Home

User's Guide for SIRENA II - Denis Alder

1. eeeeeeeeeeeeeeeeeeee eene 20 The general options GialOG 2 0 ca iieri acd ened irent tt nn dee qui enne eta nen eaa gira 20 Sensitivity analysis 35i aint eo crar er ri e e E ab ere qu e nu dte Ra t cea eee 22 VOlUMeOplIONS EE 23 Output graphs uoi et epe coun spa dM CER ep ER MAR E EE Sn ne AD se ED ADR ARR E MEE 24 The species abundance graph ssssssssssssssssesse eene nennen enne nnns 24 Size class disStriDUtiON istrada OLLI 25 Volume by species groups over time sssssssssseee nennen enne 25 Harvested volumes 3 esp t n pa rin edens Sa t prdet ce iste 26 ii BaSalareacdynamics s cust ee fae eas dap E ae eet ee a p ea OR 27 Comparison of scernarioS 1p edere Rd Pep deen P eto deer aeter TERR ee ede 28 Output tables 2 EIE 30 Table of species by diameter classes sssssssssssseeee eene nennen nnns 30 The yield tables 31 Executive menu functions ueeeeeeeeessseseseseeeeeee eene nennen nnn 32 Running pausing and cancelling simulations ess 32 Exporting and printing outputs sssssssssssssssseeseeeneenen enne rennen nnne rre 32 Exiting from Sirena 2 c detnr de eoe a Ded e ade ett een tra Best rer LYRA eet det dea ead 32 Technical basis of the model eeeeeeeesessseeseeeeeeeeeeeeeennennnn 33 IALFOCUGUOM je CREER 33
2. ssssssssseeeneee enne nnn 9 Species groups for forest management eeeeeeeeeeeeeeeeeee 10 General considerations ssssssssssssssssssssssseseeenenenne n ener nnn nne EEEE nnne nnne EEEE EEEE 10 Managing species groups in SIRENA ssssssssssssseeeeeenenme eene nnne nnn 10 Specification of harvesting operations eeeeeeeeeeeeeeseeeeee 12 The Aprovechamiento dialog Arsise aniar aaan EENAA nemen nnnm 12 Scheduling and controls on the general intensity of harvesting esesss 13 Specifications for the selection of trees cccceecceceeccecceceeeeeeeceneaeeeeeeesesecueaeeeeeeeeeeseesnneeeeees 14 Special Options rr iaar Ea 15 Specification of silvicultural treatments ccccceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeees 16 tThe dialog Screen n reto fece e etr ee eter ped ta dea eet er ee 16 Timing of treatments eir ete erectae cerea md tear cede e uote 16 SPEClal OptlOris 1d nita dent te ee eR ees daa exi eer neds Orden Ge endo 17 Specifications for trees to be treated ssssssssssssssseseeeeeeee eren 17 The scenario MANAGER coo eere eec ere one apex ee Gaeta oe x eue E nnmnnn nnmnnn nenne 18 Concepts and definitions iii et tt ei een iste uten ite etta celi Heb iere 18 Using the scenario marnaget oit eei ict bu are oo decet etucet aaa EAAS 18 Setting general program options
3. 40 despu s tratamiento utilizar escenario Nocomercial 30 S D Solo rboles defectuosos ser n tratados Aceptar z Anular Timing of treatments Figure 7 shows the layout of the screen There are three checkboxes and associated value fields related to the timing of treatment Checking the Tratamiento inicial box causes an initial treatment to be done in the specified year irrespective of whether the stand has been harvested or not If this box is left clear as in the example then treatment is done only after harvesting Treatment after each harvesting is specified by the second checkbox The period in years to elapse after each harvest is given in the field to the right of the label Tratamiento despu s aprovechamiento If this is zero treatment is done in the same year as harvesting Treatments may also be repeated within a felling cycle by checking the third box labelled Repetir tratamiento despu s A value here of 5 for example would repeat treatment to the same specifications every 5 years 16 Special options The general options controlling treatment indicated in the group of buttons at the top right of the screen Figure 7 are the same as for the harvesting screen The first button provides a quick method of switching off treatment without having to edit the screen in detail The second button is for normal operation The third causes the program to pause after each treatment I
4. Tabebuia rosea Virola koschnyi Qualea paraensis Terminalia amazonia Cordia alliodora Laetia procera Minquartia guianensis Grias cauliflora Tapirira guianensis Simira maxoni Croton killipianus Protium panamense Ferdinandusa panamensis Brosimum rubescens Mosquitoxylon jamaicense Croton smithianus Dendropanax gonatopodus Ochroma lagopus Cecropia obtusifolia Casearia arborea Henriettea odorata Inga thibaudiana Eschweilera costaricensis Cordia porcata Miconia punctata Chimarrhys latifolia Chione costarricense Lonchocarpus guatema ANOaKRWBN O ROON AHA AW OC The number of trees shown under the 7rees heading refers only to the number of trees for the key species not for all the species in the group Likewise the statistics shown are for the key species only not group means The last column shows the number of species in each group Statistics are not shown for palms as these were grouped on the basis of their botanical affinity rather than via the ordination process The species group that each individual species belongs to is contained in a database file called GROUPS This has two fields GENSPP and GRUP which give the abbreviated scientific name of the species and its allocated group This file can be used as a resource to add group numbers to any species list that is being prepared for use by SIRENA 35 Growth and mortality rates by species groups Once species groups had been determined it was possi
