Fire and Environmental Change in Tropical Ecosystems
Activities during this reporting period involve continuing analysis of data from projects to assess the continental-scale emissions of greenhouse gases from widespread burning and preparation for and implementation of an aircraft remote sensing campaign to map selective harvesting and recent deforestation in the Amazon.
At the request of IBAMA, the working group implemented an aircraft campaign in May and June of 1998 to map recent fire activity in the State of Roraima, to monitor rates of selective harvesting and deforestation over large regions of Mato Grosso and Pará, and to map the Floresta Nacional de Tapajos including recent demonstrations of low-impact timber harvesting.
This year we integrated both our Extended Dynamic Range Imaging Spectrometer, which has been designed specifically for fire measurement, with a high-resolution digital color camera system (ARTIS) aboard a Lear 35 aircraft of the Força Aérea Brasileira. The FAB Lear traveled to Kansas for modifications on the ferrings for imager ports and to California for equipment installation. In California, the aircraft and crew were hosted by James Brass and Robert Higgins at NASA Ames Research Center provided a specially modified Lear imager port and fitted both ports and sensors to the FAB Lear on short notice. The ARTIS system was made available to the program by the NASA Environmental Research and Sensor Technology Program (ERAST). The Forest Service Pacific Southwest Research Station has joined with ERAST in the deployment and application of lightweight compact remote sensing systems which are being developed for use in remotely piloted aircraft.
IBAMA personnel managed the deployment this year with early assistance from NASA and the Forest Service. The latter portion of the campaign was conducted solely by IBAMA and FAB personnel. Selective harvesting targets were successfully mapped as was the entire Floresta Nacional de Tapajos. The Lear was unable to map recent fire activity in Roraima, despite a three-week deployment there, because of persistent low-level clouds and rain throughout the period of the campaign. This year’s remote sensing campaign provided training for IBAMA personnel in management of the program and successfully integrated our technology with the FAB group tasked with environmental monitoring under SIVAM. We are now planning a second campaign in August 1998 to continue fire measurements in the Amazon region.
During this reporting period, we have continued analysis of fire occurrence across central Brazil by a change detection of ash-covered ground from successive scenes of Landsat Multi-Spectral Scanner (MSS) data. Analysis has progressed to add techniques for removing potential errors due to counting of cloud shadows with fire areas and to classify catalogued fires by vegetation type. We have also developed a frequency distribution of fire sizes for 173,000 measured fires. These data are being analyzed to determine the potential errors and utility for fire monitoring of future satellite remote sensing systems such as the Moderate Resolution Imaging Spectrometer (MODIS). We have also begun analysis of perimeter-to-area ratios for a selection of fires which is also important in remote fire characterization. Analyses continue to indicate a lower fire return interval than previously described in the literature and are pointing to the difficulty of using fire “counts” as a meaningful measure of regional fire activity.
The rates of carbon and energy flux from fires are key to our understanding of the potential role of tropical fires as an agent of global change. We have completed our analysis from experimental fires at the IBGE Reserva Ecolôgica during which we made nearly concurrent measurements of plume velocities, fire and plume temperatures, and gas concentrations. Analyses have also been completed of carbon and sensible heat flux from these and a range of fires in Cerrado and moist forest. We have integrated remote sensing-based measurements of radiant energy flux and fire temperature with the plume measurements to provide a complete picture of energy and carbon flux from a Cerrado fire. Results have demonstrated a linear relation between sensible heat and carbon flux across the range of extant fire conditions we have encountered in Brazil. Furthermore, we have shown that remote characterization of fire temperatures and fire line structure can provide useful estimates of the vertical component of wind over the fire and, from this and fire temperature, the sensible heat flux. Thus we have the potential for remotely measuring both sensible heat flux and carbon flux from major fires.
The fire measurements constitute a unique set of information that is valuable for the development and testing of fire behavior models. Toward this end, we have also begun the process of integrating these data with an interactive model of fire and atmospheric dynamics being developed by the Fluid Dynamics Group at the Los Alamos National Laboratory.
The working group has begun construction of a web site to facilitate outreach for the project and provide access to important data sets. Among the first products to be available will be selected images of fire in the Cerrado, examples of patterns of selective harvesting, and products from the Reserva Ecológica do IBGE at Brasília including a high-resolution terrain map and registered imagery of vegetation at the Reserva collected by the EDRIS spectrometer.
Presentations were made on the fire and global change assessment at a workshop on remote sensing hosted by the Forest Service, International Programs, in Ft. Collins, Colorado, and at a seminar sponsored by International Programs in Washington, D.C.
The development of a biomass estimation photo series for a range of fuel types is continuing in cooperation with Dr. Heloisa Miranda and the University of Brasilia. In 1997, 10 sites were located and inventory within Emas National Park. An additional 10 sites will be located and inventoried near Chapada Da Diamantina in 1998. This field effort will complete the data gathering and the photo series will be published sometime in 1999. Logistical support and planning has been completed for the 1998 field campaign. There are no delays or shortfalls at this time. The project is on schedule.
