FSH 5109.19 - FIRE MANAGEMENT ANALYSIS AND PLANNING HANDBOOK 9/85 WO AMENDMENT 3 42.32 - Model Calibration. A simulation model must be adjusted (or calibrated) before it is used to develop outputs for program analysis so that it reflects the actual conditions associated with each specific application. Calibration is the adjustment of selected input variables so that outputs agree with those expected for a given set of conditions. Calibrate the initial attack assessment model by conforming the model output of acres burned with historical data. Do this by systematically adjusting the rate of spread inputs for each fire intensity level (FIL). In some cases you also may need to make minor adjustments in fire occurrence between FIL's. Calibrate the model against the average annual number of acres burned by fire size class for the period represented by the historical fire behavior data. Use the fire organization and dispatch strategies in the calibration process that were typical of those used during the period that created the historical record. After calibration, use the adjusted fire frequency and rate of spread inputs for each FIL for all subsequent runs of each option for the fuel model in that fire management analysis zone. Calibration procedure: 1. For each FMAZ develop a tabulation of the historical average annual number of fires and acres burned by size class and FIL (output of FIREPLAN program FPL-FREQ for National Forests). 2. Develop for each FMAZ the 50th and 90th percentile rates of spread and fire frequency by FIL for the fuel profile typical of the period in step 1 (FIREPLAN program FPL-LEVEL2 for National Forests). 3. Document on worksheets 2 and 3 the National Forest/unit organization typical of the historic period being used, and run FPL-IAA for each FMAZ. (FPL-IAA2 can be used, but do not develop costs or net value changes for the calibration runs.) 4. Compare the expected average annual number of fires and acres burned (suppression table 2 and 3) for each size class with the historic annual average for that FMAZ. The object is to achieve reasonable consistency between the two sets of data. Try to bring the modeled average annual acres burned (by size class and total) within approximately 5 percent of historic acres. However, use judgment in evaluating deviations, and consider magnitude along with percentage. A 15-acre difference between modeled and actual annual average acres burned in a size class that amounts to a 20-percent deviation probably is not significant. On the other hand, a 7-percent difference of 120 acres may be unacceptable. In deciding what is acceptable, the key point is the potential absolute effect of the deviation on total cost plus net value change. Greater percentage deviations may be tolerated in the fire size classes with small burned-acre totals and in low value, low suppression cost areas than in the larger fire size classes or in high value, high cost areas. 5. Calibrate the model by adjusting the rate of spread (ROS) and, if necessary, fire frequency by FIL inputs. Base the decision of what to adjust on how the model output differs from the historic record. First, if modeled fire sizes and acres burned are not realistic (compared to history), or if there are too many or too few escapes, make adjustments to the ROS inputs for the applicable FIL's. If, after several adjustments, too many acres burned are still being predicted, but rates of spread and fire sizes appear reasonable, try shifting a fire (or decimal fraction) to lower FIL's. If too few acres are burned, shift occurrence to higher FIL's. Total number of fires must stay the same. 6. Start calibration with the escaped fires. Adjust ROS inputs to bring the total frequency of escaped fires in the model output (suppression table 1) equal to the historical frequency calculated in the escaped fire analysis (worksheet 14a/b). The number of escape fire events in the model output must be equal to or greater than the number of size classes identified on worksheet 14a/b. The set of frequencies for the modeled escapes should agree generally with the set identified for the individual size classes in the escape fire analysis; escape size is then assigned to the escape fire table (worksheet 14) based on corresponding frequency. In adjusting ROS to create escapes at particular FIL and ROS levels, try to choose those with frequencies that are logical matches for the escape fire sizes, that is larger fire sizes usually should be matched to higher FIL's and ROS (FPL-IAA2 User's Guide). 7. Once escaped fire frequencies are conformed to the historical record, continue the calibration process working from highest to lowest size class. Adjust ROS to get acreage for each size class, as shown in suppression table 2 of the FPL-IAA2 output, to conform to historical acres within an acceptable tolerance. You may consider adjacent size classes together when comparing modeled and historical acreages. Shift numbers of fires between FIL's to achieve conformity of acreages only as a last resort. 