10-fold within the last 5 years; correspondingly, the cost of data collection has decreased by about 10 times within the same time period. Nowadays users can obtain data with a density of >1 pulses per m2 for several hundred dollars per square mile. The dramatically decreasing cost of data collection encourages more and more users to embrace this innovative technology in their application and research. For example, North Carolina has collected statewide
lidar to help the Federal Emergency Management Agency (FEMA) update their digital flood insurance rate maps (Stoker et al., 2006). A wealth of free lidar data are also accessible to the public from the websites maintained by governmental agencies such as the U.S. Geological Survey (the Center for Lidar Information, Coordination and Knowledge: CLICK), National Oceanic and Atmospheric Administration (Coastal Service Center), and U.S. Army Corps of Engineers (the Joint Airborne Lidar Bathymetry Technical Center of Expertise: JALBTCX).
Although lidar data has become more affordable for average
users, how to effectively process the raw data and extract useful information remains a big challenge. Compared to image processing, lidar is appealing in many aspects. For example, the users do not have to worry about geometric, atmospheric, and radiometric corrections. However, lidar data have some characteristics that post new challenges. First of all, lidar is essentially a kind of vector data. Different from raster data, the spatial locations of laser points have to be explicitly stored, making the file size much larger than imagery given the same “nominal” spatial resolution. Second, how to extract useful information from these seemingly random points is a relatively new research topic. The generation of digital elevation models (bare earth) is the largest and fastest growing application of
lidar data (Stoker et al., 2006). However, the research on automating the production of bare earth is still in its infancy. To make this situation worse, until recently, researchers tended not to publish their methods (Zhang et al., 2003, Chen et al., 2007). Besides terrain mapping, there is an endless list of areas where lidar has a potential application but they have not been adequately explored. I have developed a software (dubbed Tiffs: Toolbox for Lidar Data Filtering and Forest Studies) for processing lidar data and extracting bare earth and forest structure information. I will discuss the challenges and needs for lidar data distribution, management, and processing. I hope this article can shed light on the topic, not only for other software developers, but also for data providers and end users of lidar.
Filtering methods based on morphological operations have been developed in some previous studies.... more Filtering methods based on morphological operations have
been developed in some previous studies. The biggest
challenge for these methods is how to keep the terrain
features unchanged while using large window sizes for the
morphological opening. Zhang et al. (2003) tried to achieve
this goal, but their method required the assumption that the
slope is constant. This paper presents a new method to
achieve this goal without such restrictions, and methods for
filling missing data and removing outliers are proposed.
The experimental test results using the ISPRS Commission
III/WG3 dataset show that this method performs well for
most sites, except those with missing data due to the lack of
overlap between swaths. This method also shows encouraging
results for laser data with low pulse density.
Quantifying the uncertainty of the aboveground biomass (AGB) and carbon (C) stock is crucial for ... more Quantifying the uncertainty of the aboveground biomass (AGB) and carbon (C) stock is crucial for understanding
the global C cycle and implementing the United Nations Program on Reducing Emissions from Deforestation and
Forest Degradation (UN-REDD). The uncertainty analysis of remotely sensed AGB is tricky because, if validation
plots or cross-validation is used for error assessment, the AGB of validation plots does not necessarily represent
the actual measurements but estimates of the true AGB. Leveraging a recently published pan-tropical destructively
measured tree AGB database, this study proposed a new method of characterizing the uncertainty of the remotely
sensed AGB. The method propagates errors from tree- to landscape-level by considering errors in the
whole workflow of the AGB mapping process, including allometric model development, tree measurements,
tree-level AGB prediction, plot-level AGB estimation, plot-level remote sensing based biomass model development,
remote sensing feature extraction, and pixel-level AGB prediction. Applying such a method to the tree
AGB mapped using airborne lidar over tropical forests in Ghana, we found that the AGB prediction error is over
20% at 1 ha spatial resolution, larger than the results reported in previous studies for other tropical forests. The
discrepancy between our studies and others reflects not only our focus on African tropical forests but also the
methodological differences in our uncertainty analysis, especially in the aspect of comprehensively addressing
more sources of uncertainty. This study also highlights the importance of considering the plot-level AGB estimate
uncertainty when field plots are used to calibrate remote sensing based biomass models.
