<p>Predicting the fate of the terrestrial ecosystems and their role in the Earth sy... more <p>Predicting the fate of the terrestrial ecosystems and their role in the Earth system requires a quantitative and mechanistic understanding of carbon, water, and energy exchanges between the land surface and the atmosphere. While the current generation of land surface models show skill in representing many ecosystem processes, they largely disagree in the integrated response of the terrestrial biosphere to climatic change. These disagreements may be reconciled by confronting models with the diverse and expanding suite of Earth system observations in order to better constrain the underlying processes. In light of this goal, we have implemented substantial developments to the CARbon DAta-MOdel FraMework (CARDAMOM)—a data assimilation system that optimally estimates parameters of a parsimonious ecosystem model—which expand its original scope as a diagnostic tool for estimating carbon states and fluxes into a system that can infer and predict the response of carbon, water and energy cycles to climate and CO<sub>2</sub> concentrations at seasonal-to-decadal timescales. CARDAMOM 3.0 retains all functionality and model structures of previous versions, but now features a flagship model which includes coupled carbon, water, and energy cycles, along with semi-mechanistic representations of photosynthetic assimilation, allocation, phenology, autotrophic and heterotrophic respiration, snow and cold-weather processes, and soil hydrology. Additionally, the underlying framework was substantially updated in order to facilitate community use of CARDAMOM by simplifying the interface and increasing the ease with which users can integrate new observations and develop new model structures. With these new developments, CARDAMOM 3.0 provides a versatile tool for applying information from a broad array of Earth observation data to investigate carbon, water, and energy cycles and their responses to climate and atmospheric CO<sub>2</sub> across the full range of terrestrial ecosystems, from leaf level to continental scales.</p>
L'estimation précise da la biomasse des forêts tropicales est un enjeu important pour des pro... more L'estimation précise da la biomasse des forêts tropicales est un enjeu important pour des programmes tels que REDD+ et autres politiques environnementales. Cette thèse étudie comment des métriques lidar aident à estimer la biomasse aérienne (AGB) et suit trois axes: le premier chapitre traite de la détection de la dynamique des forêts. Nous explorons quelle échelle/résolution/taille de parcelle et quelles métriques lidar sont optimales pour estimer la biomasse et ses dynamiques. Nous avons trouvé qu'une résolution d'au moins un hectare et la hauteur de canopée moyenne donnent les meilleurs résultats. Le deuxième chapitre traite des dynamiques spatiales et compare neuf sites situés dans les Néotropiques. Nous présentons une nouvelle métrique représentant l'aire des larges canopées (LCA), couplée avec la densité de bois moyenne des sites pour estimer leur biomasse. Nous montrons ainsi que les différences entre sites peuvent être dépassées et qu'un seul modèle peut ...
Abstract: Modeling sub-canopy elevation is an important step in the processing of waveform lidar ... more Abstract: Modeling sub-canopy elevation is an important step in the processing of waveform lidar data to measure three dimensional forest structure. Here, we present a methodology based on high resolution discrete-return lidar (DRL) to correct the ground elevation derived from large-footprint Laser Vegetation Imaging Sensor (LVIS) and to improve measurement of forest structure. We use data acquired over Barro Colorado Island, Panama by LVIS large-footprint lidar (LFL) in 1998 and DRL in 2009. The study found an average vertical difference of 28.7 cm between 98,040 LVIS last-return points and the discrete-return lidar ground surface across the island. The majority (82.3%) of all LVIS points matched discrete return elevations to 2 m or less. Using a multi-step process, the LVIS last-return data is filtered using an iterative approach, expanding window filter to identify outlier points which are not part of the ground surface, as well as applying vertical corrections based on terrain s...
