- Forest and Wildlife Research Center
Contents
1. cots REI TEES ROREM y uU Sep data es 10 Growth Projection Measurements hare 11 Site Index Measurements the xn ee DA ee Ee Nui 11 ba tides totes ele ie tee oe bats le 11 oti aes te eto ie Pale tate e Rund 12 Common Name Species Reference sue 3 00363 POMONA aeg 12 Append 2 52 ime rer te dien eee eu er ee oa LG Se ed 14 SAP Cover Type Definitions eo esos e enm egt NEVER 14 ADppendlIx ost edd Let UU Dui Dr RN P DU S pde htt 16 Power supply M Oe ULM EC Da he BRS utes 16 Gi Se MM Me An ee E M Du ee I 16 Hypsometer 24 5 bored uu nip LIN E LAE ERE bsc P bes 16 DME RUPIESINGSUESESUNA Ae ea TE Se eels 16 Appendix e darte BAR scd de ee pr ee a uie LG Se 17 Transponder nee Aa 17 prado M C ETC 18 Calibraon ssa reda E adea toni ven _ be a Phe E aA rg 18 Hypsormeter EE M eU ud RE FCU SARUM SR M Rus 18 18 19 CMT Operations oso2. d 19 Screen Contrasts Lee cec __ 19 Power 19 d Vates sa aues Da a ato E E tar 20 Slope Determinationy RR Aan cu Lp we es 20 ADDODOIX REE Wes Son IM A eet t AE 21 Dalia Recording SHeets s EEE ERE ROS 21 Measurement Plot Specifications and Procedures This document supplies definitions methodologies and standards for the inventory implementation of a pilot program for a forest monitoring and information system The objective of this document is to describe field procedures that will assure consistent and uniform data collection over all expected conditions in the study area It is expected that all plot data can be measured by a two person crew in approximately two hours on average including time spent traversing to and from the plot Field Crews Each field crew will consist of two individuals Each individual will be assigned a file designator The file designator identifies within the data structure which person conducted the height estimations To reduce bias of ocular estimations the same individual will perform estimations throughout the day Prior to leaving the office determine which crew member will perform height estimations Upon commencement of plot measurements the same i
3. Loplolly shortlear Pine Han commercial Other Fed Daks Other Softwog a Uther Yellow Pinea select White Yellow Poplar LL ssh OT HH OT 5H ur uu SEL RD s nr TE Lohgleaft slash Pine Other Hard Soft Hardwoods Other White select Red Dake Bort Tupelo Black Gum Figure 11 Example of report generated by MsFMIS from the knowledge based growth and yield user interface 14 Error matrix assessing LandSat derived cover types using high resolution aerial imagery Classified Data Reference Data Non Hardwood Mixed Pine Regeneration Total Product User s Forest Accuracy 96 Accuracy 96 Non Forest 56 0 1 0 0 57 83 58 93 33 Hardwood 4 55 0 0 0 59 100 00 93 22 Mixed 0 0 54 0 0 54 98 18 96 49 Pine 0 0 0 58 0 58 100 00 100 00 Regeneration 8 0 0 0 58 66 89 66 86 67 Total 68 55 55 58 58 294 N A N A Kappa Statistic 0 9193 Error matrix assessing LandSat derived cover types using field plot Classified Data Reference Data Non Hardwood Mixed Pine Regeneration Total Product User s Forest Accuracy 96 Accuracy 96 Non Forest 86 4 18 1 0 109 90 53 78 90 Hardwood 1 95 6 3 1 106 49 48 79 62 Mixed 6 81 173 73 0 333 77 58 51 95 Pine 1 3 12 107 0 123 46 52 86 99 Regeneration 1 9 14 46 56 126 98 25 44 44 Total 95 192 223 230 57 797 N A N A Kappa Statistic 0 5490 Age Class Assessments Age class accuracy was assesse
4. Pinus taeda Pinus echinata Pinus palustris Pinus elliotii Taxodium spp Juniperus virginiana Quercus spp erythrobalanus Quercus pagodifolia Quercus shumardii Quercus velutina Quercus nuttallii Quercus falcata Quercus marilandica Quercus laevis Quercus spp leucobalanos Quercus alba Quercus stellata Quercus prinus Quercus lyrata Quercus durandii Quercus nigra Quercus phellos Quercus laurifolia Quercus muehlenbergii Quercus virginiana Liquidambar styraciflua Fraxinus spp Platanus occidentallis Liriodendron tulipifera Nyssa aquatica Nyssa sylvatica Celtis spp Carya spp Carya illinoensis Juglans spp Populus deltoides Tilia spp Cherry Prunus serotina WHE A Yi 5 Persimmon Diospyros virginiana 40 Sassafras Sassafras albidum 41 Magnolia Magnolia spp 42 Maple Boxelder Acer spp 43 Willow Salix spp 44 Elm Ulmus spp 45 Beech Fagus spp 46 Birch Betula spp 47 Dogwood Cornus spp 48 Red Bay Persea borbonia 49 Holly spp 50 Mulberry Morus spp 51 Locust Robinia Gleditsia spp 52 Osage Orange Maclura pomifera 53 Hornbeam Carpinus Ostrya spp 54 Redbud Cercis canadensis 55 Catalpa Catalpa bignonioides 56 Exotics Albizia Sapium Paulownia spp 57 Other Hard Broadleaf Crataegus Chionanthus spp 58
5. Cover Type 00 b ele ae aes 5 Ownership 12 1 iss ba Doe UR De 5 Physioetaphlic Position eb e ut eo e RP E rx 5 ODOrabilby xa ye ROME RR AE RE E LA LI D MU AE 6 1 Longitude 24 ocu oh Sete a Re QE 6 Plot c tton ko s et auo coc en BTS 6 Tidivid at Tree usos HUN 7 Hn crc EP 7 oie os oc Baa Re eme LU I I 7 Species sede vA A eaae ds RAPERE am 7 Diameter zu 8 eee REL el 8 Pulpwood Height to Absolute AR XY 8 Pulpwood Usable obe ODORE ADORA s S 8 Saw timber Height to Absolute 435v xe X Y XXX V PEPPER Y xA YR s 8 Saw timber Usable e can ex E PRONUM Mad Ob AE e ard ERE e a UR 8 Rb PAX s e t pd 8 Broken sen deo t teste MER ee Se 9 Field Procedures oo yas RECON ere cu p eae pedum a ets 9
6. to stem length being measured to an absolute upper stem diam eter limit stem length was also measured to a usable limit typifying local mill harvesting practices Five and ten year diameter growth rates were measured to allow projection for equivalent periods Site index measurements were recorded to quantify productivity potential Portable field computers were used to collect data in an electronic format This increased the efficiency of the inventory by eliminating the need for key punching of the data and the possibility of transcription errors Data collection software was developed to accommodate all the information collected and validate the data entries in real time Figure 6 As data were entered into the recorder they were checked against a set of values appropriate for that measurement and the user was prompted to verify the entry if the input value was outside the expected range Ultrasonic measuring equipment was employed for distance and height measurements to decrease time spent on the plot Tree heights were measured with Vertex hypsometers which eliminated the need to establish the baseline distance required by clinometers Distance measuring devices DME 201s were used to establish plot radii and to check border trees without pulling a logger s tape through brush and around obstacles Figure 6 Screens from the Pro4000 field computer a main menu b enter edit screen for inventory codes c full screen view of tree tally
7. DME 201 will be carried to prevent delays caused by depleted power sources It is the responsibility of each field crew to maintain the equipment In the event equipment is lost or becomes non serviceable the field coordinator should be notified as soon as possible Land Navigation Each plot location has been randomly selected from remote sensing information to fulfill accuracy require ments for the strata it represents Latitude and longitude coordinates are used to identify plot centers Gross navigation to each plot is achieved using available maps depicting vehicular access routes It is extremely unlikely that plot access will be available to motor vehicles If permission from the landowner can be obtained prior to site visitation it may be possible to minimize the amount of time spent traversing to the plot location At no time will a field crew be authorized to tamper with vandalize or destroy fencing or gates defining property boundaries to facilitate access Damage or destruction of property with the intent to aid navigation is not permitted In the event of long traverses trail marking to assist with egress is restricted to flagging the line of travel provided flagging is biodegradable and not used excessively Equipment Calibration The Vertex hypsometer and the DME 201 require calibration with ambient environmental conditions to compute distances accurately Calibration will be conducted each time a field crew exits the vehicle in
8. Other Soft Broadleaf Foresteria Aralia spp 59 Unknown 60 Appendix 2 SAF Cover Type Definitions Cover Type Description Code Lob Short Loblolly and shortleaf pine together comprise a majority of the stocking although the proportion of each varies loblolly is usually dominant 1 Loblolly Comprised of either pure stands of loblolly pine or mixtures in which loblolly makes up the majority 2 Shortleaf Shortleaf provides the majority of stocking but is rare except in young stands or on very dry sites 3 Lob Hdwd Loblolly is not dominant but provides 20 percent or more of the stocking in mixture with hardwoods 4 Short Oak Shortleaf is not dominant but provides 20 percent or more of the stocking in mixture with oaks 5 Long Slash Longleaf and slash pine make up a majority of the stocking with slash being subordinate to longleaf 6 Longleaf Longleaf pine is pure or comprises a majority of the trees in the overstory 7 Slash Slash pine is pure or comprises a majority of the stocking 8 Long Scrub Longleaf pine and scrub oaks comprise this typically younger type 9 Slash Hdwd Slash pine and a variable mixture of hardwoods comprise the majority of the stocking 10 ERC Hdwd Eastern redcedar and mixed hardwoods comprise the majority of stocking with varying proportions 11 ERC Eastern redcedar is pure or comprises the majority of the stocking 12 S Scruboak Type consists of a mixture of scrub oaks where fire has been excluded n