This study examines how education benefits health through social well-being in Hawaii where the c... more This study examines how education benefits health through social well-being in Hawaii where the centrality of community life is underscored. The 2007 Hawaii Health Survey with linked zip-code information was used to investigate the effects of education at both individual and neighborhood levels using mixed-effects models. Geographic Information System was applied to map the geographical distributions of education, social well-being, and health. It was found that individual-level education benefits mental health and its effects are largely mediated by respondents' employment status and their social well-being (social integration, social contribution, social actualization, and social coherence). Both individual and neighborhood-level education promotes physical health and their effects are partially mediated by economic well-being and two indicators of social well-being (social integration and social coherence). Results of this study suggest the independent effects of two levels of education on physical health and the importance of education and social well-being to both mental and physical health in the State of Hawaii.
Photogrammetric Engineering and Remote Sensing, Aug 2006
This study presents a new method of detecting individual treetops from lidar data and applies ma... more This study presents a new method of detecting individual
treetops from lidar data and applies marker-controlled
watershed segmentation into isolating individual trees in
savanna woodland. The treetops were detected by searching
local maxima in a canopy maxima model (CMM) with variable
window sizes. Different from previous methods, the
variable windows sizes were determined by the lower-limit
of the prediction intervals of the regression curve between
crown size and tree height. The canopy maxima model was
created to reduce the commission errors of treetop detection.
Treetops were also detected based on the fact that they
are typically located around the center of crowns. The
tree delineation accuracy was evaluated by a five-fold,
cross-validation method. Results showed that the absolute
accuracy of tree isolation was 64.1 percent, which was
much higher than the accuracy of the method, which only
searched local maxima within window sizes determined
by the regression curve (37.0 percent).
Estimating tree aboveground biomass (AGB) and carbon (C) stocks using remote sensing is a critica... more Estimating tree aboveground biomass (AGB) and carbon (C) stocks using remote sensing is a critical
component for understanding the global C cycle and mitigating climate change. However, the importance
of allometry for remote sensing of AGB has not been recognized until recently. The overarching goals of
this study are to understand the differences and relationships among three national-scale allometric
methods (CRM, Jenkins, and the regional models) of the Forest Inventory and Analysis (FIA) program
in the U.S. and to examine the impacts of using alternative allometry on the fitting statistics of remote
sensing-based woody AGB models. Airborne lidar data from three study sites in the Pacific Northwest,
USA were used to predict woody AGB estimated from the different allometric methods. It was found that
the CRM and Jenkins estimates of woody AGB are related via the CRM adjustment factor. In terms of
lidar-biomass modeling, CRM had the smallest model errors, while the Jenkins method had the largest
ones and the regional method was between. The best model fitting from CRM is attributed to its inclusion
of tree height in calculating merchantable stem volume and the strong dependence of non-merchantable
stem biomass on merchantable stem biomass. This study also argues that it is important to characterize
the allometric model errors for gaining a complete understanding of the remotely-sensed AGB prediction
errors.
PLoS Biology, 2013
Ongoing greenhouse gas emissions can modify climate processes and induce shifts in ocean temperat... more Ongoing greenhouse gas emissions can modify climate processes and induce shifts in ocean temperature, pH, oxygen concentration, and productivity, which in turn could alter biological and social systems. Here, we provide a synoptic global assessment of the simultaneous changes in future ocean biogeochemical variables over marine biota and their broader implications for people. We analyzed modern Earth System Models forced by greenhouse gas concentration pathways until 2100 and showed that the entire world's ocean surface will be simultaneously impacted by varying intensities of ocean warming, acidification, oxygen depletion, or shortfalls in productivity. In contrast, only a small fraction of the world's ocean surface, mostly in polar regions, will experience increased oxygenation and productivity, while almost nowhere will there be ocean cooling or pH elevation. We compiled the global distribution of 32 marine habitats and biodiversity hotspots and found that they would all experience simultaneous exposure to changes in multiple biogeochemical variables. This superposition highlights the high risk for synergistic ecosystem responses, the suite of physiological adaptations needed to cope with future climate change, and the potential for reorganization of global biodiversity patterns. If co-occurring biogeochemical changes influence the delivery of ocean goods and services, then they could also have a considerable effect on human welfare. Approximately 470 to 870 million of the poorest people in the world rely heavily on the ocean for food, jobs, and revenues and live in countries that will be most affected by simultaneous changes in ocean biogeochemistry. These results highlight the high risk of degradation of marine ecosystems and associated human hardship expected in a future following current trends in anthropogenic greenhouse gas emissions.