5. as follows 1 The code used for the inventory eg INV 2 The name for the type sample comprising any descriptive text eg Muestreo diagnostico The sample plot size in ha eg 0 3 The smallest diameter in cm measured on the whole plot eg 30 5 The size of a plot used for subsampling within the main plot to a smaller diameter limit eg 0 075 6 The name of the species list file applicable to the data excluding the SPL extension eg CODFORSA AU Species groups for forest management General considerations Within the SIRENA 2 model each species is represented individually However natural tropical forests typically comprise 300 400 species of which 50 100 may be of commercial interest In order to make prescriptions for harvesting and silviculture easy and in order to present ouputs and results simply species need to be grouped Generally the groups should comprise species with similar commercial demand and similar management and silvicultural characteristics Groups appear in the legends of graphs and in other contexts where very long names are not very helpful Generally the names given to species groups should be short mnemonic codes of not more than 8 10 characters The name of one group should not be a complete subset of the name of another group For example A and A SIRENA will confuse group names if one is a subset of another Group names are also sorted in outputs and lists in alphabetical order This m
6. dThe datalbase 5 a a REHAB SN m 33 The formation of species groups ssssssssssssssseee eene e ener ESSEE EEEE EEES nennen 34 Growth and mortality rates by species groups sssssse mem emn 36 The basal area increment model sssssssssssssesseseseeeeneeeeen eene nnns 37 Basal areamortality 1 25 t iiit n tpi nate c HE ERU D BOR ER epo AN HERE Perte ne s 37 Recruitment basal area ene eren enn ne nemerrn ener ener nennen nens 38 Crown classification of trees ssssssssssssssssseeeseeneneen eene nnne nnn nnns 39 bogging damage 1 1 tetti ced etum dete p ci tem dtu m nd di te cete Eie tem edet dete oy 40 Supplying coefficients to the model sss nennen nennen 40 Conclusion The application of SIRENA eeeeereeeeee nnne 42 Appendix Troubleshooting ccceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeseeeenaeeeeeees 43 iii List of figures Figure T The login screen inneil cto nd d e ae ete 3 Figure 2 The file selection dialog box used to locate a directory sse 4 Higure3 The SIRENA 2 menu System etie titel d ele eee eee a ed eet 5 Figure 4 Dialog to select a forest basis file sse ener enne 6 Figure 5 Screen for managing species groups essen eene ener eere 11 Figure 6 Screen to define harvesting options eese eere 12 Figure 7 Dialog for silvicultural treatment option
7. the required records from the database This can be deleted if its presence is inconvenient After creation of ouffile FBs it can be edited with a text editor to add the following lines of information 1 A code for the inventory type INV and DIAG are defined for CODEFORSA inventario preliminar or muestreo diagnostico for example New codes and inventory types can be created as described in the following section 2 A zero or a forest type code This information is not used by SIRENA 2 3 The title of the dataset This can be any text 4 A zero or the effective forest area This information is not used by SIRENA 2 The file should be checked with the text editor to ensure that there are no blank lines between plots There should also be no spare blank lines at the end of the file The data should be separated by commas not semi colons or any other delimiter Decimal points must be marked with the dot character not a comma None of the data needs to be enclosed in quotation marks Any quotation marks found should be deleted using the search and replace function of the editor The file must be saved as a text only file Saving it as a standard word processor file rich text file or other format will cause SIRENA to fail when reading the file Defining new inventory or sampling types New inventory types are created by editing the file iNvrYP sN2 with a text editor Each line represents one type of plot There are six items on each line