L'estimation precise da la biomasse des forets tropicales est un enjeu important pour des pro... more L'estimation precise da la biomasse des forets tropicales est un enjeu important pour des programmes tels que REDD+ et autres politiques environnementales. Cette these etudie comment des metriques lidar aident a estimer la biomasse aerienne (AGB) et suit trois axes: le premier chapitre traite de la detection de la dynamique des forets. Nous explorons quelle echelle/resolution/taille de parcelle et quelles metriques lidar sont optimales pour estimer la biomasse et ses dynamiques. Nous avons trouve qu'une resolution d'au moins un hectare et la hauteur de canopee moyenne donnent les meilleurs resultats. Le deuxieme chapitre traite des dynamiques spatiales et compare neuf sites situes dans les Neotropiques. Nous presentons une nouvelle metrique representant l'aire des larges canopees (LCA), couplee avec la densite de bois moyenne des sites pour estimer leur biomasse. Nous montrons ainsi que les differences entre sites peuvent etre depassees et qu'un seul modele peut ...
ABSTRACT Quantification of sub-canopy topography and forest structure is important for developing... more ABSTRACT Quantification of sub-canopy topography and forest structure is important for developing a better understanding of how forest ecosystems function. This study focuses on a three-step method to adapt discrete return lidar (DRL) filtering techniques to Laser Vegetation Imaging Sensor (LVIS) large-footprint lidar (LFL) waveforms to improve the accuracy of both sub-canopy digital elevation models (DEMs), as well as forest structure measurements. The results of the experiment demonstrate that LFL ground surfaces can be effectively filtered using methods adapted from DRL point filtering methods, and the resulting data will produce more accurate digital elevation models, as well as improved estimates of forest structure. The first step quantifies the slope present at the center of each LFL pulse, and the average error expected at each particular degree of slope is modeled. Areas of high terrain slope show consistently more error in LFL ground detection, and empirical relationships between terrain angle and expected LVIS ground detection error are established. These relationships are then used to create an algorithm for LFL ground elevation correction. The second step uses an iterative, expanding window filter to identify outlier points which are not part of the ground surface, as well as manual editing to identify laser pulses which are not at ground level. The semi-automated methods improved the LVIS DEM accuracy significantly by identifying significant outliers in the LVIS point cloud. The final step develops an approach which utilizes both the filtered LFL DEMs, and the modeled error introduced by terrain slope to improve both sub-canopy elevation models, and above ground LFL waveform metrics. DRL and LVIS data from Barro Colorado Island, Panama, and La Selva, Costa Rica were used to develop and test the algorithm. Acknowledgements: Special thanks to Dr. Jim Dilling for providing the DRL lidar data for Barro Colorado Island.
There is a large uncertainty in carbon allocation and biomass accumulation in forest ecosystems. ... more There is a large uncertainty in carbon allocation and biomass accumulation in forest ecosystems. With the recent availability of remotely sensed biomass estimates, we now can test some of the hypotheses commonly implemented in various ecosystem models. We used biomass estimates derived by integrating MODIS, GLAS and PALSAR data to verify above-ground biomass estimates simulated by a number of ecosystem
IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium
Radar measurements of the Earth’s land surface are sensitive to water in the vegetation and soil:... more Radar measurements of the Earth’s land surface are sensitive to water in the vegetation and soil: each frequency is jointly sensitive to a range of soil moisture and vegetation water content terms, which are often ignored in order to retrieve a single quantity of interest. Here, we explore the joint capability of multifrequency radar observations of the terrestrial land surface through an observing system simulation experiment (OSSE) case study. Specifically, we investigate the added value of temporal constraints on the carbon (C) and water (H2O) cycles through the joint use of K, C, L and P band measurements to retrieve fundamental land surface C and H2O state variables and process parameters. We use the CARbon DAta-MOdel fraMework (CARDAMOM) to represent the temporal evolution of C and H2O state variables and associated process inter-dependencies. Our results indicate that overall, the assimilation of 4-bands leads to substantial uncertainty reductions relative to single band experiments.
Additional file 1. Detailed information on LULC map, stratification map, field data, remote sensi... more Additional file 1. Detailed information on LULC map, stratification map, field data, remote sensing predictors, FDI map and methodology.