The relationship between lidar-derived metrics and biomass could vary across different vegetation... more The relationship between lidar-derived metrics and biomass could vary across different vegetation types.
However, in many studies, there are usually a limited number of field plots associated with each vegetation
type, making it difficult to fit reliable statistical models for each vegetation type. To address this problem, this
study used mixed-effects modeling to integrate airborne lidar data and vegetation types derived from aerial
photographs for biomass mapping over a forest site in the Sierra Nevada mountain range in California, USA. It
was found that the incorporation of vegetation types via mixed-effects models can improve biomass estimation
from sparse samples. Compared to the use of lidar data alone in multiplicative models, the mixed-effectsmodels
could increase the R2 from 0.77 to 0.83 with RMSE (root mean square error) reduced by 10% (from 80.8 to
72.2 Mg/ha) when the lidar metrics derived from all returns were used. It was also found that the SAF (Society
of American Forest) cover types are as powerful as the NVC (National Vegetation Classification) alliance-level
vegetation types in themixed-effectsmodeling of biomass, implying that the future mapping of vegetation classes
could focus on dominant species. This research can be extended to investigate the synergistic use of high spatial
resolution satellite imagery, digital image classification, and airborne lidar data for more automatic mapping of
vegetation types, biomass, and carbon.
The challenge to retrieve canopy height from large-footprint satellite lidar waveforms over mount... more The challenge to retrieve canopy height from large-footprint satellite lidar waveforms over mountainous
areas is formidable given the complex interaction of terrain and vegetation. This study explores the potential
of GLAS (Geoscience Laser Altimeter System) for retrieving maximum canopy height over mountainous areas
in the Pacific Coast region, including two conifers sites of tall and closed canopy and one broadleaf woodland
site of shorter and sparse canopy. Both direct methods and statistical models are developed and tested using
spatially extensive coincident airborne lidar data. The major findings include: 1) the direct methods tend to
overestimate the canopy height and are complicated by the identification of waveform signal start and
terrain ground elevation, 2) the exploratory data analysis indicates that the edge-extent linear regression
models have better generalizability than the edge-extent nonlinear models at the inter-site level, 3) the
inter-site level test with mixed-effects models reveals that the edge-extent linear models have statisticallyjustified
generalizability between the two conifer sites but not between the conifer and woodland sites,
4) the intra-site level test indicates that the edge-extent linear models have statistically-justified
generalizability across different vegetation community types within any given site; this, combined with
3), unveils that the statistical modeling of maximum canopy height over large areas with edge-extent linear
models only need to consider broad vegetation differences (such as woodlands versus conifer forests instead
of different vegetation communities within woodlands or conifer forests), and 5) the simulations indicate
that the errors and uncertainty in canopy height estimation can be significantly reduced by decreasing the
footprint size. It is recommended that the footprint size of the next-generation satellite lidar systems be at
least 10 m or so if we want to achieve meter-level accuracy of maximum canopy height estimation using
direct and statistical methods.
Gaussian decomposition has been used to extract terrain elevation from waveforms of the satellite... more Gaussian decomposition has been used to extract terrain elevation from waveforms of the satellite lidar GLAS (Geoscience Laser Altimeter System), on board ICESat (Ice, Cloud, and land Elevation Satellite). The common assumption is that one of the extracted Gaussian peaks, especially the lowest one, corresponds to the ground. However, Gaussian decomposition is usually complicated due to the broadened signals from both terrain and objects above over sloped areas. It is a critical and pressing research issue to quantify and understand the correspondence between Gaussian peaks and ground elevation. This study uses 2000 km2 airborne lidar data to assess the lowest two GLAS Gaussian peaks for terrain elevation estimation over mountainous forest areas in North Carolina. Airborne lidar data were used to extract not only ground elevation, but also terrain and canopy features such as slope and canopy height. Based on the analysis of a total of 500 GLAS shots, it was found that (1) the lowest peak tends to underestimate ground elevation; terrain steepness (slope) and canopy height have the highest correlation with the underestimation, (2) the second to the lowest peak is, on average, closer to the ground elevation over mountainous forest areas, and (3) the stronger peak among the lowest two is closest to the ground for both open terrain and mountainous forest areas. It is expected that this assessment will shed light on future algorithm improvements and/or better use of the GLAS products for terrain elevation estimation.