9. data identified forested areas and assessed the age and composition of the areas Roller 2000 Forest area mapping was based on spectral categorization of the leaf on 1998 Thematic Mapper TM data Forest age was assessed using a first and last date forest stratified hybrid change detection procedure developed by Veridian ERIM International This procedure yielded six forest age classes based on the available LandSat Multispectral Scanner MSS and TM data Forest composition assessments were based on 1999 leaf off TM data The assessment was performed using a forest and age stratified spectral based procedure Forest composition was assessed with respect to the relative amount of evergreen and deciduous forest present within a pixel Inventory and Plot Allocation Sample Size Estimation The initial estimate of sample size for each of the four counties in the pilot project was determined from information provided by the 1992 FIA inventory of these counties The variability calculated for total cubic foot volume within a county was used to estimate the number of plots needed to meet an allowable sampling error of 10 at the 95 confidence level The sample size formula for simple random sampling rather than the formula for stratified sampling was used The former provided a more conservative estimate given some doubts about applicability of the information gleaned from the FIA database The resulting estimates of sample size varied from a high
10. of 98 plots for Choctaw County to a low of 79 plots for Oktibbeha County Again based on the conservative approach regarding the preliminary database it was decided that 100 plots would be allocated to each county Plans were also made to follow this initial allocation with the establishment of additional plots as necessary once each county s volume data were analyzed The final plot allocation was 695 plots in the four counties Plots were allocated proportionally based upon area in each cover type by county ERDAS Imagine software was utilized to randomly choose each plot location In an effort to allow for image geometric correctional errors and the known capabilities of the Differential Global Positioning Systems DGPS the criterior for choosing these plots was that each must be centered in a homogeneous 180 by 180 m area Geographic coordinates projected in Mississippi Transverse Mercator MSTM for the center of each area served as plot centers These coordinates were re projected into latitude longitude and formatted for the DGPS field units Field Equipment GPS Systems The Corvallis Microtechnology Incorporated CMT PC5L GPS and the Juniper Systems Landmark GPS were utilized to navigate to 695 random plot locations across the four county area Figure 2 Equipment failure either from physical damage to cabling or malfunction accounted for the largest percentage of reduction in field productivity Selective Availability GPS sa
11. plot overlaps the boundary of a seed tree and clear cut stand The plot should be moved into the seed tree stand 4 A plot overlaps the boundary of a pine and mixed pine hardwood stand The plot should be moved into the mixed pine hardwood stand 5 A plot includes area converted to a non forestry application e g agriculture roads etc i Inaccessibility of plot center is not a reason for re positioning Plot locations may fall within or include watercourses or natural catchments If the plot cannot be established without jeopardizing the safety of the crew then a notation must be made and the field supervisor notified Plot Description Each plot sampled will be identified with a unique number The plot number will index all data entered into the field recorder In addition to plot number descriptive information for each plot will be entered Slope Size Class Origin SAF Type Ownership Physiographic Position Logging Operability and Stand Condition will be catalogued within each plot Slope Identifies the average slope for the plot Slope observations are made with the hypsometer on a line parallel to the slope Procedures for obtaining slope measurements are detailed in Appendix 7 Stand Condition Describes the current condition of the stand with respect to damage and harvesting activi ties Explicit codes are defined in table 1 Stand level damage categories descriptions and data code Damage Type Descrip
12. route to a new plot location The procedures for calibrating each instrument are included in Appendix 5 Because both instruments can utilize either transponder it is more efficient to calibrate both instruments together The stipulation with calibration is that the unit has time to adjust to ambient temperature and humidity conditions before calibrating This is pertinent whenever the unit is initialized at the beginning of the day or has been in a vehicle for extended periods of time Allow the units to adjust to environmental conditions for 10 minutes before calibration Plot Establishment The first determination to be made upon arrival at plot center is whether the plot will encompass a homoge nous sampling unit If there is no significant variation in type size or age class then plot center is established at the point navigated to with the GPS unit If significant variation is encountered the plot center must be moved 1 chain from the original plot center perpendicular to the stand line into the stand occu pying a majority of the plot Micro pockets less than 10 acres within a contiguous stand are considered to be part of the surrounding stand Examples where plot locations should be moved 1 A plot overlaps the type boundary of a plantation and natural pine stand The plot should be moved into the natural stand 2 A plot overlaps the boundary of a thinned and unthinned stand The plot should be moved into the unthinned stand 3 A
13. screen d window view of tree tally screen e plot information screen and f sample tree screen Field Training Training and certification plots were established to represent the possible plot conditions encountered by field crews conducting the inventory Training plots demonstrated various conditions that required a qualita tive measurement or a subjective decision to be made The Mississippi Forestry Commission assigned 13 people to assist with data collection Junior and senior level undergraduates from the College of Forest Resources were also employed to collect the field data Two person inventory crews were trained on the certification plots for four consecutive days in order to be certi fied competent with the equipment and inventory procedures Certification procedures required that an inventory crew navigate to and establish an inventory control plot within specified data collection criteria of time and precision The inventory data were then printed and the crew and field supervisor returned to the plot If there were no errors associated with qualitative or subjec tive criteria detected the crew was certified Verification Once data collection began the integrity of the inventory data became the primary priority Inventory data were downloaded and screened after each day of collection It was discovered that the Corvallis Micro Technology PC5L field computer ram disks were subject to radio frequency interference generated by
14. that location was moved from the preset Before exiting the GPS software make a notation of the new corrected Latitude Longitude of plot center The new coordinates must be entered into these fields if the plot center was moved from the preset navigation point Plot Measurements Three concentric overlapping plot sizes are incorpo rated within each plot A 1 5th 0 20 acre plot with a radius of 52 7 feet is utilized to sample merchantable timber 4 6 inches in d b h and larger A 1 20th 0 05 acre plot with a radius of 26 3 feet is used to inventory sub merchantable trees between 1 and 4 5 inches in d b h A 1 100th 0 01 acre plot with a radius of 11 8 feet is used to record regeneration 0 0 to 1 inch in d b h Refer to figure 1 for a representation of the plot design Two sweeps will be made around the plot One sweep is made for merchantable and sub merchantable stems The other sweep is made within the regenera tion plot Regeneration plots are only established on locations where the Size Class is designated Reproduction Sweeps will start on the radial line with a north azimuth to ensure there are no dupli cate tallies 1 100th acre plot radius 1 20th acre plot radius 1 5th acre plot radius Figure 1 Plot layout depicting effective radii Individual Tree Tally Each live tree will have the following observations recorded Product species d b h total height pulpwood height to an abs
15. volumes are reported by the associated species group The current USDA Forest Service definitions of forest type species group associations were utilized A full implementation of the prototype is expected to be completed within the next two years Tasks include the addition of a large number of growth and yield databases further refinement of predictive algorithms and formulation of reliability estimates and their comparison to tradi tional modeling techniques nomen hecurecy Assessment The accuracy assessment of the remote sensing products supplied by Veridian ERIM International were completed using the field plot data historical aerial photographs and current high resolution imagery The field plot data and high resolution imagery were used to evaluate the forest cover type products Historical aerial photographs were used to assess the age class products This accuracy assessment was performed by the Spatial Information Technologies Laboratory of Mississippi State University Cover Type Assessments Overall accuracy based on the high resolution aerial imagery was estimated to be 93 54 Table 2 while the accuracy estimated from the field plot evaluations was 64 87 Table 3 The differences in these figures can be attributed to how each evaluation was taken In the field the assessment is made from under the canopy looking up Assessments from the high resolution imagery are made from above the canopy looking down Anothe