<p>Predicting the fate of the terrestrial ecosystems and their role in the Earth sy... more <p>Predicting the fate of the terrestrial ecosystems and their role in the Earth system requires a quantitative and mechanistic understanding of carbon, water, and energy exchanges between the land surface and the atmosphere. While the current generation of land surface models show skill in representing many ecosystem processes, they largely disagree in the integrated response of the terrestrial biosphere to climatic change. These disagreements may be reconciled by confronting models with the diverse and expanding suite of Earth system observations in order to better constrain the underlying processes. In light of this goal, we have implemented substantial developments to the CARbon DAta-MOdel FraMework (CARDAMOM)—a data assimilation system that optimally estimates parameters of a parsimonious ecosystem model—which expand its original scope as a diagnostic tool for estimating carbon states and fluxes into a system that can infer and predict the response of carbon, water and energy cycles to climate and CO<sub>2</sub> concentrations at seasonal-to-decadal timescales. CARDAMOM 3.0 retains all functionality and model structures of previous versions, but now features a flagship model which includes coupled carbon, water, and energy cycles, along with semi-mechanistic representations of photosynthetic assimilation, allocation, phenology, autotrophic and heterotrophic respiration, snow and cold-weather processes, and soil hydrology. Additionally, the underlying framework was substantially updated in order to facilitate community use of CARDAMOM by simplifying the interface and increasing the ease with which users can integrate new observations and develop new model structures. With these new developments, CARDAMOM 3.0 provides a versatile tool for applying information from a broad array of Earth observation data to investigate carbon, water, and energy cycles and their responses to climate and atmospheric CO<sub>2</sub> across the full range of terrestrial ecosystems, from leaf level to continental scales.</p>
L'estimation précise da la biomasse des forêts tropicales est un enjeu important pour des pro... more L'estimation précise da la biomasse des forêts tropicales est un enjeu important pour des programmes tels que REDD+ et autres politiques environnementales. Cette thèse étudie comment des métriques lidar aident à estimer la biomasse aérienne (AGB) et suit trois axes: le premier chapitre traite de la détection de la dynamique des forêts. Nous explorons quelle échelle/résolution/taille de parcelle et quelles métriques lidar sont optimales pour estimer la biomasse et ses dynamiques. Nous avons trouvé qu'une résolution d'au moins un hectare et la hauteur de canopée moyenne donnent les meilleurs résultats. Le deuxième chapitre traite des dynamiques spatiales et compare neuf sites situés dans les Néotropiques. Nous présentons une nouvelle métrique représentant l'aire des larges canopées (LCA), couplée avec la densité de bois moyenne des sites pour estimer leur biomasse. Nous montrons ainsi que les différences entre sites peuvent être dépassées et qu'un seul modèle peut ...
Abstract: Modeling sub-canopy elevation is an important step in the processing of waveform lidar ... more Abstract: Modeling sub-canopy elevation is an important step in the processing of waveform lidar data to measure three dimensional forest structure. Here, we present a methodology based on high resolution discrete-return lidar (DRL) to correct the ground elevation derived from large-footprint Laser Vegetation Imaging Sensor (LVIS) and to improve measurement of forest structure. We use data acquired over Barro Colorado Island, Panama by LVIS large-footprint lidar (LFL) in 1998 and DRL in 2009. The study found an average vertical difference of 28.7 cm between 98,040 LVIS last-return points and the discrete-return lidar ground surface across the island. The majority (82.3%) of all LVIS points matched discrete return elevations to 2 m or less. Using a multi-step process, the LVIS last-return data is filtered using an iterative approach, expanding window filter to identify outlier points which are not part of the ground surface, as well as applying vertical corrections based on terrain s...