Filtering is a crucial step in lidar data processing. The Edge-based Morphological (EM) filtering... more Filtering is a crucial step in lidar data processing. The Edge-based Morphological (EM) filtering method proposed by Chen et al. (2007, Photogrammetric Engineering and Remote Sensing, 73, pp. 175–185) is fast and can be applied to different land use and land cover types. However, it requires a large number of
parameters. It is challenging for average users to tune these parameters without a good understanding of the algorithm. This study introduces a new method to identify buildings so that the total number of parameters to be tuned is reduced
from 7 to 2. Even with fewer parameters being tuned, it was found that the average filtering error slightly decreased compared to the original algorithm when tested with the benchmark dataset provided by the International Society
for Photogrammetry and Remote Sensing (ISPRS) Commission III/WG3. This is a useful contribution to the original algorithm given that it can achieve increased accuracy in a simpler way for users.
This study proposes a new metric called canopy geometric volume G, which is derived from small-fo... more This study proposes a new metric called canopy geometric
volume G, which is derived from small-footprint lidar data,
for estimating individual-tree basal area and stem volume.
Based on the plant allometry relationship, we found that
basal area B is exponentially related to G (B 1G3⁄4, where
1 is a constant) and stem volume V is proportional to
G (V 2G, where 2 is a constant). The models based on
these relationships were compared with a number of models
based on tree height and/or crown diameter. The models
were tested over individual trees in a deciduous oak woodland
in California in the case that individual tree crowns are
either correctly or incorrectly segmented. When trees are
incorrectly segmented, the theoretical model B 1G3⁄4 has
the best performance (adjusted R2, 0.78) and the model
V 2G has the second to the best performance ( 0.78).
When trees are correctly segmented, the theoretical models
are among the top three models for estimating basal area
( 0.77) and stem volume ( 0.79). Overall, these
theoretical models are the best when considering a number
of factors such as the performance, the model parsimony,
and the sensitivity to errors in tree crown segmentation.
Further research is needed to test these models over sites
with multiple species.
Lidar is changing the paradigm of terrain mapping and gaining popularity in many applications suc... more Lidar is changing the paradigm of terrain mapping and gaining
popularity in many applications such as floodplain mapping,
hydrology, geomorphology, forest inventory, urban planning, and landscape ecology. One of the major barriers for a wider application of lidar used to be the high cost of data acquisition. However, this problem has been greatly alleviated with the thrilling developments in hardware. The first commercial airborne lidar system was introduced just ten years ago (Flood, 2001). Now, the latest system is capable of transmitting 100,000 pulses per second from an altitude of up to 2km. The pulse repetition rate has reached a maximum of more than 150 kHz and has increased by about 10-fold within the last 5 years; correspondingly, the cost of data collection has decreased by about 10 times within the same time period. Nowadays users can obtain data with a density of >1 pulses per m2 for several hundred dollars per square mile. The dramatically decreasing cost of data collection encourages more and more users to embrace this innovative technology in their application and research. For example, North Carolina has collected statewide
lidar to help the Federal Emergency Management Agency (FEMA) update their digital flood insurance rate maps (Stoker et al., 2006). A wealth of free lidar data are also accessible to the public from the websites maintained by governmental agencies such as the U.S. Geological Survey (the Center for Lidar Information, Coordination and Knowledge: CLICK), National Oceanic and Atmospheric Administration (Coastal Service Center), and U.S. Army Corps of Engineers (the Joint Airborne Lidar Bathymetry Technical Center of Expertise: JALBTCX).
Although lidar data has become more affordable for average
users, how to effectively process the raw data and extract useful information remains a big challenge. Compared to image processing, lidar is appealing in many aspects. For example, the users do not have to worry about geometric, atmospheric, and radiometric corrections. However, lidar data have some characteristics that post new challenges. First of all, lidar is essentially a kind of vector data. Different from raster data, the spatial locations of laser points have to be explicitly stored, making the file size much larger than imagery given the same “nominal” spatial resolution. Second, how to extract useful information from these seemingly random points is a relatively new research topic. The generation of digital elevation models (bare earth) is the largest and fastest growing application of
lidar data (Stoker et al., 2006). However, the research on automating the production of bare earth is still in its infancy. To make this situation worse, until recently, researchers tended not to publish their methods (Zhang et al., 2003, Chen et al., 2007). Besides terrain mapping, there is an endless list of areas where lidar has a potential application but they have not been adequately explored. I have developed a software (dubbed Tiffs: Toolbox for Lidar Data Filtering and Forest Studies) for processing lidar data and extracting bare earth and forest structure information. I will discuss the challenges and needs for lidar data distribution, management, and processing. I hope this article can shed light on the topic, not only for other software developers, but also for data providers and end users of lidar.
