Progress 02/01/10 to 10/01/13
Outputs
OUTPUTS: The overall goal of the project involved characterization of the long-term biomass/carbon sequestration potential of the redwood/Douglas-fir ecosystem of the Pacific Northwest. In addition to canopy metrics that were derived using LiDAR data and subsequently used in biomass estimation, important correlations were produced to relate canopy height to DBH. Activities: We (graduate student and I) developed a comprehensive geodatabase for the study sites located in Prairie Creek State Park, Humboldt county, CA, and two other sites in Jackson Demonstration State Forest, Mendocino, CA. We (graduate student and I) gained indepth working knowledge of USFS FUSION, TreeVaw, and ENVI/IDL software packages for LiDAR data analysis. PI and two graduate students collected field level GPS data (Tree Height and dbh) for three study sites. We (graduate student and I) developed a regression model that estimates dbh relationship to tree height. We (graduate student and I) applied the regression model to estimate biomass with old-growth study site in Prairie Creek watershed. PI integrated the data and analytical methods in two courses teaching about 200 undergraduate students and graduate students The PI mentored 2 graduate students and 2 undergraduate students on LiDAR remote sensing over the period. Events: We (1 graduate student and PI) presented some of the preliminary results of the LiDAR analysis via a poster at the American Geophysical Union conference in San Francisco, Dec 13-17, 2010. We presented research project scope and results to managers at Jackson State Demonstration Forest meeting. PI presented LiDAR application in forest ecology to land managers, scientists, and agency personnel at public meeting organized by the Humboldt Bay Initiative. Services: We consulted with managers at Redwood National and State Parks, Save-the-Redwood League about our research projects and about correlating LiDAR-derived forest metrics to field level measurements. Also, we contacted managers and database/GIS analysts at Jackson Demonstration State Forests and Redwood Science Lab on geospatial data sets and LiDAR data. Products: In addition to an updated and georeferenced for the study sites, the project work has resulted in demonstrating the utility and application of LiDAR remote sensing in forest ecology. Based on the TreeVaw analysis, the ratio of paired tree to total considered trees is (36:57 = 61.4%) if limit of distance for pairing tree is within 3 m and (25:57 = 43.9%) at the limited distance of 2.5 m. Additionally, regression analysis between LiDAR-derived tree height and field measurements of dbh resulted in R2 values of 0.92 and 0.74 for Douglas-fir and redwood species respectively. At the Prairie Creek study site, LiDAR-derived heights ranged from 85.58 m - 116.0 m with an average of 93.20 m while ground measurements ranged between 85.06 m - 97.81 m with an average of 92.31 m. Based on this, above ground biomass estimated in our study ranged from 319.62 MgHa-1 to 5761.94 MgHa-1 and are within the range reported in the literature for the redwoods of the Pacific Northwest (Gonzalez 2010). PARTICIPANTS: Individuals Mahesh Rao (PI), Associate Professor, Associate Professor, Department of Forestry and Wildland Resources, Humboldt State University. Jomals John, Research Assistant, Department of Forestry and Wildland Resources, Humboldt State University. Hai Vuong, Research Assistant, Department of Forestry and Wildland Resources, Humboldt State University. Erik Stewart, Research Assistant, Department of Forestry and Wildland Resources, Humboldt State University. Mark Lawler, Research Assistant, Department of Forestry and Wildland Resources Partner Organizations Mr. Bill Kruse - Save the Redwood League Dr. Steve Sillett, Department of Forestry and Wildland Resources, Humboldt State University Dr. Phillip van Mantgem, Research Ecologist, USGS Western Ecological Research Center Ms. Judy Wartela, Ecologist, Redwood National and State Parks Mr. Daryl Van Dyke, GIS Specialist, Redwood National & State Parks Mr. Lathrop Leonard, Forest Ecologist, California State Parks Training or Professional Development Graduate and Undergraduate Students were trained (as listed above). Collaboration with land managers and scientists in agencies listed above. TARGET AUDIENCES: The target audiences for this project are academics and research scientists involved in forest ecology, conservation practitioners, and public and private forest land managers. The project created opportunities for classroom, laboratory, and field-based experiential learning for graduate and undergraduate students involved in field surveying, forest inventory and forest structure measurement using remote sensing techniques. PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
Preliminary results of Above Ground Biomass Analysis: Research involving biomass estimation is important for better understanding carbon sequestration and related climate change impacts. Our research provides valuable information about LiDAR-based methods and demonstrates techniques for better estimation and characterization of forestlands, particularly the old-growth redwoods and douglas-fir forests of the Pacific Northwest. Improved techniques involving LiDAR-derived canopy metrics are definitely not only effective but efficient and cost-effective in quantifying biomass that are further useful in landscape level models aimed at forest health and restoration. Additionally, various landscape models related to climate change benefit from accurate predictive biomass models developed using LiDAR data since optical sensors usually suffer from response saturation. Our research efforts focused on the Prairie Creek and Casper Creek Watersheds provide an ideal framework to amass exhaustive data both at canopy and landscape levels. Our methods and results have demonstrated that these old-growth redwood forests are huge sinks for biomass accumulation and subsequently carbon sequestration, thereby providing predictive model using LiDAR-derived estimates. Our research serves as a basis for regional managers and scientists at state and local agencies to further acquire remote sensing data for future temporal analysis. Our work involving LiDAR remote sensing has characterized old-growth canopy height as a surface model along with specific relationship (regression model) to estimating DBH. This greatly benefits in rapid and efficient estimation of site-specific allometric equations for not only biomass but other related ecological services in forest environments. Our work pertaining to LiDAR remote sensing for above ground biomass estimation contributes to the growing knowledge base of ecological applications of active remote sensing. However, our study is unique in the sense hardly much literature exists regarding LiDAR application for biomass estimation of the redwoods. Our research provides data models to capture redwood canopy height which is a very important canopy metric for biomass estimation.