8. If it is not possible to calibrate the model within acceptable limits with the above procedures, check the basic fire history data and the forces used (kind, number, and FIL at which dispatched) for inaccuracies. 42.33 - Analysis Program Outputs. The FPL-IAA2 program can provide the following outputs for each FMAZ for each option: 1. Containment time and size and units used for each simulated fire, by rate of spread, FIL, and representative location. 2. Final size and expected average annual frequency, emergency suppression (FFF) cost, and net resource value change (NVC) for each representative fire, by rate of spread, FIL, and representative location. 3. Summary of expected average annual number of fires and acres burned by FIL and fire size class A-E+. 4. Summary of expected annual acres burned, suppression costs, and NVC by representative location. 5. Summary of expected average annual number of uses of each suppression force unit, by FIL, and the associated annual FFF cost. In addition, the program is capable of storing output data for individual FMAZ and summarizing it for the National Forest/unit as a whole. 43 - FIRE EFFECTS AND NET RESOURCE VALUE CHANGE. Use resource specialists to develop estimates of the potential effects of fire on planned resource outputs for each FMAZ based on documented land and resource management objectives. The estimates must reflect the physical and economic change (increase or decrease) in number/quality of outputs as a result of fire at each fire intensity level (FIL) based on the difference between a situation with and without fire. Base estimates on an average acre in the area represented on which the resource is present and used or managed for use. For resources other than timber, the estimates are annual outputs. For timber, estimates are for changes in the amount of intermediate and final harvests. Determine the consequences of fire for lands for which there is a legal requirement to provide protection or for which protection is provided under contractual agreement for which reimbursement is received. Also include downstream effects, if any, due to fires on National Forest System lands. Convert physical output changes to dollar value changes per acre burned for each FIL for input to the program. The program calculates expected total average annual net resource value change for each FMAZ based on the expected burned acres for each option. 43.1 - Resources With Dollar Values. Determine dollar value changes for commercial timber, commercial forage, water use, water storage, human use of wildlife and fish and recreation resources, and improvements (structures). Unless there are well documented local values, use the current Resource Planning Act (RPA) program values for use in National Forest planning in determining value changes for other than timber and improvements. Use local stumpage values for timber. Evaluate improvements destroyed or damaged using current local costs for repair or replacement. Express values as constant dollars for the same fiscal year as cost data. When commercial timber values are unavailable to compute as directed in FSH 1909.12, use worksheet 6 to develop commercial timber values. Develop per acre values for immature stands of timber on worksheet 7. Calculate and record all other value changes directly on worksheet IAA2-11 for each representative location. Base all discounting (or compounding) upon a 4-percent rate of interest. Assume mature timber values and all improvements values are lost immediately. Express losses in water storage capacity as the average annual loss due to fire-induced siltation over a 25-year period. The calculation process on worksheet IAA2-ll for other resources assumes the significant economic impact is within a 7-year period, and the discount factor used treats losses (for simplicity of the calculation) as if they occurred in the third year. If calculation of losses over a longer period is more appropriate (and the present value difference is significant), a longer period may be used. In such cases, the discount factor on the worksheet is not valid, and you must calculate the total present value of the individual annual costs or benefits for the period used. Develop net value change (NVC) inputs for the program as an average per acre for the area the data represents (either FMAZ or representative location). 43.11 - Commercial Timber 1. Stumpage Values--Worksheet 6. (Optional.) Use this worksheet to calculate the average local mature timber stumpage price when commercial values are unavailable to compute as directed in FSH 1909.12. 