L'estimation precise da la biomasse des forets tropicales est un enjeu important pour des pro... more L'estimation precise da la biomasse des forets tropicales est un enjeu important pour des programmes tels que REDD+ et autres politiques environnementales. Cette these etudie comment des metriques lidar aident a estimer la biomasse aerienne (AGB) et suit trois axes: le premier chapitre traite de la detection de la dynamique des forets. Nous explorons quelle echelle/resolution/taille de parcelle et quelles metriques lidar sont optimales pour estimer la biomasse et ses dynamiques. Nous avons trouve qu'une resolution d'au moins un hectare et la hauteur de canopee moyenne donnent les meilleurs resultats. Le deuxieme chapitre traite des dynamiques spatiales et compare neuf sites situes dans les Neotropiques. Nous presentons une nouvelle metrique representant l'aire des larges canopees (LCA), couplee avec la densite de bois moyenne des sites pour estimer leur biomasse. Nous montrons ainsi que les differences entre sites peuvent etre depassees et qu'un seul modele peut ...
ABSTRACT Quantification of sub-canopy topography and forest structure is important for developing... more ABSTRACT Quantification of sub-canopy topography and forest structure is important for developing a better understanding of how forest ecosystems function. This study focuses on a three-step method to adapt discrete return lidar (DRL) filtering techniques to Laser Vegetation Imaging Sensor (LVIS) large-footprint lidar (LFL) waveforms to improve the accuracy of both sub-canopy digital elevation models (DEMs), as well as forest structure measurements. The results of the experiment demonstrate that LFL ground surfaces can be effectively filtered using methods adapted from DRL point filtering methods, and the resulting data will produce more accurate digital elevation models, as well as improved estimates of forest structure. The first step quantifies the slope present at the center of each LFL pulse, and the average error expected at each particular degree of slope is modeled. Areas of high terrain slope show consistently more error in LFL ground detection, and empirical relationships between terrain angle and expected LVIS ground detection error are established. These relationships are then used to create an algorithm for LFL ground elevation correction. The second step uses an iterative, expanding window filter to identify outlier points which are not part of the ground surface, as well as manual editing to identify laser pulses which are not at ground level. The semi-automated methods improved the LVIS DEM accuracy significantly by identifying significant outliers in the LVIS point cloud. The final step develops an approach which utilizes both the filtered LFL DEMs, and the modeled error introduced by terrain slope to improve both sub-canopy elevation models, and above ground LFL waveform metrics. DRL and LVIS data from Barro Colorado Island, Panama, and La Selva, Costa Rica were used to develop and test the algorithm. Acknowledgements: Special thanks to Dr. Jim Dilling for providing the DRL lidar data for Barro Colorado Island.
There is a large uncertainty in carbon allocation and biomass accumulation in forest ecosystems. ... more There is a large uncertainty in carbon allocation and biomass accumulation in forest ecosystems. With the recent availability of remotely sensed biomass estimates, we now can test some of the hypotheses commonly implemented in various ecosystem models. We used biomass estimates derived by integrating MODIS, GLAS and PALSAR data to verify above-ground biomass estimates simulated by a number of ecosystem
IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium
Radar measurements of the Earth’s land surface are sensitive to water in the vegetation and soil:... more Radar measurements of the Earth’s land surface are sensitive to water in the vegetation and soil: each frequency is jointly sensitive to a range of soil moisture and vegetation water content terms, which are often ignored in order to retrieve a single quantity of interest. Here, we explore the joint capability of multifrequency radar observations of the terrestrial land surface through an observing system simulation experiment (OSSE) case study. Specifically, we investigate the added value of temporal constraints on the carbon (C) and water (H2O) cycles through the joint use of K, C, L and P band measurements to retrieve fundamental land surface C and H2O state variables and process parameters. We use the CARbon DAta-MOdel fraMework (CARDAMOM) to represent the temporal evolution of C and H2O state variables and associated process inter-dependencies. Our results indicate that overall, the assimilation of 4-bands leads to substantial uncertainty reductions relative to single band experiments.
Additional file 1. Detailed information on LULC map, stratification map, field data, remote sensi... more Additional file 1. Detailed information on LULC map, stratification map, field data, remote sensing predictors, FDI map and methodology.
Uploads
Papers