KEY TERMS: air quality; economic value; floodplain management; runoff; spatial aggregation; spatial scale; urban forest.
Like most southern cities, Baton Rouge has developed on land that was primarily forest. The undisturbed landscape was a mixture of upland pine forest broken up by riparian bottomland cypress and hardwood communities. Today, this land cover pattern is further divided with urban and suburban land uses. The influence of land uses on natural processes are well documented (Dunne and Leopold, 1978; USEPA Pacific Northwest Environmental Research Laboratory, 1974; Omernik, 1976, 1977). As land cover changes, through natural succession or human intervention, hydrologic and atmospheric cycles are altered (Betz Environmental Engineers, 1977; Hewlett, 1982; Johnston, 1989). Many of these changes can be quantified. Then, economic value can be assigned based on the cost to mitigate negative changes or costs saved due to positive benefits. QuantiTree (Davey Resource Group, et al., 1996) uses this approach to calculate the value of urban forest land.
The status of the East Baton Rouge Parish urban forest has been assessed (Davey Resource Group, et al., 1996). These data provide a baseline for urban forest value. The initial economic analysis estimates the total annual benefits (TAB) to the Parish to be $191,957,100 (This figure is based on several criteria and is not inclusive of all benefits, see below). These values change directly with changes in the amount of forest land. This allows for alternative land use scenarios to be evaluated for their utility to the community. This study examines the area known as Bayou Duplantier, in Baton Rouge, Louisiana (see figure 1.).
Figure 1. Drainage pattern of East Baton Rouge Parish, Louisiana.
The Bayou Duplantier study area lies between the two lines.
As a spatial phenomenon, the urban forest is a mosaic of various tree communities. Hydrologic and other environmental responses are based on leaf area and canopy characteristics. This study compares the economic value of the Bayou Duplantier generated from derived, aggregated data and from site survey. The two approaches differ by a factor of five. The results confirm the importance of scale choice and spatial aggregation for analytical factors based on spatial data.
Figure 2.. gif of East Baton Rouge Parish Urban Forest Canopy
The stand-level polygons were aggregated into larger polygons based on Parish planning districts. The original QuantiTree analysis was performed on these data. This is the baseline scenario for change analysis. To estimate the change in value if the tees in Bayou Duplantier were removed, an estimate of the stand characteristics was made. Using the Horizon Plan, and estimating that Distinct 13 contains approximately 35% of the study area a tree count was derived. Assuming a uniform density, a figure was calculated for the area in District 14. Initially, no actual data were available for the study area. All model input values were derived from the aggregated data.
The stand density was assessed by surveying four sample plots along the length of the corridor. These averaged 844 trees per acre. The floodplain is almost completely covered with cypress and bottomland hardwood. A 95 percent coverage was assumed over the 460 acre study area. The cover is virtually 100 percent deciduous. It is assumed to be constant for the entire area. The difference in benefits between these data and the baseline are assumed to form a lower boundary (conservative estimate) for the economic contribution of the extant urban forest.
TABLE 1. ANNUAL IMPACT MITIGATION
PROVIDED BY THE BAYOU DUPLANTIER RIPARIAN CORRIDOR
ESTIMATED AT TWO DIFFERENT SPATIAL SCALES
Parish-wide Site-specific
Category Survey Survey Units
Particulates bellow 10 microns 29.3 129.3 tons/year NO2 6.9 30.4 tons/year CO 0.0003 0.0015 tons/year SO2 3.2 14.7 tons/year O3 21.3 94.2 tons/year Carbon Storage 300.0 1,600 tons/year Runoff 47.46 318.24 million gal./year Electric Power 8.27 46.57 million KWh/year Gas Utilization 133.7 753.2 MBtu/year
The approach used by QuantiTree is to estimate an economic value for each of the quantifiable categories as a function of the control strategy used to mitigate their impact on the local environment. For example, as forest cover is decreased, runoff is known to increase (Dunne and Leopold, 1978). Building a retention structure is one method to retain the volume of water on a site, thus accounting for the water that would have been recycled through evapotranspiration by the forest. This minimizes the impact of runoff to other locations. Furthermore, the cost of building the structure is proportional to the volume of water to be controlled. QuantiTree uses this approach to estimate the economic impact due to the change in expected runoff when the forest canopy is altered. Likewise, as the urban forest changes, the various categories in table 1 generate updated expected values. The economic analysis generated by the site-specific survey and the aggregated, parish-wide information are compared in table 2. QuantiTree is a non-spatial distributed parameter model. There is no topology to relate the interaction between spatial units. As a result, increases and decreases in canopy should have a linear, additive effect to maintain continuity. If the spatial information is adequately represented in the aggregated units, changing any section, as in this example, should result in relatively scale independent outcomes. As the results show, this is not the case.