2. Immature Timber Values--Worksheet 7. Use this worksheet to establish the maximum potential value loss per acre in immature stands, by species and age group. Calculate actual loss estimates for each FIL on worksheet IAA2-9. Develop values for different species and age class groups present on the unit on separate copies of worksheet 7 (ex. 1). Include only timber stands on commercial forest lands that are available, capable, and suitable. The objective is to define the future volumes of commercial timber that would have been harvested in the absence of fire, to translate those volumes into future values, calculate the present (discounted) value of those expected receipts, and adjust the result in a manner that recognizes that expected future receipts will be delayed rather than irrevocably lost. Specific steps to complete worksheet 7 are: a. Descriptors. Enter at top of worksheet as indicated. Specify species (group) and indicate with a check whether there are poletimber or seedling and sapling stands. Enter in footnote 1 the average annual real price increase (ARPI) in timber value as defined in directions for National Forest planning. One sheet covers all FIL's. b. Body of Table. One or two intermediate harvests and a final harvest beyond current stand age may be included. Complete a separate line for each. Leave unused lines blank. (1) Column 1. Expected average volume per acre for each harvest. (2) Column 2. Current stumpage price per unit volume for the harvested product. Generally, express volumes in MBF. An alternative measure of volume (cubic feet, cords) may be used as long as measure and unit price are consistent. (3) Column 3. The number of years from present stand age to the time expected harvest would occur in absence of fire. (4) Column 4. Price increase factor from appropriate column in the table in exhibit 1 for the number of years in column 3. (5) Column 6. Discount factor from the "1.04" column in the table in exhibit 1 for the number of years in column 3. c. Calculating Maximum Value Per Acre. Develop entries for column 7 on lines 4, 5, 6, 7, and 8 as indicated on the worksheet. Calculate line 5 by raising 1.04 and 1+ARPI to a power equal to the stand age (use exhibit 2 table), and divide. For example, for ARPI = .025 and a 40-year-old stand, line 5 would be 4.801/2.685, or 1.79. 3. Mature Timber Stands--Worksheet IAA2-8. Use this worksheet (ex. 3) to develop per acre estimates of mature timber stand value losses from fire. Consider only timber stands in the FMAZ that could have been harvested and sold under current market prices and are found on land classified as available, capable, and suitable. One worksheet may be used to develop estimates for a mixed stand by the stand as a whole or by individual species; it also may be used for up to three stands of single or mixed species in the FMAZ. Convert cubic feet to MBM on the basis of one cubic foot equals 5 board feet. A cord equals .5 MBM. Use a separate worksheet for each FIL that will occur in the FMAZ (except you may use one worksheet for two or more FIL's with similar fire effects). a. Column 1. Average final harvest sawtimber (or equivalent) volume per acre of forested land. b. Column 2. Percent by area that the species or group constitutes of the total mature timber on the area being represented (col. 3, worksheet IAA2-11). c. Column 3. The average proportion of the volume per acre of the primary species that a fire would kill at the FIL this worksheet represents. d. Column 4. The average proportion of the mortality of that species or stand which would not be salvaged following a fire at this FIL. e. Column 6. The mature green timber stumpage price (from worksheet 6) for this species/stand. Use columns 8 through 12 when the average per unit stumpage value of salvaged fire-killed timber would be significantly less than the per unit green timber stumpage: f. Column 8. The average proportion of the mortality reported in column 3 that would be salvageable. Line entries in columns 4 and 8 must equal 1.0. g. Column 10. The difference in per unit value between green timber stumpage and fire-killed timber stumpage, expressed as a positive number. 4. Immature Stands--Worksheet IAA2-9. Calculate the present value of the future harvest of timber destroyed by the fire. As with mature timber calculations, include only timber acreage classified as available, capable, and suitable. Use the three species (group) divisions on the worksheet to reflect differences in species values from worksheet 7, or differences in fire effects on different age stands for a single species or species group. Follow these steps to complete worksheet IAA2-9 (ex. 4): a. Columns 2, 5, 8. Enter the decimal fraction each species group/age constitutes of the total immature stand area on the area being represented (FMAZ or representative location). b. Columns 3, 6, 9. Enter the decimal fraction of the stand for each group that would, on the average, fall into each mortality category (lines 1-3), given a fire in the stand at the indicated worksheet FIL. c. Columns 4, 7, 10. Enter appropriate values from worksheets 7 on line 5, and complete indicated calculations. SEE PAPER COPY FOR EXHIBITS 1-4 43.12 - Commercial Forage. Estimate potential losses only for areas currently