16. Mississippi s SEE Orest mventory Pilot Program Use of Computer and Spatial Technologies in Large Area Inventories By Robert C Parker Patrick A Glass H Alexis Londo David L Evans Keith L Belli Thomas G Matney and Emily B Schultz Forest and Wildlife Research Center Mississippi State University The Forest and Wildlife Research Center at Mississippi State University was established by the Mississippi Legislature with the passage of the renewable Natural Resources Research Act of 1994 The mission of the Center is to conduct research and technical assistance programs relevant to the efficient management and utilization of the forest wildlife and fisheries of the state and region and the protection and enhancement of the natural environment associated with these resources The FWRC scientists conduct this research in laboratories and forests administered by the University and cooperating agencies and industries throughout the country Research results are made available to potential users through the University s educational program and through Center publications such as this which are directed as appropriate to forest landowners and managers manufacturers and users of forest products leaders of government and industry the scientific community and the general public Dr Bob L Karr is interim director of the Forest and Wildlife Research Center Authors Robert C Parker and Emily Schultz are asso
17. a For Sci 37 6 1497 1512 Van Deusen P C 1997 Annual forest inventory statistical concepts with emphasis on multiple imputation Can J For Res 27 379 384 17 Fi PE M T Ip rocedures 18 Pilot Program for Forest Monitoring and Information System Volume 1 written by Patrick A Glass Revised October 21 1999 Pilot Program for a Forest Monitoring and Information System Measurement Plot Specifications and Procedures 2325 45 eid Fh nhs es 1 CU EUIS UU RED 1 Feld Equip ment 1 Land Navigation hee Pa e ag DOW rue aUe a 2 Equipment Calibration 283 2 Plot Establishment oua 2 Plot ID esScri OM okie tee et Aon LE UM M A M ag oie 3 ete E LB Ne E OR yt EET 3 Stand bse ee 3 LABS so tede EE avete nri oe 3 5 a aa 4 SAT OUB ek Li de eae d diua qa VAT Ronan ta cet hae Th d oua qao 4 SAF
18. affiliation age disability or veteran status 1
19. aged The tree displays no apparent damage 1 Insect Infestation of the tree as evidenced by pitch tubes bore holes webbed foliage etc 2 Disease Infection of the tree as evidenced by brooming cankers conks etc 3 Fusiform Infection of the tree by Cronartium Fusiforme 4 Fire Damage to the bole and or lower canopy from heat excluding typical damage from prescribed burns 5 Storm Damage by excessive winds from hurricanes or tornadoes is prevalent blowdown windthrow 6 Ice Damage to the upper stem and crown caused by ice accumulation breakage in upper crown 7 Mechanical Damage caused by human activity 12 Decay Decay of portions of the tree due to exposure 13 Lightning Direct strikes evidenced by spiraling scars on the stem 14 Erosion Exposure of root system by flowing water or wind removal of soil 15 Broken Top The stem diameter where the break occurs For broadleaf species this is defined where the largest diameter stem has been broken Field Procedures Upon exiting the vehicle turn the G P S unit on enter G P S software if not resident by typing GPS at the prompt Select Navigation from the main menu and allow the unit to initialize During the time of initial ization the hypsometer and DME 201 s will be calibrated After initialization the primary navigation screen will appear Select the route containing the plot locations then select the way point plot to be sampled The unit will then show the azimu
20. areas Schreuder and Thomas 1991 and is accurate for assessing the forest resources of a region or state The last inventory data available for Mississippi were collected in 1993 and 1994 and included 3190 plots Reams and Van Deusen 1999 The existing federal data collected were not designed to be statistically precise at the county level Current inventories also do not describe the distribu tion of the forests in the state The objectives of the Mississippi Forest Inventory Pilot Program were 1 To develop a forest inventory and information distribution system based on remote sensing GIS and GPS technologies in concert with widely accepted field data collection and computation techniques that can be extended to a statewide system and provide county level timber inventory information at a sampling error of 10 at the 95 level of confidence 2 To make original and derived data products available with on line technologies that will allow users to estimate timber supplies and monitor forest resource change ExbSrimental Design TU 5 di 7 This research was conducted a contiguous four region the east central part of Tx Eu involved the counties of Choctaw Clay Oktibbeha and Winston Figure 1 LII m The pilot study was accomplished through use LandSat satellite imagery in conjunction with a field inventory of forest resources for a multi cou
21. ata on a county or regional level Figures 7 and 8 All information from the field plots is entered into a database and summarized depending on the user s needs Irregular polygons or county boundaries can be used to define an interest area Embedded in this system are routines that generate stand and stock tables per acre values and totals are given with the associated precision statistics for the area and volume estimates The four report dynamic link libraries to be included with the release version of MsFMIS will allow the client to derive virtually any volume acreage reproduction or growth estimates showing point and relia bility statistics For report types that are needed on a continuing basis custom programming will be added to make the report generation a one step process The report generation and database application interfaces APIs will be published so clients can customize the generated reports to suit their needs ci ieee TEH ee 2d LINE p E vau F Lars amd Dele i Figure 8 Example screen from MsFMIS depicting the selection of an irregular polygon to define an area from which to generate inventory information based on other criteria indicated by check boxes Based on Veridian ERIM International products clients can view the spatial distribution of forest age and compositi
22. ate golf courses etc 8 Tribal Acreage that is owned or operated by federally recognized Native American tribes including Choctaw Chickasaw etc 9 Unknown Acreage that ownership cannot be readily identified 10 Physiographic Position Classifies the geographic position of the plot Table 6 supplies descriptions that will define relative position Strict definitions are not applicable to identify physiographic position because of the variability experienced state wide Topographic position categories category description and data codes Physiography Description Code Upland Drier xeric sites found on top of ridges and side slopes 1 Bottom Wet hydric sites found along rivers and streams 2 Terrace Mesic sites that by default are not upland or bottom 3 Logging Operability Classifies the terrain for season and equipment accessibility Table 7 lists the appli cable designators for logging operability Logging operability categories category description and data code Logging Operability Description Year Round Accessible to logging equipment throughout the calendar year Summer Fall Accessible to logging equipment during these seasons only Crawler Soil or slope precludes the use of rubber tired skidders High Lead Cable harvesting systems are required Inoperable Cannot be harvested economically or efficiently Code WH nm Latitude Longitude Identifies the plot center location in the event
23. becomes un usable in typical industry applications 1 2 3 4 This height will always be less than or equal to the height to an absolute top nearest foot This height is also recorded for both pulpwood and saw timber trees Examples of usability limits are given below The stem forks at 5 inches DOB The stem forks and none of the multiple stems are 5 inches DOB The stem is broken The stem has crook and there is not 5 5 feet of stem pulpwood above the crook Saw timber Height to Absolute Top For conifer species the height where the main stem is 6 inches DOB for broadleaf species the height where the main stem is 10 inches DOB nearest foot Saw timber Usable Height The height at which the stem becomes un usable in typical industry application 1 2 3 4 5 6 This height will always be less than or equal to height to an absolute top nearest foot Examples of usability limits are given below The stem forks and none of the multiple stems meet minimum saw timber diameter The stem incurs crook The stem is broken There is a branch whorl of 3 or more branches A branch is less than 45 degrees off vertical and is one third the stem diameter where it branches A branch is equal to one half the stem diameter Damage Table 9 provides damage categories that will be used to relate tree level damage assessments Tree level damage categories descriptions and data code Damage Type Description Code Undam
24. ciate professors in the Department of Forestry David L Evans Keith L Belli and Thomas G Matney are professors in the Department of Forestry H Alexis Londo is a Research Associate II in the Department of Forestry Patrick A Glass is Director of Operations for the Mississippi Institute for Forest Inventory To Order Copies Copies of this and other Forest and Wildlife Research Center publications are available from Publications Office Forest and Wildlife Research Center Box 9680 Mississippi State MS 39762 9680 Please indicate author s title and publication number if known Publications are also available at our web site at www cfr msstate edu Citation Parker R C Glass Londo D L Evans Belli T G Matney Schultz 2005 Mississippi s forest inventory pilot program Use of computer and spatial technologies in large area inventories Forest and Wildlife Research Center Bulletin FO 274 Mississippi State University 43pp Research Bulletin FO 274 FOREST AND WILDLIFE RESEARCH CENTER Mississippi State University MississipBiS Forest inysntery Pilot Program Use of Computer and Spatial Technologies in Large Area Inventories By Robert C Parker Patrick A Glass H Alexis Londo David L Evans Keith L Belli Thomas G Matney and Emily B Schultz Forest and Wildlife Research Center Mississippi State University Rilsiract The objectives of the Mississ