The analysis and assessment of the exposure of rural residents to smoke at a site in Rondonia has been presented at the Pacific Northwest Air Pollution Control Association, Forest Fire and Meteorology Conference, and at several training sessions in the United States. A manuscript for submittal to the Journal of Air and Waste Management has been completed. Continued interest by the Brazilian government and large areas of smoke impact from burning in 1997 and 1998 may lead to large support and a continuation of the original effort.
Initiated a new research partnership with the Tropical Forest Foundation from Belem, Pará to conduct a fire risk assessment on low impact and traditional harvesting at the Floresta Nacional do Tapajos in the state of Pará. The risk assessment will compare the change in flammability between low impact harvesting, traditional selective logging, and undisturbed primary forest. The fire risk assessment will be made on the same study plots that the Institute of Tropical Forestry is conducting an economic evaluation of the two logging systems at the FLONA Tapajos.
Continued the development of combustion algorithms of tropical biomass in cooperation with the Combustion and Propulsion Laboratory of the Brazil's Instituto Nacional de Pesquisas Espaciais (INPE), Cachoeira Paulista, Sao Paulo. Plans were initiated to implement a complete experiment 1998 and 1999 to study flaming and smoldering combustion of slash in three experimental plots at their experimental farm outside of Alta Floresta, Mato Grosso. Also, we assisted in the design and construction of a wind tunnel at INPE's facilities in Cachoeira Paulista to conduct burn under controlled conditions The burns will be used to study biomass combustion, fire physics, and flammability on the interface of the primary forest and deforested areas. Arrangements were made for logistic support the 1998 field season. This project will result in a fire risk rating system to predict fire danger during drought years in the Amazon basin.
Continued the support of a Brazilian doctoral student at the University of Washington and their dissertation research on the development of a fire risk model for Amazon forest landscapes considering forest physiology, fire meteorology, land use, and climate change.
SERVICE, PACIFIC NORTHWEST RESEARCH
SERVICE, PACIFIC SOUTHWEST RESEARCH STATION
AMES RESEARCH CENTER
CENTER FOR ATMOSPHERIC RESEARCH
Sustainable Forest Management Technologies and Practices
Within the Brazilian Amazon, the Tropical Forest Foundation (TFF) has been developing low-impact logging (LIL) methods which are being tested in several regions. Training programs have been conducted by TFF to transfer LIL knowledge and skills to local loggers and forestry professionals. However, because a comprehensive economic evaluation of LIL methods versus conventional logging (CL) methods has not been undertaken, the economic incentives for adopting the new methods are unknown.
The overall goal of this research is to evaluate the economic competitiveness of LIL methods relative to CL methods in the Brazilian Amazon. This will be accomplished by conducting a comparative financial analysis of the costs and benefits of representative LIL models relative to representative CL firms in several regions of the Brazilian Amazon. Representative LIL models are constructed using data from one or more LIL block. Likewise, representative CL firms are constructed using data from one or more CL operations. The preliminary report focuses on only one region, the “old frontier” region of Paragominas in Pará state, where we compare one LIL model with one representative CL firm.
To evaluate the costs and benefits of LIL versus CL systems, the Tropical Forest Foundation (TFF) established several 100 hectare cutting blocks at Fazenda Cauaxi near Paragominas. Block 1 (established in 1996) was the conventional logging block, and Blocks 2 (established in 1995), 3 (established in 1996) and 4 (established in 1997) were the LIL blocks. Pre-harvest inventories of commercial and potentially commercial trees were conducted on all blocks and, according to the Brazilian forestry code, permanent plots were established representing 1% of the area in each of the blocks.
Pre-harvest inventories were conducted for all commercial and potentially commercial trees that were greater than or equal to 35 cm d.b.h. on blocks 1 and 3. As part of the damage assessment, post-harvest inventories were also conducted.
To calculate the cost of each logging activity on a cubic meter basis for the LIL block, we needed to determine the productivity for each activity. This entailed calculating the rate at which the appropriate crew or individuals completed each activity. We obtained rates for all of the pre-harvest and harvest planning activities in hectares per hour (ha/hr). For harvest activities, we were able to calculate productivity in terms of cubic meters per hour (m3/hr). Productivity data was then converted to cost data using appropriate cost per hour data.
For the CL block, we used several data sets to construct representative productivity and cost information. Data were provided by TFF conducted surveys of CL operations in the region, IMAZON data, and data provided by EMBRAPA.
Analysis showed that the harvesting intensity on the CL block was somewhat higher than the harvesting intensity on the LIL block. Both the average number of trees harvested and the average volume harvested on the CL block were greater than the average number of trees and volume harvested on the LIL block. However, the volume harvested per tree harvested on the LIL block was larger, on average, than the volume per tree harvested from the CL block. This may be due to better recovery (less waste) in LIL operations, harvesting larger trees, or both.