TABLE 2. ANNUAL ECONOMIC BENEFITS
ROVIDED BY THE BAYOU DUPLANTIER RIPARIAN CORRIDOR
Category Parish-wide Survey Site-specific Survey
Air Quality $70,000 $311,400
Carbon Storage 10,500 56,000
Hydrologic 471,100 3,182,300
Energy Conservation 427,000 2,421,800
Total Annual Benefits $978,600 $5,971,500
Difference: $4,992,900 (510%)
Table 1 compares the environmental impact summary for the 460 acre site for the two levels of spatial resolution. The differences are rather striking. The impacts from the planning district aggregated data are generally five times smaller than the site-surveyed data. Since the surveyed information is clearly more accurate than the derived data the later must be incorrect. This is reflected in the TAB figures (see table 2.). The summed comparison shows that the aggregated data produce a TAB of $978,600, while the site-surveyed data show a value of $5,971,500 per year.
In this case, the aerial photography should have provided adequate resolution to capture the scale of the urban forest canopy. The 1:1,800 scale can adequately identify tree groupings and canopy closure. The problem arises from the spatial aggregation of these data to the large planning districts and their disaggregation for use in the model. Figure 2) shows a map of both canopy characteristics and the District boundaries. As a spatial process, the canopy operates at a much finer scale than the Districts can represent.
In addition to the problem of representing the variation in canopy cover, spatial topography is not present in these studies. The juxtaposition of various forest stand types can be critical in generating hydrologic responses and other environmental impacts. This is clearly demonstrated each time the area receives a heavy precipitation event.
The application discussed here is a typical use of [spatial] environmental data. Hydrologic modeling has been a leader in the use of GIS to account for the spatial relationships of factors that influence hydrologic responses. A great deal of environmental data are collected by remote sensing and aerial photography and then converted into a GIS format. Hydrologists have recognized the utility of GIS in keeping track of spatial correspondence of watershed characteristics (Fast, et. al, 1995; USEPA Chesapeake Bay Program, 1993) and have moved away from traditional, lumped parameter modeling. Often the GIS is simply used for data collection, aggregation, and display. Data of different scales are often mixed without regard for their combined effect.
In the final analysis, spatial aggregation must be used with care. Maps are models. As in other modeling applications the scale of information processing must be adequate to support the information necessary to make proper management decisions
Davey Resource Group and John Emerson and Associates. Quantifying and Communicating Urban Forest Benefits; Benefit and Cost Analysis. Prepared for Baton Rouge Green, Baton Rouge, Louisiana. March, 1996.
Dunne, T. and L. B. Leopold, 1978. Water in Environmental Planning, Published by W. H. Freeman and Company, San Francisco, California.
East Baton Rouge City-Parish Planning Commission, 1992. Horizon: Final Plan Report.
East Baton Rouge Parish Tree Commission, 1995. East Baton Rouge Parish Urban Forest Management Plan. Report prepared by: Reich Associates, ACRT Inc., C. Blanche, and D. Earle. March 31, 1995. 44 pages. 4 appendices.
Fast, M. A. and T. K. Rajala, 1995. "Decision support system for multiobjective riparian/wetland corridor planning." In: Seminar Publication, National Conference on Environmental Problem-Solving with Geographic Information Systems. September 21-23, 1994, Cincinnati, Ohio. Pages 213-217.
Hewlett, J. D., 1982. Principles of Forest Hydrology. Published by the University of Georgia Press, Atones.
Johnston, C. A., 1989. Human Impacts to Minnesota Wetlands. Environmental Research Laboratory, Duluth, Minnesota. EPA Report Number EPA/600/J-89/519
Mitchell, J. E., 1991. A Regionalization Approach Based on Empirical Bayes Techniques with Applications to Hydrologic Forecasting, Ph.D. Dissertation, Duke University, Durham. North Carolina. 320 pages.
Mitchell, J. E., 1995. "GIS policy and uncertainty; Where do we draw the twenty-five-inch line?" In: Seminar Publication, National Conference on Environmental Problem-Solving with Geographic Information Systems. September 21-23, 1994, Cincinnati, Ohio. Pages 1-12.
Omernik, J. M., 1976. Influence of Land Use on Stream Nutrient Levels. Environmental Research Laboratory, Corvallis, Oregon. EPA Report Number EPA/600/3-76/014.
Omernik, J. M., 1977. Nonpoint Source-Stream Nutrient Level Relationships: A Nationwide Study. Environmental Research Laboratory, Corvallis, Oregon. EPA Report Number EPA/600/3-77/105.
USEPA Chesapeake Bay Program, 1993. Role and Function of Forest Buffers in the Chesapeake Bay Basin for Nonpoint Source Management. CBP Report Number CBP/TRS-91/93.
USEPA Pacific Northwest Environmental Research Laboratory, 1974 Relationships Between Drainage Area Characteristics and Non-Point Source Nutrients in Streams. Working Paper-25.
1. James E. Mitchell
Institute for Environmental Studies
Center for Coastal Energy and Environmental Resources
Louisiana State University
42 Atkinson Hall
Baton Rouge, LA 70803
Email: jmitche@unix1.sncc.lsu.edu
WWW: http://www.leeric.lsu.edu/ies/homepage.htm
home page