25. d using historical aerial photographs The only available statistic for this assessment was the overall accuracy for each age class Table 4 These accuracies range from 9096 to 9496 Again these were assessed from above the canopy looking down Because of the temporal discrepancies between LandSat and the photographs judgments about stand age were assessed from the original imagery For example if a stand was harvested in the middle of the age class the photographs would show a mature stand of timber however by comparing the two satellite images from the beginning and end of an age class a determination could be made 19 Accuracies for age classes assessed using historical aerial photographs Age Class Accuracy Years Years Included Year of Photography 0 3 1996 1999 1996 90 4 6 1995 1993 1992 92 7 12 1992 1987 1985 92 13 9 1986 1980 1980 94 20 26 1979 1973 1969 94 27 Prior to 1973 1969 96 Overall Accuracy Overall the accuracy of the remote sensing products was within acceptable limits The difference between classification techniques using aerial imagery and field data techniques generally show up as differences in the comparative statistics As image processing capabilities and the resolution of satellite imagery increase these products will be more reliable Conclusions ___ ant Future Directions Integration of remote sensing data with GPS technologies is valuable for increasing inventory efficiency and level o
26. e enter key Select menu item to start recording inventory data When asked for job name enter your file designator 01 When asked about previous specs press followed by enter Enter the plot number and press enter Press the key to access the PLOT INFORMATION screen Press N followed with enter Input the appropriate data into the various fields as required Remember to record new plot location coordinates if plot center was moved When all data have been entered for plot description then press F4 to return to the individual tree tally screen Starting from the radial line with a north azimuth proceed in a clockwise direction to record merchantable and sub merchantable tree data The first two trees tallied are to be nondestructively marked for subsequent height measurements Borderline trees are to be tallied if the distance from the center of the stem to plot center is less than or equal to plot limiting distance Classify the tree into the highest product category for which it qualifies Identify the species of the tree if possible If species identification is not possible or the species is not included in the species list then use the general codes 1 8 16 58 59 to identify the tree If genus identification is not possible then classify the species as unknown Measure the diameter of the tree to the nearest tenth of an inch If the tree qualifies as pulpwood then estimate the total height absolute merchantab
27. ed cover types There are 35 separate cover types that can be encountered For the inventory the definitive guide is Forest Cover Types of the United States and Canada published by the SAF 1980 Abbreviated descriptions are included with the code designations in Appendix 2 Ownership Identifies the primary ownership entity Identification of ownership is not paramount to performance of the inventory Minimal time should be spent discerning this variable If ownership is obvious then identification should be made Table 5 depicts the 10 categories that will describe all ownership possibilities Land ownership categories category description and data codes Ownership Description Code Private Non Industrial Acreage that does not appear to be intensively managed for forestry 1 Industrial Acreage that has the appearance of intensive management for forestry 2 USFS Acreage that is owned or operated by the USDA Forest Service 3 USFWS Acreage that is owned or operated by the USDI Fish and Wildlife Service 4 Other Fed Acreage that is owned or operated by any other federal entity including BLM National Park Service etc 5 State Gov Acreage that is owned or operated by the State including state parks 16th section land etc 6 Municipal Acreage that is owned or operated by municipalities including city parks public golf courses etc 7 Urban Acreage that is owned or dwelled upon for residential purposes including subdivisions priv
28. ees and measurements are taken at d b h If the stem visible separation forks at or above d b h then the stem is treated as a single tree and the diameter is taken at the base of the fissure scar If the tree exhibits exces sive butt swell as usually encountered with cypress and gum species then the diameter is measured 1 foot above the point where the stem resumes normal form Parasitic vegetation clinging vines etc should be removed to assure that only the tree diameter is measured Total Height The height of the tree nearest foot to meristematic tip or uppermost leader Most conifers will exhibit one stem terminating with an apical meristem If the main stem is intact then height is measured to the apical tip of the tree If the main stem has been damaged then height is measured to the point where the breakage occurs Broadleaf trees typically do not possess a single upper stem They will however usually have a dominant leader If a dominant leader can be identified then height is meas ured to its tip If the top has been broken out then height is measured to the point of breakage on the largest diameter limb Pulpwood Height to Absolute Top For conifer species the height where the main stem is 3 inches DOB for broadleaf species the height where the main stem or largest diameter limb is 4 inches DOB nearest foot This height is recorded for both pulpwood and saw timber trees Pulpwood Usable Height The height at which the stem
29. erchantable No commercial tree species greater than 4 5 inches in d b h are encountered Pulpwood Saw timber Non Timber within the radius of a 1 5th acre plot The majority of commercial tree species occupying the 1 5th acre plot are 4 6 to 10 6 inches in d b h The majority of commercial tree species occupying the 1 5th acre plot are greater than 10 6 inches in d b h The site has been converted to a non forestry application Origin Identifies the origin of the stand Table 3 identifies acceptable origin codes Origin regeneration method categories category description and data codes Origin Description Natural Regeneration occurred from natural processes Planted Undetermined Regeneration method is not readily apparent Regeneration occurred from planted stock 2 3 Code Code 1 SAT Code Identifies the canopy coverage in a broader perspective than the SAF cover type Table 4 provides guidelines for defining SAT codes SAT categories category description and data codes SAT Codes Description Pine Canopy composition is greater than 80 coniferous Pine Hdwd Canopy composition is between 60 and 80 coniferous Mixed Canopy composition is between 40 and 60 coniferous Hdwd Pine Canopy composition is between 20 and 40 coniferous Hdwd Canopy composition is less than 20 coniferous Code amp WH eR SAF Cover Type Identifies the canopy species composition using SAF accept
30. est age and forest composition information resulted in relating age to composition for any particular forested area This information was used in assessing the sustainability of the forested resources in the study area Results of Harvesting Sustainability The average annualized rate of forested harvesting is summarized in Table 1 An increase in forest The average annualized rate of forest harvesting in the early 1980 s and the last half of the harvesting by observer time interval 1990 s was evident This information was based upon 1 the forested area at the beginning of the Time Interval Percent Forest Area Harvested Per Year study period 2 the amount of harvested area 1972 1979 0 32 detected throughout the study and 3 an assumed 1979 1986 0 83 harvest rotation of 40 years standard at the begin a 1986 1992 0 17 ning of study Based upon this information the forest harvesting is occurring at a sustainable rate 1992 1995 0 13 2 5 per year in the study area This is an average 1995 1998 0 88 annualized rate of forest harvesting by area of 0 54 These observations relate to clear tell harvest systems and do not reflect intermediate thinning during the full rotation of a timber stand 11 Knowledge based prototype for estimating growth yield A prototype for a knowledge based growth and yield KBGY prediction system was developed to supply managed stand growth projections and planning and decision sup
31. f precision The benefits of this approach are two fold first no effort was expended on sampling non forested areas and secondly sampling intensity was matched with the variability of the target resource This coupling allowed for a true random sample of forest area by type over a large landscape Based on the remote sensing forest typing precision goals were attained Future directions for this product include validation expansion and integration with other programs Expansion goals are aimed at applications in other states and regions Finally the integration of FIA data will further increase sampling efficiency Cochran 2002 16 literature HSn r o ted Cochran J K 2002 County level inventories can FIA help Thesis Department of Forestry Mississippi State University Hartsell A and J London 1995 Forest statistics for Mississippi counties 1994 USDA Forest Service So For Exp Stn Res Bull SO 190 89 p Munn I and S H Bullard 1997 Pulp and paper industries multi billion dollar contributions to Mississippi s economy Tree Talk 19 2 12 13 Roller N 2000 Intermediate multispectral satellite sensors J For 98 6 32 35 Reams G A and P C Van Deusen 1999 The southern annual forest inventory system J of Ag Bio and Envir Stat 4 4 346 360 Schreuder H T and C E Thomas 1991 Establishing cause effect relationships using forest survey dat