managed for livestock grazing (active allotments). Calculate the expected average annual animal use months (AUM's) per acre without fire (col. 5, worksheet IAA2-11); for each FIL, estimate the AUM's per acre that would be available and utilized by domestic livestock during each of the 7 years following a fire. Enter the sum of this 7-year output on worksheet 11, column 8. For example: at FIL 2, an immediate first-year reduction to 2 AUM's per acre followed by a subsequent 6 years of 8, 7, 5, 4, 4, and 4 AUM's. The sum of these outputs over the 7 years equals 34 AUM's. Enter "34" on worksheet 11, column 8 for that FIL. If a period longer than 7 years is necessary to reflect significant economic effects, follow the same procedure, except discount each year to the present using a standard present value formula; do not multiply total by the discount factor on worksheet 11. 43.13 - Water. Calculate the value change for water separately in terms of both yield and storage. 1. Water Yield. Fire over significant area often has the effect of increasing the runoff. However, you must include only potential increases in water yields that would be subsequently used. Do not consider runoff that could not be captured or would not be used. Calculate for the watershed of the represented area the increased water runoff per acre (in acre-feet), by FIL, that would be captured and used in each of the first 7 years after a fire. Enter the total per-acre output for the 7-year period on worksheet 11, column 8 for each FIL. To help in developing estimates, use data developed in the commercial timber and forage portions to calculate percentage of vegetation removed and vegetative response. 2. Water Storage. This output estimates the change in available water caused by siltation of water impoundment facilities following a fire. Consider only water that would have been used, and assume no postfire rehabilitation. If a fire at one or more FIL's would not cause a storage loss, enter zero on worksheet 11, column 9 in the appropriate space. If fire would cause reduced storage capacity and water lost from use as a consequence, enter on worksheet 11, column 7 the per-acre prefire average annual acre-feet of stored water from the area. For each FIL, enter in column 8 the reduced average annual (for a 25-year postfire period) per acre output that it would be possible to store, given the fire-induced accelerated siltation of the reservoir. This is the only category where you enter the average annual change. In all other categories, enter the total change over the period effects are estimated. 43.14 - Wildlife and Fish Resources. For each of the wildlife and fish subcategories listed on worksheet 11 that would be affected by fire, estimate the average annual pre-fire output in recreation user days (RUD) (standard 12-hour recreation days used in FORPLAN) and thousands of pounds of commercial anadromous fish in column 5. Estimate the annual per-acre output (use) on a burned area (for each FIL) for each of 7 years following a fire, and enter the total in column 8. Consider the change that would occur as a function of impaired or enhanced habitat productivity (such as increased big game populations due to improved forage), or hunter access. Do not estimate any increase or reduction in use due to fire if use would, in all probability, simply be shifted to or from another part of the general area. 43.15 - Recreation Resources. For each of the recreation subcategories on worksheet 11, estimate the current (pre-fire) per-acre average annual use, in RVD's, on areas managed for recreation in the area represented. For each FIL, estimate the annual use per acre for each of 7 postfire years, reflecting the change due to fire, and enter in column 8. However, do not show as a reduction any use that would be shifted to other lands in the general area. The capacity of developed facilities is not a direct estimate of use unless those facilities are actually used. If a recreation facility is included as an improvement loss, include only the change in recreation use from the time of the fire to the time of its likely replacement. 43.16 - Improvements. Estimate (by FIL if possible) the actual historic wildfire dollar improvement losses for each FMAZ for the past 5 to 10 years. Include only improvements that require repair or replacement because of the fire, fire-induced flooding, or soil movement. Do not include loss of function by water impoundment structures. Water storage loss is reflected in water resource effects. Exclude significant historic losses that are not part of potential future losses, such as unreplaced major structures. Estimate the cost of the necessary repair or, in the case of total loss, the depreciated value. Divide the total dollar improvement loss by the total acres burned for the FMAZ (by FIL if losses are estimated by FIL) during the same period. Enter the resulting average per-acre figure in column 13 of worksheet IAA2-11. 