Publications
- Vuong, H., an M. Rao. 2013. Uncertainty analysis of biomass estimates based on LiDAR. Remote Sensing of Environment. In preparation.
- E. Stewart and Rao, M. 2010. Estimating Above Ground Biomass using LiDAR in the Northcoast Redwood Forests. Abstract B33A-0395. Presented at 2010 Fall Meeting, AGU, San Francisco, Calif., 13-17 Dec.
- John J., M. Rao. 2012. Biomass Estimation of redwoods using a small foot print LiDAR data. Sensors. In Review.
|
Progress 01/01/11 to 12/31/11
Outputs
OUTPUTS: Major activities related to the past several months included identifying pertinent algorithms relevant to the pacific northcoast environments. Modern technology gives us an edge over the past generations, who contributed immensely to link the natural resources and economic aspects. Today, development of remote sensing and most modern software help us in obtaining more accurate data and facilitate analysis more simple than ever before. InVEST (Integrated Valuation of Ecosystem Services and Tradeoffs) is one such powerful tools developed in the open-source environment to function as a decision-making system. InVEST facilitates the quantification of goods and services from ecosystems under alternative land use scenarios and links this information to policy and management. A great deal of time was spent trying to understand the ecological and economic production functions. Additionally, software routines using land cover for the Prairie Creek study area were experimented for carbon outputs. Some of the salient features of InVEST that were noteworthy included the ability to estimate net amount of carbon stored , the total biomass removed from harvested parcels, and market and social values of the carbon sequestered in remaining stock. However, certain limitations were also identified - assumption of linear change in carbon sequestration over time, potentially inaccurate discounting rates, important biophysical conditions such activity of soil organisms are not included. "National Biomass and Carbon Dataset (NBCD)" providing a circa year 2000 high-resolution baseline data set of woody vegetation height (basal area weighted height), aboveground live dry biomass and derived carbon stock in the conterminous United States was downloaded and uploaded into ARC map for comparison. Further research with FUSION software involved sensitivity analysis for the cell-size and median to generate different renditions of the bare earth as well as the canopy height models. Then the FUSION files were converted to Arc GIS format. Analysis surfaces such as slope, aspect and curvature were created in Arc Map. The ortho image of the area was uploaded into Arc map and the LiDAR data converted into ArcGIS format was brought in. A tree list was generated from about 245 data points identified. Their coordinates, height, height to crown base, maximum and minimum crown width were recorded. A maximum tree height of 115.85 was estimated using the canopy height model. These coordinates will be checked with the ground truth and the height obtained from the fusion software will be used in allometric equations. The new version of InVEST software will be used to calculate the biomass and Carbon. Additionally, a detailed review of literature was conducted to shortlist allometric equations and algorithms specific to the Redwood ecosystems. The focus was to relate and identify sensitive parameters in those allometric equations that can be estimated using LiDAR data. PARTICIPANTS: A new graduate student has joined the research effort to characterize and quantify carbon pools using geospatial modeling tools including InVEST and FUSION. Additionally, a post-doctoral researcher will provide his expert knowledge in carbon modeling using the CENTURY and DAYCENT models. TARGET AUDIENCES: Not relevant to this project. PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
An important approach identified during the literature search included the use of InVEST modeling routines along with important algorithms specific to the region. We intend to further explore these modeling results with extensive ground measurements that will be conducted during the summer months. Additionally, we plan to collaborate with the Stanford team to closely work in fine-tuning the InVEST model for its specific use in the pacific northwest. Additionally, a post-doctoral researcher on visiting status will provide important input in terms of implementing CENTURY and DAYCENT models for biomass estimation for the study area. The existing geodatabase developed will be manipulated to generate point locations within the study area to simulate both short and long carbon pools. These estimates will be again correlated with the InVEST and NBCD estimates.