32. g and press Enter A bar graphic and the estimated charge percentage are displayed Press F3 to confirm the setting then press F5 to return to normal operation Appendix 7 Slope Determination Activate the transponder on the plot center stake Move parallel with the slope to a point on the circumfer ence of the 1 5th acre plot Activate the hypsometer and verify the radial distance to plot center Then shoot a height on the top of the transponder The slope will show up on the screen directly under the angle label If the slope is less than 100 there will be a percent designator with the reading If the slope is greater than 100 there will not be a percent designator with the reading Data Recording Sheets Individual Tree Data Crew Designator Date Lat Long N T gt T gt iz Crew Designator Plot Ave Slope Stand Code Size Class Origin Code Sat Code SAF Type Ownership Physio Pos Logging Oper Latitude Longitude Growth Data SP DBH TOTH AGE Site FWRC Forest and Wildlife Research Center Mississippi State University Mississippi State UNIVERSITY Mississippi State University does not discriminate on the basis of race color religion national origin sex sexual orientation or group
33. h instruments utilize ultrasonic frequencies Only one transponder can be in operation at any given time If both transponders are in operation measurements will be invalid Calibration Hypsometer Activate the transponder and tape off a distance of 33 feet Press and hold the On Off button After the unit activates the display will show Auto Distance Press and hold the Undo button until the display shows the distance The distance will replace a series of question marks When the calibration distance is shown the unit is calibrated DME 201 Activate the transponder and tape off a distance of 33 feet Repeatedly press the red dot until the number 9 appears in the display The unit will display 4 dashes followed by 2 dashes When the calibration distance appears in the display the unit is calibrated Auxiliary CMT Operations Screen Contrast Screen contrast can be adjusted any time the unit is in operation Press the SH1 key then press the Enter key to access the unit setup screen Scroll to the contrast setting and press Enter Adjust the contrast with the left and right arrow keys until visibility is appropriate for current light conditions Press F3 to confirm the setting then press F5 to return to normal operation Power Supply Remaining battery power can be checked any time the unit is in operation Press the SH1 key then press the Enter key to access the unit setup screen Scroll to the Powermeter settin
34. he majority of stocking Willow oak water oak and laurel oak in varying proportions comprise a majority of the stocking Live oak is pure or the majority of stocking Cypress is pure or comprises a majority of the stocking Tupelo is pure or constitutes a majority of the stocking Cottonwood is pure or comprises a majority of the stocking Sycamore sweetgum and elm together comprise a majority of the stocking with varying proportions Willow species comprise a majority of the stocking Naturally occurring land areas excluding agricultural lands and pastures that are devoid of tree species Code 25 26 27 28 29 30 31 32 33 34 35 Appendix 3 Power Supply G P S The system consists of three separately powered integrated components The field recorder has an internal power supply the GPS receiver utilizes an external battery physically attached to the field recorder the external antenna utilizes a battery and signal decoder carried in the backpack frame The signal decoder attaches to the GPS receiver with a DB 9 connector on the COM 2 port It is expected that the equipment will operate for 10 hours on a single charge To insure maximum battery life turn off the satellite decoder after establishing plot center Turn all other equipment off when not in use Therefore it is necessary that the equipment be recharged every evening Procedure cards will be supplied to explain the necessary connec tions for recha
35. ices If site index is supplied predictions will be based only on growth and yield plots possessing that site index The database update facility was developed as a WIN32 program in Visual C This module facilitates the addition of growth and yield data sets to the database component of the knowledge based system As the database grows predictions can be made for an increasing number of forest types and management condi tions with greater confidence An ISAM indexed sequential access method database structure was utilized because of its superior access and retrieval speed as compared to relational database structures It is antici pated that the database component of the full implementation will be extremely large and retrieval time will be an important issue Microsoft s Open Database Connectivity ODBC standard for client side database access and a standard Structured Query Language SQL embedded in Visual C code were employed for writing the database access code necessary to construct the search engine and predictive algorithms 12 The report writer for the growth and yield system Figure 11 was designed to display volume information in a format as similar as possible to the main MsFMIS interface so as to appear as a seamless link Volumes are reported on a per acre basis for inside and outside bark estimates in cubic feet of pulpwood and sawtimber and for three different board foot log rules Depending upon the user selected forest type
36. ippi Forest Inventory Pilot Program were 1 to develop a forest inventory and infor mation distribution system based on remote sensing GIS and GPS technologies in concert with widely accepted field data collection and computation techniques that can be extended to a statewide system and provide county level timber inventory information at a sampling error of 10 and 2 to utilize computer technology to make original and derived data products available with on line technologies that will allow users to estimate timber supplies and monitor forest resource change Assessment of LandSat satellite imagery over successive time periods facilitated the stratification of forest areas by age class and forest cover type Other remote sensing technologies permitted the allocation of truly random plots using stratified random sampling criteria Allocation of sample plots was designed to achieve a 1096 sampling error at a county level for total cubic volume at the 95 confidence level within the Pilot Program area involving 4 counties of east central Mississippi GPS units coupled to a geostationary satellite were used to navigate on a real time basis to plot locations with sub meter accuracy Inventory data were recorded on field computers containing specialized software The satellite image estimates of forest acreage by age classes of conifers hardwoods and mixed conifer hardwoods were combined with the inventory data to obtain post stratified area and volume e
37. le pulpwood height and usable pulpwood height If a tree qualifies as saw timber Peelers Poles and Railroad Ties are considered saw timber for height measurements then estimate total height absolute merchantable pulpwood height usable pulpwood height absolute merchantable saw timber height and usable saw timber height If the tree is sub merchantable measure diameter and estimate total height If the tree is a cull then measure diameter only Perform an inventory of reproductive material only on plots for which the size class is designated as repro duction after all merchantable and sub merchantable stems have been tallied Reproduction size classes are entered into the TH field and the number of trees for that species size class combination are entered into the PHa field If there are more than 20 individuals encountered for any species size class combination then enter 20 and proceed with the next combination Height Sub Sample The first two trees that were tallied and marked are measured with the hypsometer for total absolute and usable heights The measurements are to made by the same individual that estimated the heights De activate the DME 201 transponder by removing it from the range pole Activate the hypsometer transponder by swiveling the pins outward from the casing After measuring d b h attach the transponder to the tree at d b h Move away from the tree in a direction that will allow the clearest line of sight for as much of
38. leaf and slash predominately found in southern coastal plain 6 Railroad Tie Hardwood species red and white oaks that are not suitable for saw timber but higher quality than pulpwood 7 Cull Any tree that has obvious defects that preclude its inclusion in any of the other product classifications i e large areas of decay very poor form etc 8 Height Sample The first two non cull trees merchantable and or sub merchantable tallied on each plot are carefully measured for all pertinent heights This sample is taken after the reproduction plot information is recorded Heights are measured by the same individual that estimated heights 9 Species A numeric code from 1 60 that identifies the species of the tree The predominant commercial species are assigned a code to be input Additionally genus codes are supplied in the event that species identification is not possible Appendix 1 supplies the code common name and species name Diameter The diameter of the stem at breast height in tenths of an inch Diameter is measured at 4 5 feet above ground level on the uphill side of the stem If there is a deformity at d b h then diameter is preferably measured 1 foot above the point where the stem resumes normal growth If this point is inac cessible then the diameter is measured 1 foot below the point where the abnormality occurs If the stem forks visible separation below d b h then each of the multiple stems are treated as separate tr