Filtering methods based on morphological operations have been developed in some previous studies.... more Filtering methods based on morphological operations have
been developed in some previous studies. The biggest
challenge for these methods is how to keep the terrain
features unchanged while using large window sizes for the
morphological opening. Zhang et al. (2003) tried to achieve
this goal, but their method required the assumption that the
slope is constant. This paper presents a new method to
achieve this goal without such restrictions, and methods for
filling missing data and removing outliers are proposed.
The experimental test results using the ISPRS Commission
III/WG3 dataset show that this method performs well for
most sites, except those with missing data due to the lack of
overlap between swaths. This method also shows encouraging
results for laser data with low pulse density.
Quantifying the uncertainty of the aboveground biomass (AGB) and carbon (C) stock is crucial for ... more Quantifying the uncertainty of the aboveground biomass (AGB) and carbon (C) stock is crucial for understanding
the global C cycle and implementing the United Nations Program on Reducing Emissions from Deforestation and
Forest Degradation (UN-REDD). The uncertainty analysis of remotely sensed AGB is tricky because, if validation
plots or cross-validation is used for error assessment, the AGB of validation plots does not necessarily represent
the actual measurements but estimates of the true AGB. Leveraging a recently published pan-tropical destructively
measured tree AGB database, this study proposed a new method of characterizing the uncertainty of the remotely
sensed AGB. The method propagates errors from tree- to landscape-level by considering errors in the
whole workflow of the AGB mapping process, including allometric model development, tree measurements,
tree-level AGB prediction, plot-level AGB estimation, plot-level remote sensing based biomass model development,
remote sensing feature extraction, and pixel-level AGB prediction. Applying such a method to the tree
AGB mapped using airborne lidar over tropical forests in Ghana, we found that the AGB prediction error is over
20% at 1 ha spatial resolution, larger than the results reported in previous studies for other tropical forests. The
discrepancy between our studies and others reflects not only our focus on African tropical forests but also the
methodological differences in our uncertainty analysis, especially in the aspect of comprehensively addressing
more sources of uncertainty. This study also highlights the importance of considering the plot-level AGB estimate
uncertainty when field plots are used to calibrate remote sensing based biomass models.
This study examines how education benefits health through social well-being in Hawaii where the c... more This study examines how education benefits health through social well-being in Hawaii where the centrality of community life is underscored. The 2007 Hawaii Health Survey with linked zip-code information was used to investigate the effects of education at both individual and neighborhood levels using mixed-effects models. Geographic Information System was applied to map the geographical distributions of education, social well-being, and health. It was found that individual-level education benefits mental health and its effects are largely mediated by respondents' employment status and their social well-being (social integration, social contribution, social actualization, and social coherence). Both individual and neighborhood-level education promotes physical health and their effects are partially mediated by economic well-being and two indicators of social well-being (social integration and social coherence). Results of this study suggest the independent effects of two levels of education on physical health and the importance of education and social well-being to both mental and physical health in the State of Hawaii.
Photogrammetric Engineering and Remote Sensing, Aug 2006
This study presents a new method of detecting individual treetops from lidar data and applies ma... more This study presents a new method of detecting individual
treetops from lidar data and applies marker-controlled
watershed segmentation into isolating individual trees in
savanna woodland. The treetops were detected by searching
local maxima in a canopy maxima model (CMM) with variable
window sizes. Different from previous methods, the
variable windows sizes were determined by the lower-limit
of the prediction intervals of the regression curve between
crown size and tree height. The canopy maxima model was
created to reduce the commission errors of treetop detection.
Treetops were also detected based on the fact that they
are typically located around the center of crowns. The
tree delineation accuracy was evaluated by a five-fold,
cross-validation method. Results showed that the absolute
accuracy of tree isolation was 64.1 percent, which was
much higher than the accuracy of the method, which only
searched local maxima within window sizes determined
by the regression curve (37.0 percent).