43.2 - Total Net Value Change per Acre Burned. (Worksheet IAA2- 11.) Summarize individual resource values and fire consequences, and develop the average per acre burned net value change for each FIL on this worksheet (ex. 1). Complete one worksheet for each FMAZ or, where different consequences are being assigned to parts of an FMAZ, for each representative location. Enter the total area (acres) to which the data apply (either the total FMAZ or the area represented by a representative location) at the top of the worksheet. Enter in column 3 the acres within that area on which there is management or planned use of each resource. Divide the acres on each line of column 3 by the total area to calculate the composite acre weighting factor for the resource in column 12. The column 5 entry for each resource (except timber) is the average annual per-acre pre-fire use of that resource for the area. Be certain to include negative signs where appropriate for values entered in columns 9 and 13. Column 13 is the average per acre net value change for each FIL. Enter these figures for each resource into the value change table described in the FPL-IAA2 User's Manual. The FPL-IAA2 program multiplies these values by the expected acres burned at each FIL to calculate total net value change for each FMAZ. SEE PAPER COPY FOR EXHIBIT 1 43.3 - Value Change Summary. The total expected annual net value change for each option must be totaled by FMAZ and for the National Forest/unit as a whole. This may be done in FPL-IAA2 through use of the Fort Collins Computer Center SPSS programs described in the FPL-IAA2 User Guide. 43.4 - Resources Without Dollar Value. The general approach to use in determining the cost-efficient option is to express changes in resource values in dollar terms. It is sometimes necessary also to consider the consequences of options with respect to specific resources or other factors that are of concern in a particular area and for which there are no marketplace product equivalents from which a dollar value can be constructed. These include: 1. Environmental Values. a. Air quality: visibility; impact on State standards. b. Fish and wildlife: long-term habitat impacts; effects on threatened and endangered species. c. Visual: foreground/background aesthetics. d. Classified areas: effects on wilderness ecology; impacts on research natural areas. e. Soils: erosion and changes in long-term productivity. f. Vegetation: long-term effects of repeated fire or absence of fire on the general ecology. 2. Social, Political, and Technical Values. a. Specific public concern about wildfire in a particular area. b. Public safety. c. Special or unique property: protection of archaeological, historical, or other sites of special concern. d. Stability of resource outputs to support local communities. e. Risk and consequences of failure of planned efficient protection program. f. Other specific concerns identified and documented in the land and resource management planning process. Two steps to use in considering these issues are described in sections 43.41 and 43.42. The first, which is more applicable in complex situations, helps to provide a relative quantitative measure on the potential impact of fire on the nondollar resource outputs on an area; the second, and more important step, provides essential information on the potential cost of meeting nondollar output objectives. 43.41 - Subjective Evaluation of Nondollar Resource Impacts. Use this evaluation in considering the consequences of a fire program option on the unit as a whole. Evaluate environmental, social, political, and technical factors that would be affected over time by fire on the unit. Use an interdisciplinary team approach and limit attention to values which would be affected differently by different fire program options. 1. Consider the extent and importance of each nondollar value in the management of the area and assign weights on the following scale: 0 = not significantly affected by the fire management program. 1 = low importance, but needs to be considered. 3 = moderate importance. 5 = a major value and concern. 2. For each element rated 1, 3, or 5, assign a value from +5 to -5, based on the overall potential net impact that fire has upon the element on that unit (+5 is a high net positive effect (benefit), -5 a high negative effect, and 0 no, or equal positive and negative, effect). 3. Multiply the element value by its weight and record the algebraic product. Algebraically sum all entries for the option. 