39. n handling DME Devices The Haglof Forester DME 201 Figure 3 was used to measure plot radii and horizontal distances for tree height measurement Haglof developed the DME 201 for use as a distance measuring device in areas where terrain is uniform It is a one handed receiving device that utilizes a transponder located at plot Figure 2 Juniper Pro4000 left fe ELLE center that resonates at 25 kHz and CMT PC5L Field computers with a 360 broadcast area with CMT GPS antenna Height Measuring Devices The Haglof Vertex and Vertex III hypsometers were used for tree height measurement Figure 4 They are one handed receiving devices that utilize a transponder located on the tree at some predetermined height typically breast height and resonate at 25 kHz with a 60 broadcast area to establish baseline distances Using the receiver transponder combination eliminates the need to measure baselines and the 60 broadcast angle for the transponder allows a wide range of motion for obtaining a clear line of sight The ultrasonic equipment demonstrated a high degree of precision Figure 3 Forestor DME 201 bottom operating through a wide range of environmental conditions with 360 transponders Measurement errors were obvious to opera tors and could be attributed to either a depleted power supply or multiple transponder interference DGPS and Navigation Procedures DGPS allowed the use of stratified random sampling proced
40. ndividual will complete all height estimations for that plot Field Equipment Each field crew will be assigned a complete set of field equipment In addition to equipment a backup system compass clipboard field data sheets and note pads will be accessible to prevent loss of time due to equipment failure A 5 receiver will be used to navigate to plot locations and a field recorder will be utilized to record inventory data Tree heights will be measured with a Vertex hypsometer The data will be used to obtain local volume tables and site index estimations on a subsample of trees on each plot Hypsometers will also be used to obtain a measurement of slope for each plot DME 201 units will be used to check radial distances for boundary trees Because of the nested plot design and diameter limited plot radii each tree s radial distance will be verified Two increment borers 12 inch and 18 inch will be issued to each crew The 18 inch borer will be held in reserve for large trees The borers will be used to obtain 5 and 10 year radial growth and age at breast height for site index A bark gauge will be used to determine single bark thickness Logger s tapes will be supplied for diameter at breast height d b h measurements They will also be used to measure distances in the event the DME 2015 or the hypsometers fail Clinometers will be carried as a precaution against equipment failure Spare batteries for the hypsometer and the
41. nt or co dominant oak tree is sampled If oak is not present then a dominant or co dominant tree from the predominant commercial species is sampled Record the species d b h age at breast height and utilizing the hypsometer total height If the tree that was selected for largest diameter is either a dominant or co dominant and representative of the predominate species then the growth and site index increment can be combined After all measurements are recorded press F4 to return to the tree tally screen Plot Exit All plot information should be collected at this time Press F4 to save the plot data press F5 to exit the plot tally entry routine and press enter to return to the main menu Press enter again to return to the prompt At the gt prompt type GPS to activate the G P S software if the next point is to be navigated to a plot or turn the unit off if returning to the vehicle Appendix 1 Common Name Species Reference Common Name Misc Pine Loblolly Shortleaf Longleaf Slash Cypress E Red Cedar Other Red Oak Cherrybark Oak Shumard Oak Black Oak Nuttall Oak Red Oak Blackjack Oak Turkey Oak Other White Oak White Oak Post Oak Swamp Chestnut Cow Oak Overcup Oak Durand Oak Water Oak Willow Oak Laurel Oak Chinkapin Oak Live Oak Sweetgum Ash Sycamore Yellow Tulip Poplar Tupelo Blackgum Sugarberry Hackberry Hickory Pecan Walnut Cottonwood Basswood Black Species Name Pinus spp
42. nty area Digital classifications of satellite data were used to allo cate field sample locations for sampling by project crews and cooperators from the Mississippi Forestry Commission All data were incorporated into an Internet World Wide Web site The primary methods included 1 Building a current forest type map for the four county area of the state 2 Augmenting the forest type map with ground inventory photo plots to generate a geospatial database that can be used to deter mine volume and distribution by timber type and ownership 3 Making the information developed from objectives one and two available to the public on an Internet site 4 Designing a knowledge based expert system for estimating the growth of the mapped forest types from existing growth and yield data and other sources 5 Designing a system that would combine spatial and resource data components into a visual online system that would allow the public to assess forest productivity for any user defined land area State of Mississippi Choctaw fom Oktibbeha M DE Winston Four County Study Area bp tad Figure 1 Four county region for the Forest Inventory Pilot Program POM elation Spatial Technologies Veridian ERIM International a contract partner geo corrected the LandSat
43. olute top usable pulpwood height saw timber height to an absolute top usable saw timber height apparent damage and broken top diameter if present Plot The unique number assigned to this particular set of Latitude Longitude coordinates Product A numeric code that describes the primary merchantability of the stem Table 8 defines the specifi cations for the product classes Product merchantability categories category description and data code Product Category Description Code Reproduction All commercial species that are less than 1 0 inch d b h These trees are tallied according to height class 1 Class 1 Commercial species less than 1 foot in height Class 2 Commercial species between 1 and 3 feet in height Class3 Commercial species greater than 3 feet in height Sub merchantable All commercial species that are between 1 and 4 5 inches d b h Total height for these trees is estimated to the nearest foot 2 Pulpwood All commercial species that are 4 6 inches d b h or greater that do not meet the qualifications for a product of greater value 3 Saw Timber Pine species that are 7 6 inches d b h or greater and broadleaf species that are 11 6 inches d b h or greater and of sufficient quality to be utilized as saw timber 4 Peeler Trees that are 15 6 inches d b h or greater and have 16 feet of clear butt log 5 Pole Trees that are 11 6 inches d b h or greater free of defect and sweep Restricted to pine species long
44. on within the study area Figure 9 The spatial distributions can be viewed using ESRI s ArcExplorer software ArcExplorer is freeware and it can be downloaded from www esri com T Inventory analysis with MsFMIS was accomplished through an interface that allows the user to define several criteria for compiling results The user has the ability to select from several inventory databases Currently the USDA USFS 1992 Southeastern FIA database and the Mississippi Pilot Inventory database are available for inventory calculations A unique feature of MsFMIS is the versatility in selecting the area for analysis An inter active map is provided that allows the selection of a single county or multiple counties or the selection of a geographic region by outlining the area of interest with a polygon The only restriction placed on the regional selection method is the total area within the polygon must be larger than 200 square miles This insures that an adequate number of sample plots are included in the computations After the region under consideration has been defined the user can then expand or restrict the analysis based on the attributes of interest Four separate reports will be available with the commercial version Area Regeneration Growth and Volume however the current version only reports timber volumes with standard stand and stock tables A typical application scenario for forest industry would be to determine whether an area
45. ot allowing longleaf regeneration 13 Oak Pine Various oaks dominate the stocking with pine comprising less than 20 percent 14 Oak Hick Oaks and hickories are prevalent throughout the stocking 15 PO Blkjck Post oak and blackjack oak comprise the majority of the stocking 16 White Oak White oak dominates the stocking with other components less than 20 percent 17 WO RO Hick White and red oaks dominate the canopy with stocking levels varying among the species 18 YP WO RO Yellow poplar white oak and red oak together comprise the majority of the stocking 19 SG YP Sweetgum and yellow poplar comprise the majority of the overstory and the stocking 20 O G C Oaks gums and cypress comprise the majority of the stocking with varying percentages of composition 21 SCO Chrybk Swamp chestnut oak and cherrybark oak constitute a majority of the stocking 22 SG WiO Sweetgum and willow oak comprise a majority of the stocking with sweetgum being dominant to willow oak 23 SB AE GA Sugarberry elm and ash constitute a majority of the stocking Sugarberry is replaced by hackberry in the northern part of the range 24 Cover Type Description OvO Hick Cyp Tup Swb STu Rb WiO WaO LO Live Oak Cyp Tup CotWood Syc SG AE Willow NonStock Overcup oak and hickory make up a majority of the stocking Cypress and tupelo together comprise the majority of the stocking Combinations of sweetbay tupelo and redbay with sweetbay dominant make up t