Nearly twice as much ground area was disturbed by heavy equipment CL operations as was disturbed by LIL operations. Because regeneration is severely impacted by these sources of ground area disturbance, it is expected that LIL operations will have a significantly smaller impact on regeneration of future stands.
In addition to gains in skidding and log deck productivity due to planning, it was found that LIL operations significantly reduce the amount of wood wasted during the harvest operation due to more careful felling, bucking and skidding operations. Increased costs associated with more careful harvesting and bucking operations are more than compensated by the increase in volume and value recovered.
It was discovered that felling is the most important cause of tree mortality due to harvesting operations. The proportion of trees in the residual stand killed by felling on the CL block is several times larger than the proportion of trees in the residual stand killed by felling on the LIL block.
The preliminary analysis indicated that the total costs of CL operations are greater than the total costs of LIL operations on a per m3 and per hectare basis (but not on a per tree basis). This is because the gain in efficiency from skidding and log deck operations more than offset the increase in cost due to planning. LIL is more cost efficient than CL, and if LIL and CL operations were constrained to harvest the same number of trees, LIL would also generate larger net revenues per hectare relative to CL.
The Tropical Forest Foundation completed scheduled training for industry, community and IBAMA personnel on low-impact logging techniques on the two 96-hectare demonstration logging sites on the Tapajos National Forest near Santarem in the State of Pará. The training included harvest crews in October for local workers living in or near the FLONA, In addition two practical courses were held covering related activities such as inventory, data processing, map making, planning and low impact harvesting techniques. Safety was included in the curriculum.
Infrastructure supporting the harvest was completed during the period. This included primary roads, secondary roads, log decks and primary skid trails according to the approved IBAMA harvest plan.
The low impact harvesting of the two 96-hectare blocks was completed by the end of December with an outstanding safety record.
Some progress, although not as much as anticipated, was made on the evaluation of the impacts of harvesting on the Tapajos National Forest. We are in a holding pattern awaiting the LBA agreement. This work is lead by Dr. Michael Keller from the International Institute of Tropical Forestry.
Dr. Keller visited Brazil in November 1997 and planned a follow up trip for May 1998 to attend meetings of the international LBA Science Steering Committee. Dr. Keller is co-chair of this committee which is charged with the task of coordinating overall activities of all LBA modules from Europe, the U.S., Brazil, and other South American nations. The LBA project is very closely related to the overall evaluation efforts on the Tapajos National Forest, and Dr. Keller’s leadership has significantly leveraged the number of scientists who are or will be working on the LBA-Ecology Module. The module includes over 120 investigators (mainly from the U.S., Brazil and Ecuador) with an expected budget of $U.S. 8 million annually for three years. An estimated $US 2 million annually is planned for the Santarem/Tapajos component of the project. These efforts will significantly increase the depth and breadth of scientific evaluation of harvesting impacts.
Notable accomplishments during this period include:
Activities of individual researchers included the following:
Capacity Building Opportunities
During this period some of the difficulties related to compliance with the permitting requirements under the Law of Scientific Expeditions administered by CNPq were resolved. Both the Neff and Asner agreements are being conducted under the Law of Scientific expeditions administered by CNPq. Many of the investigators working under the LBA program are awaiting execution of the final agreement between Brazil and the U.S. and the expedited clearance procedures contained therein. Other USDA Forest Service investigators have initiated CNPq permitting through their Brazilian collaborators.
As yet IBAMA has not completed the edital process for the timber harvest on the Tapajos National Forest. However progress has been made and issues related to the adjacent communities were largely resolved by actions under Brazilian law to exclude some areas from the planned harvest. A revised edital will be issued shortly, and initiation of harvesting activities is anticipated in the 1999 dry season. Meanwhile alternative strategies based on harvesting sites on private lands adjacent to the FLONA are being explored should the Tapajos harvest not be completed in time.
A total of 7 people were trained in economic methods for research and analysis to collect and process the data described above.
Reports for this period include:
Boltz, F., T. Holmes and D. Carter. 1998. The costs and benefits of low-impact logging relative to conventional logging practices in the Brazilian Amazon. Proceedings of the 1998 Southern Forest Economics Workshop, Raleigh, NC.
Holmes, T., G. Blate, J. Zweede, and R. Junior. 1998. The costs and benefits of low-impact logging relative to conventional logging practices in the Brazilian Amazon. Preliminary Report to the Tropical Forest Foundation, Washington D.C.
Successful partnerships between TFF, IMAZON and EMBRAPA were formed. The collaborative effort will provide valuable information to the forest industry and to public forestry decision-makers.
USAID supported investment in forest management activities have been significantly leveraged.
INSTITUTE OF TROPICAL FORESTRY
SERVICE, SOUTHERN RESEARCH STATION
For more information on these projects, please contact Jan Engert, International Programs/Brazil Coordinator USDA Forest Service 1099 14th Street NW, Suite 5500W Washington, D.C. 20005 Phone: (202)273-4752 Fax: (202)273-4748 email: email@example.com