Publications
- No publications reported this period
|
Progress 01/01/10 to 12/31/10
Outputs
OUTPUTS: The overall goal of the project involved characterization of the long-term biomass/carbon sequestration potential of the redwood ecosystem of the pacific northwest. In terms of the specific objectives related to the geospatial database and analysis, some of the outputs of the project for the past year involved geospatial data search and collection, geodatabase setup, and preliminary analysis. Project activities involved collection of geospatial data from various sources including the USDA-NRCS Geospatial Data Gateway. A geodatabase was developed for the Prairie Creek Redwood Forests within ArcGIS environment. Preprocessing activities pertaining to the dataset included subsetting the data, georeferencing. Some of the related activities involved performing a literature review and familiarizing with the USFS software, FUSION. Fusion is a LiDAR viewing and analysis software suite developed by the USFS Pacific Northwest Research Station - Silviculture and Forest Models Team. The activities were centered around gaining an understanding of the software for purposes of application for this project. This involved downloading and installing the software and executing different software routines geared toward processing LiDAR data including Canopy Height Model (CHM) and terrain model. Preliminary analysis were performed using the Prairie Creek Redwood geodatabase. The PI mentored two undergraduate students on geospatial database and LiDAR application. One of those undergraduate students is very interested in pursuing the research and developing ideas for a master's thesis. Events: We presented some of the preliminary results of the LiDAR analysis via a poster at the American Geophysical Union conference in San Francisco, Dec 13-17, 2-010. Services: We contacted and consulted with managers at Redwood National and State Parks, Save-the-Redwood League about our research projects and about correlating LiDAR-derived forest metrics to field level measurements. Products: Geodatabase for the study area. This included relevant geospatial data for the Prarie Creek study area. Some of the geospatial datasets included elevation, NAIP imagery, soils, Landsat data, and also LiDAR data. Other products of the project included a working knowledge of the USFS FUSION software, and a bank of literature on LiDAR applications for biomass/carbon estimation. PARTICIPANTS: Mahesh Rao (PI), Associate Professor, Associate Professor, Department of Forestry and Wildland Resources, Humboldt State University. Erik Stewart, Research Assistant, Department of Forestry and Wildland Resources, Humboldt State University. Andelyn Jesson, Research Assistant, Department of Forestry and Wildland Resources, Humboldt State University. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
Preliminary results of Above Ground Biomass Analysis: Our preliminary analysis involved developing a geospatial database, including LiDAR data collected in 2007 for the study site, and analyzing the data using USFS Fusion software. The study area comprised of a one hectare section of mixed hardwood trees in the Del Norte Coast Redwood State Park, located in Crescent City, CA. In addition to the LIDAR data, field data pertaining to DBH, basal area, species collected by the State Park were used in the analysis. A series of analytical steps were executed using the USFS FUSION software to produce some intermediate data such as bare earth model, canopy height model, canopy coverage model, and canopy maxima derived list of trees. A total of over 1000 trees were estimated, and then with thinning (to eliminate errors due to low vegetation > 3.0 meters tall), a total of 950 trees were delineated. Ground measurements were imported as a point based shapefile and then compared to the treetop heights created from LiDAR data to the actual ground referenced data. The results were promising as most estimated treetops were within 1-3 meters of the ground measurements and generally within 3-5m of the actual tree height. Correlation/Regression analysis indicated a strong relationship (R2=0.88) between the actual tree height and the LiDAR derived estimates. Similarly, analysis of the basal area indicated a strong relationship (R2=0.77). Finally, above ground biomass was estimated using the LiDAR-derived canopy metrics and allometric equations (Jenkins, et al.) Our results are roughly in line with several other studies of biomass in redwood and mixed hardwood forests in the Pacific Northwest. Further research into the accuracy of LiDAR derived modeling of forest metrics is needed to confirm the results and support further research in the area.
Publications
- E. Stewart and Rao, M. 2010. Estimating Above Ground Biomass using LiDAR in the Northcoast Redwood Forests. Abstract B33A-0395. Presented at 2010 Fall Meeting, AGU, San Francisco, Calif., 13-17 Dec.
|