46. port into a database structure An expert system computer algorithm was used to screen the data collected by the field crews and create the database The algorithm examined the field data applied error checking rules set to detect and correct errors wrote the corrected data to the database and printed the original and corrected record for manual correction of the field data files This expert system utilized the same external files and settings as did the report gener ator for complete system compatibility and portability of the error checking facility to any region or scale Data Delivery Tools An interactive dynamic report generator called Mississippi Forest Monitoring and Inventory MsFMIS dynamic report generator was developed to synthesize and deliver information to system users The software can derive reports from an inventory database stored on a public server or optionally on a locally stored inventory database It allows the user to select a reporting area on either a county group basis Figure 7 or as an irregular polygon Figure 8 After a report type is selected the interface presents to the user changes to match the selected report type The user can then select the desired stand and tree selection criteria appro priate for the report species groups forest types land ownerships size classes products and others A beta version of the program can be downloaded from the Web site www cfr msstate edu To maintain compati bility
47. port tools for industry private landowners and public agencies The prediction system was constructed in five component modules 1 MS Windows user interface that will be linked to the main MsFMIS interface 2 a database update facility 3 the database 4 search engines and predictive algorithms and 5 a report writer The knowledge based growth and yield user interface Figure 10 was implemented in a Visual C e i a H ia ra e J Ls 1 1 Ez uim m Madenin 1 m manm Li z a Figure 10 Example screen from MsFMIS depicting the knowledge based growth and yield user interface programming environment The interface includes both required and optional user inputs The user selects forest type stand origin current and future age and report type for which growth and yield predictions will be made Either current basal area per acre or current trees per acre must also be entered Site index is an optional input that when known can further refine yield projections The system assumes depending on the type of user varying degrees of current stand information may be available for input The more inputs that are supplied the more specific the predictions that can be made For example if site index is not entered predictions will be broadly based across a range of site ind
48. r cause for error could be the registration errors between the remote sensing product and the DGPS Finally the errors between the cover types may be a result of the compilation of the five field plot cover type categories into the three cover type categories produced in the remote sensing products 13 Knowledge Based Reporter Forest Type z Current Age 6T Origin Wetureal Stand Future Age 87 Current Trees cre 120 Future Basal i 124 50 Fucure Trees 110 Predicted Yolumea Per Acre By Species Group Species Fu ET SI Total Tatal CEDU 10 2 1 1 b 12 2 11 2 1 Dayle Sotib Int 1 4 1151 Z lal 184 3 219 1 175 1 228 1 185 0 559 as 1058 Basel is in feet FH pulpuonund volume iz acre Feet fron stump to 3 inch top ET zawtimher volume is per arre cubic feet stump to d inch top Total total volume is cubic feet acre from stunp to O inch tup Doyle Scribnec and Intecnationsl 1 4 are boacd feet per acre Barlog trees ace defined 6 4 inches greater Pulpwood tree are defined as 4 4 inches or greater Species group codes ASH Balt Basan od CTU Cobtonwoad Exotic Beech BLW Black Walnut CYP Cpress Hickory LE YAH HE UT RD DU T 3W UI YP zEL W
49. rging Hypsometer The 60 transponder for the hypsometer has 9 volt battery for its power source The hypsometer uses two 1 5 volt AA batteries to power the unit To replace the battery in the transponder slide the battery cover out of the chamber replace the battery with the proper polarity and slide the cover back into place To replace the batteries in the receiver unscrew the large silver battery cover with a coin replace the batteries with the proper polarity and screw the battery cover back into place DME 201 Both the transponder and the handheld unit use a 9 volt battery for their power supplies To change the battery for the transponder unscrew the top of the unit disconnect the 9 volt battery and replace screw the unit back together To replace the battery in the receiver slide the battery cover completely off the unit replace the battery with the proper polarity replace the battery cover Appendix 4 Transponder Activation The transponder for the DME 201 is an orange cylinder that sits on top of a telescoping range pole with an activation mount To activate the transponder firmly seat it onto the range pole There is no visual indica tion that the transponder is activated The transponder for the hypsometer is a blue rectangular box To activate the transponder swivel the mounting pins located in recesses on the back outward fully There is a small red LED on the right side that blinks when the unit is active Bot
50. rom initial to most recent dates Determination of forest age was performed in two parts 1 aging of forested land that was detected in the first year in the sequence of the imagery and 2 aging of land that had become forested after the first year in the sequence Land that was forested in the first year in the sequence and never detected to be harvested while still being forested in the most recent image was determined to be in the oldest class Subsequent classes were aged from the last observed harvest or pre forested date determined from the intermediate image dates Forest Composition Forest composition was determined using the two most recent satellite image acquisi 10 a P UE om a po oo Tiy ET ai MP MES gt po 2 n TAE acu Er NE 1 1 Figure 9 Examples of forest age and composition derived for Choctaw County Mississippi tions Using the standard normalized difference vegetation index NDVI to detect the amount of evergreen versus deciduous trees in a forested pixel a continuous variable for all forested pixels was assessed Using ancillary photogrametric information and knowledge of the vegetation growth patterns in Mississippi it was possible to group the forested areas as pine mixed or hardwood forest Age and Composition Integration Integrating for
51. stimates by tree species forest cover types ownership class and or forest origin for user defined areas with user accessible Web site software Procedures developed in this pilot project can be used in larger land base applications and for integration with and supple mentation of existing inventory programs to meet user defined precision criteria Br duction Forestry is a leading industry in Mississippi Munn and Bullard 1997 Data exist for systematic inventory plots and county summaries Hartsell and London 1995 but there is no accurate representation of the spatial distri bution of timber volume resources Shifts in timber availability have dramatic impacts on the State s economy and environment Forest change and resource location impact Mississippi s forest based industries and activities they support There is no current system in place to track changes in the spatial distribution of harvesting resource depletion and successful regeneration future availability environmental recovery The United States Forest Service USFS currently conducts Forest Inventory and Analysis FIA surveys Reams and Van Deusen 1999 which are designed to inventory forest resources nationwide Historically these inventories have been scheduled every 6 to 15 years with a mode of 10 years Van Deusen 1997 Inventory plots are distributed systematically throughout a given region Information computed from these inventories is designed for large
52. tellite geometry and canopy coverage had minimal effect on productivity Field recorder power requirements were comparable providing 10 hours of intermittent usage without failure External power supplies for the DGPS decoding system proved to be bulky and cumbersome The CMT field delivery system consisting of a rigid frame backpack proved more efficient for transporting the additional inventory equipment than did the flexible backpack of the Juniper system Prior to deactivation of Selective Availability the rapidity of change in the pseudorange broadcast caused position fixes calculated by the CMT unit to fluctuate a known point by one to five meters in horizontal position When Selective Availability was turned off the CMT position fixes became more stable at less than two meters in horizontal offset The Juniper system which displays the position computed within the Trimble AG 132 receiver consistently reported positions around a known point within one meter regardless of Selective Availability influence When used under tree canopy conditions both the CMT and Juniper GPS systems performed best with the large Trimble dome antenna Field Computers The CMT PC5L and Juniper Pro4000 field computers Figure 2 were used with speciality software MSINVEN to record the tree and plot data and transfer the data to a desktop computer INVENCOM The software was written and compiled with Quick Basic 4 5 and contained assembler routines for scree