4. Rank options from the option with the largest sum to that with the smallest sum. 43.42 - Using Costs as an Indicator of Output Value. Identifying priority nondollar resource concerns and evaluating how fire program options affect those resources provide only part of the information needed. The economic alternatives associated with meeting those concerns must also be addressed. One measure of the potential value of sustaining a particular level of environmental, social, or political output is the management cost required to achieve it. With this information the decisionmaker can choose whether the particular output is worth the cost that would have to be paid for it, or whether there are better competing uses for those dollars. While this does not explicitly value the output, it does implicitly establish some indication of the worth of the resource relative to other outputs at that time and place. For example, if the number of days that smoke from wildfires will impact an airshed is a concern, and some tangible relationship can be established between numbers of large fires and days of unacceptable air quality, then the analysis can provide useful cost and benefit information for the line officer who must make the program decision. First, run the analysis based on market value resources to establish the economically efficient level without consideration of the air quality resource. If the number of large fires is above the number acceptable for the proposed air quality objective, run alternative fire program options until you identify one that meets the air quality objective. The difference in fire program cost between the first and last option is the cost of reducing nonacceptable days to the desired level. The cost difference between the program option run selected and the first option run reflects the minimum value of that output from a fire management standpoint under the management objectives and budget constraints considered. The importance of this approach is that the benefits of proposed outputs are explicitly matched with the fire-related costs of achieving those benefits. Where this is not done, subsequent program decisions are made without awareness of their potential program cost. 44 - EVALUATION OF FIRE PROGRAM OPTIONS. Options are evaluated on the basis of expected program outputs, as identified in the analysis process, and consideration of other, more qualitative factors. The primary criterion is economic efficiency; that is, minimization of present value of total costs plus net resource value change based on market resource values. Other criteria may override the economic one in the selection of the preferred program. However, the further you depart from the identified efficient program budget level in selecting a less efficient option, the greater the justification there must be for giving other considerations more weight. In doing this, keep in mind that the difference in C+NVC between the efficient and the selected option is the annual cost or value of the nondollar element(s) involved. In evaluating program options to identify the preferred one, consider: 1. How well the option meets the overall management goals and objectives and fire management direction of the forest plan alternative for which it was developed. 2. The economic efficiency (total C+NVC) of the option. The option with the lowest C+NVC must be considered as preferred unless there are specific overriding management constraints or nondollar factors that make it undesirable; however, from a cost- effective standpoint, a difference of 5 percent or less in C+NVC may be considered essentially equal. 3. The technical feasibility of implementing the planned program forces and activities. 4. The risk and magnitude of potential consequences if estimated program effectiveness is not achieved operationally. 5. The relationship and acceptability of the planned budget level and of the predicted FFF and NVC. A program with a high budget and a low NVC may be less acceptable than a program with a lower budget and a higher NVC, the C+NVC of both being essentially equal. 45 - DOCUMENTATION. Document the following outputs from the analysis as part of the planning process record for subsequent use in National Forest planning, program planning and budget development, and fire program implementation (fire management action plan) and monitoring: 1. Identify the organization facilities, activities, and costs for the selected (preferred) option and document these on Form FS-1930-1, Multi-year RPA/PD&B System Plan; Form FS-5100-2, Financial and Organization Plan; or suitable similar Regional format. Include only the National Forest (or equivalent unit) level budgeted fire management program costs (include as separate items overhead, assessments, and fire protection funds allocated to decision variables other than protection and fuels) for each functional activity. Separate investment from operating and maintenance costs. If the preferred option is not the one with the minimum C+NVC, briefly document the reason for its selection and include the difference in budget and total C+NVC between it and the most efficient (minimum C+NVC) option. 2. For each option evaluated document: a. Total fire protection (FFP) budget. b. Expected annual emergency fire funds (FFF). c. Expected annual net resource value change (NVC). d. Total C+NVC. e. Expected annual acres burned by fire size, intensity.