Current Projects

Modeling conifer and hardwood crown architecture

PIs  Greg Biging                    Funding Source  Intramural

Models of tree crowns are used in several fields of resource management. For instance, Biging and Dobbertin used models of tree crowns in their studies of tree competition in mixed conifers. Crown models have been used to predict wildlife habitat and as a predictor of bird abundance in the lower canopy  Crown models have also been used in studies of within tree and between tree shading of solar radiation within a forest canopy.  This project seeks to model conifer and hardwood crowns with stochastic models.  The base data is derived from digital photographs and modeled using time-series and other techniques.  One goal of this project to improve the current generation of competition models used in forest simulation.

Advanced Forest Inventory and Monitoring Using an Airborne Digital Camera Integrated with a Low-Cost Positioning/Attitude Reference System 

PIs  Greg Biging and Peng Gong  Funding Source  NCASI/USFS
Duration  1999-2001 

Cooperators:  Champion International Corporation, Mendocino Redwood Company, Gualala Redwoods Inc,  HJW,  and Jetway Graphics Systems Co.

The main objective of this project is to provide low-cost yet accurate estimates of as many important forest inventory parameters as it is feasible to measure and infer with an airborne digital frame camera and high spatial resolution (1-4 m) satellite data, and to compare these results to those obtainable with photogrammetric analysis of large-scale aerial photography.

We propose integrating a high-resolution digital camera into forest inventory
procedures to measure as many important forest parameters from the imagery as
feasibly can be done. Using a digital camera in this way produces significant cost savings over traditional forest inventory relying solely on ground sampling. Having both the merits of the high geometric quality of aerial photographs and the multispectral capability of some line scanning sensors, digital camera images may be analyzed using both digital photogrammetric techniques for 3D measurements and multispectral analysis for image enhancement, classification, and quantitative information extraction.

Monitoring of Impacts of Flood-Irrigation of Meadows in East Walker River Basin of California 

Co PIs  Peng Gong and Greg Biging    Funding Source NASA

Duration   1999-2001

CAMFER researchers led by Peng Gong and Greg Biging join researchers from UC Davis and UC Santa Barbara study the potential thermal loading of downstream waters due to irrigation flooding of an extensive meadow in an eastern Sierra valley near Bridgeport, California. Aerial photography, satellite imagery and possibly new EO-1 Hyperion hyperspectral data will be used to map vegetation types, estimate water table levels, while thermal remote sensing data will be used to determine the diurnal surface temperature changes on both land and water surfaces calibrated with ground observations. Field data will be compared with data retrieved from MAS/MASTER data, ASTER data and MODIS data. Results will be used to modify the SSTEMP stream temperature prediction model, widely used by fisheries biologists in the eastern Sierra for a wet meadow system. A user-queriable GIS database targeted at ranchers and land managers will also be developed. 

EO-1 and Landsat Imagery for Monitoring California's Conifer Forest and Hardwood Rangeland 

PIs  Peng Gong and Greg Biging    Funding Source    NASA

Duration   2000-2002  

Using study sites and research results from our existing forest and
hardwood rangeland monitoring projects in the Sierra Nevada mountains and its foothills to validate information extracted from EO-1 and Landsat data. Particularly, we will examine the capability of EO-1 Hyperion data in forest classification, crown closure, canopy chemistry and LAI estimation.  We will evaluate the utility of Hyperion data for synthesizing Landsat multispectral data; validate forest parameters extracted from Hyperion, ALI, TM and simulated TM data; and compare Hyperion performance with that of other satellite sensors for conifer forest and hardwood rangeland monitoring. Furthermore, we hope the results of this research will contribute to the remote sensing andecological community with some important tools and insights on the effective use of hyperspectral remote sensing in forest ecology, environmental monitoring, natural resource management and global change studies.

Fuzzy Classification and Remote Sensing   

PIs  Javier Montero and Greg Biging     

Funding Source  UC/UCM  Collaborative Grant

CoPIs  Peng Gong, Jesus San Miguel-Ayanz, Vincenzo Cutello, Jacinto Gonzalez-Pachon, Dr. Javier Yanez, Ana Del Amo

Duration   2000-2001  

This project will further develop fuzzy classification techniques that will advance our ability to accurately classify earth land cover/land use classes derived from satellite data. The proposed fuzzy classification system will be adapted from theory to applications with a series of meaningful tests using satellite imagery from the US and Spain for classifying forest and woodlands. We will start by exploiting theoretical issues and applications initiated mainly in papers by Amo-Montero-Biging, Amo-Montero-Cutello and González-Montero-Yáñez. We found that managing this kind of information requires sophisticated visual models for learning which aid classification system improvement.   The algorithmic complexity of the associated techniques needs careful analysis to ensure operational feasibility.