53. th and range to the plot location from your current position A directional aid is displayed on the navigation screen in the form of an arrow inside a circle next to a fixed arrow orientated with the top of the screen When the two arrows are parallel the unit is oriented in the direction of the plot Proceed in the direction indicated the unit is set to sound a proximity alarm and display a message when within 10 feet of plot center Utilizing the range indicator minimize the distance to the plot center Temporarily mark the plot center and assess for homogeneity of forest characteristics If the plot can be established at that point then exit the G P S software by pressing esc until questioned for confirmation of exit Press F5 to confirm exit If the plot has to be moved make the necessary determination for direction and move the plot When the plot center is established to contain homogenous stand conditions escape esc to the main menu the G P S software Scroll to the Collect Data menu item and press enter Press F3 to access the base data and when prompted for a file name press return After a few seconds the base location data screen will appear From this screen note the latitude and longitude for the new plot center It will be input during the plot inventory Press esc until returned to the main menu and then exit the G P S software Start the field recording software by typing PLOT at the prompt and pressing th
54. the radio units in the Mississippi Forestry Commission vehicles The impact of this interference ranged from scrambling some of the data fields to complete erasure of the file allocation table This necessitated that these field computers be turned off or located outside a radius of 30 feet during radio transmissions Validity of the data was corroborated by check cruising at a 2096 intensity Specially trained field crews were assigned check plots and collected data from these plots as specified by the procedure manual The two sets of inventory data were then compared and the original data were considered valid if discrepancies were within pre established limits To reduce the total amount of time spent collecting data on a plot stem lengths were ocularly estimated To detect and compensate for any bias in these estimations the first two trees encountered on a plot were measured using hypsometers after the ocularly estimated field tally was completed The corresponding trees were then paired by individual cruiser and correction curves were developed and applied to the estimated lengths of all trees on the plot Data Format After collection and verification of the inventory data was completed the information was organized for the database Three separate data files were formed for the database general stand description information plot mensuration data and growth and site data These three files were in ASCII format and readily available for im
55. the tree as possible Distance from the tree should allow for less than a 45 angle from the horizontal when sighting the top of the tree i e distance equal to the height of the tree Sight the transponder with the hypsometer until the targeting dot disappears Then sight the top of the tree until the targeting dot disappears Continue to sight additional heights as necessary until all height measurements obtainable from that tree are complete Repeat the process for the second tree Under no circumstances are height subsample measure ments performed prior to height estimations for the plot tally Growth Projection Measurements Three merchantable and one sub merchantable tree will be sampled for growth at each plot The dominant species occupying the plot will determine the species of the sample trees The largest merchantable diameter tree smallest merchantable diameter tree and the merchantable tree closest to plot center excluding the largest and smallest diameter trees will be sampled A sub merchantable tree of the dominant species capable of supplying a 5 year radial growth increment will also be sampled To access the plot growth entry screen press F2 Verify that the plot number is correct For each tree record the species d b h single bark thickness and 5 and 10 year radial growth increment Site Index Measurements On pine sites a dominant or co dominant pine tree is sampled On hardwood sites if oak is present then a domina
56. tion Code Undamaged There is no apparent damage within the stand 1 Insect Infestation of the stand as evidenced by pitch tubes bore holes webbed foliage etc 2 Disease Infection of the stand as evidenced by brooming cankers conks etc 3 Fusiform Infection of the stand by Cronartium fusiforme 4 Fire Damage to the boles and or lower canopy from heat excluding typical damage from prescribed burns 5 Storm Damage by excessive winds from hurricanes or tornadoes is prevalent blowdown windthrow 6 Ice Damage to the upper stems and crowns caused by ice accumulation breakage in upper crown 7 Salvage Focused harvesting operations have occurred within the stand 8 Thinning Generalized harvesting has occurred throughout the stand 9 Clear cut The stand has recently been harvested and no routinely used method of regeneration is apparent 10 Seed Tree The stand has recently been harvested and seed trees for regeneration are present 11 Size Class Classifies the merchantability class for the plot based on overall size composition The predomi nate size class encountered on the plot defines the size class Table 2 lists acceptable size class designations Plot merchantability classes class description and data code for plot size class Size Class Class Description Reproduction No commercial tree species greater than 1 inch in d b h are encountered within the radius of a 1 5th acre plot excludes seed tree stands Sub M
57. ures After the satellite imagery was classified and the required number of plots within each stratum was allocated the coordinates for the plots were organized into navigation files for the DGPS receivers Two separate DGPS systems were used for navigation Figure 5 thus Figure 4 Vertex III left and Vertex right hypsometers requiring translation of the coordinate information from a standard text format into proprietary formats that the receivers would accept Inventory crews utilized image maps generated from the satellite data and standard topographic maps for navigation to the approxi mate plot location Within a range of one mile from the plot location the crews switched on the DGPS units and utilized the navigational aids provided by the interface to locate the plot Field and Tree Data Collection To assure consistent and uniform data collection across all encountered conditions a procedures manual was developed Appendix A Inventory information was collected in two phases to allow for a complete description of the forested area The first phase observations included the general characteristics of the site i e slope stand type etc to ensure the data represented a homogenous sampling unit Mensuration data were collected t during the second phase Measurements were collected for the Figure 5 CMT and Juniper left DGPS computation of tree volume inside and outside bark In addition field delivery systems
58. with previous FIA databases and have an interface that is familiar to potential clients the FIA codes and names were adopted for the study Currently the MsFMIS produces stand and stock tables acreage tables regeneration acreage and productive potential stand and stock tabular reports by the selected criteria Other dynamic link library dll report modules will be added as needed Rept alee Beg ord Ieee Barris 10 ao Ta TEH es H Sk Figure 7 Example screen from MsFMIS depicting region selection such as county boundaries upper left to define an area from which to generate inventory information based on other criteria indicated by check boxes While the beta version MsFMIS software applies only to the small pilot study area the program code was designed to be transportable to any scale or region All the inventory codes user interface settings GIS files file locations and other locale specific settings reside in external files or the system registry This externaliza tion of locale specific settings allows the program to be adapted to another region by simply modifying the external settings No changes in the source code of the program are required An interactive Web based data delivery system was created using the FIA data standards Developed in Microsoft Windows interface format the data delivery system is designed to provide forest resource d
59. would be capable of supporting an additional milling facility Their analysis objective would be to determine the total volume available that is capable of meeting milling specifications Utilizing MsFMIS the region of interest would be identified and the generalized criteria Species Forest Type Ownership Stand Size Origin Tree Product Logging Operability Tree and Stand Level Damages and Physiographics would be designated These selec tion criteria would then be used to extract and process the inventory information from the database to produce standardized volume reports Additionally MsFMIS computes the associated statistics and sampling errors for each density and volume category Particular attention should be paid to the sampling errors asso ciated with the various density and volume estimates The sampling error presented represents the combined variance structure from the acreage estimate and the volume estimate Imagery based Products Forest Cover Mapping Forest cover maps were produced for the first and last date imagery Forest cover mapping was developed using all spectral bands from the MSS and TM imagery Spectral signatures of forest cover from the imagery were derived using unsupervised clustering Maximum likelihood categorization was used to label the clusters which were based upon phenological image changes and ancillary photogram metric information Forest Age Forest age was determined using the time sequence of images f
Download Pdf Manuals
Related Search