8. 10 afios 50cm 40 rem 3 m2 ha lim Ferlo II 40 B comercial i Exoticas Vol menes aprovechados Gavilan m3 ha 20 B Nocomercial Palmas Potencial Volumen danado todos especies Basal area dynamics The graph of basal area dynamics is invoked from the menu bar Din micas de area basal It illustrates the components of the stand that the model uses in its calculations to estimate growth An example of this graph for the same management regime as the preceding cases is shown in Figure 16 The standing basal area is subdivided into trees of the upper canopy corresponding to crown classes 4 and 5 according to Dawkins system and the lower canopy or crown classes 1 3 Defective trees of either category are shown above these two layers This includes trees of bad form or trees which have been significantly damaged by logging Each of these three layers has its own distinctive species dependent growth and mortality function within the model Generally most active diameter growth occurs in the upper canopy trees When running the model it may appear that the lower canopy basal area is increasing whilst that of the upper canopy remains relatively constant This results from the fact that as the upper canopy trees grow the smaller ones will be pushed into the lower canopy by competition and hence will become more static in their diameter growth In the process the proportion of the stand basal area in
9. 6434 1 7799 0832 4 12 68 9224 24 LJ 25 15 185 30 4 28 0807 0055 0211 0017 20 2723 0324 083 0 061 0 027 34 0 0416 0 0418 0 2676 323 0 706 0 518 192 0 0212 0 0354 0 0423 230 0 629 0 293 134 0 0155 0 0140 0 0142 656 0 866 0 467 124 0 0263 0 0231 0 0499 297 0 430 0 463 125 0 0117 0 0013 0 0252 560 1 029 0 467 131 0 0297 0 0338 0 3179 525 1 375 0 344 94 0 0071 0 0000 0 0071 194 0 346 0 156 104 0 0098 0 0194 0 0229 404 0 754 0 117 155 0 0140 0 0054 0 0363 344 0 645 0 354 89 0 0255 0 0328 0 0785 876 1 161 0 342 86 0 0292 0 0228 0 4143 384 0 511 0 571 95 0 0192 0 0245 0 0669 200 0 341 0 115 81 0 0356 0 0160 0 0563 401 0 652 0 375 110 0 0257 0 0107 0 0231 226 0 276 0 138 71 0 0241 0 0166 0 0522 543 1 033 0 616 70 0 0114 0 0214 0 1043 272 0 392 0 155 55 0 0446 0 0283 0 1270 476 1 457 3 425 44 0 1211 0 0000 0 1211 368 0 924 0 672 76 0 0370 0 0352 0 1171 178 0 198 0 133 72 0 0189 0 0149 0 0809 233 0 124 0 051 120 0 0572 0 0303 0 0572 318 0 657 0 167 72 0 0352 0 0186 0 0765 164 2 683 1 507 53 0 0841 0 1444 0 0583 199 0 102 0 088 45 0 0205 0 0279 0 0440 953 2 387 1 755 56 0 5466 0 0730 0 5466 099 1 227 0 000 58 0 2083 0 1510 0 7614 379 0 391 0 253 41 0 0184 0 0037 0 0713 256 0 257 0 213 48 0 0128 0 0246 0 0369 466 0 757 0 388 50 0 0971 0 1035 0 2197 176 0 142 0 021 31 0 0112 0 0271 0 0270 466 0 904 0 384 31 0 0125 0 0230 0 0125 366 0 406 0 301 29 0 0106 0 0260 0 0106 335 0 280 0 178 41 0 0263 0
10. Gavilan B Nocomercia Palmas Potencial Harvested volumes The graph of harvested volumes is produced by selecting the menu bar Volumen aprovechados This shows a bar for each felling cycle subdivided into quantities for each species group as illustrated in Figure 15 The values are cumulative and show the volumes actually extracted The volume of an individual species will be the difference between the top and bottom values for its section of the bar The topmost section shows volumes destroyed for all species among the remnant population due to skid trails felling gaps and severe damage to standing trees This example shows a typical pattern of yields in which the most valuable species tend to be taken in the earlier cycles and less valuable ones in later cycles The graph also shows a moderate decline in yields over time These tendencies of a simple management regime can be corrected by placing more stringent limits on the proportion of valuable species removed and by somewhat reducing the overall level of harvest It should be noted that the felling damage shown on this graph is calculated for all species and for all size classes represented in the simulation This will be down to 10 cm diameter in the present version It illustrates the point that any felling carries with it a significant cost to the remaining growing stock in terms of trees and volumes lost 26 Figure 15 Graph of harvested volumes Aprovechar cade
11. an option to pause the simulation after the harvest has been performed for a more leisurely analysis of the various output screens and an option to switch to a different management regime scenario after the next specified harvest These groups will be considered in more detail in the next sections Scheduling and controls on the general intensity of harvesting In Figure 6 the box labelled Control de aprovechamiento includes the functions which control the timing of harvesting Each one has a check box on the left and a value on the right Note that if the checkbox is not marked that particular aspect of scheduling will be ignored regardless of any value in the field at the right The options control harvesting as follows A o del primer aprovechamiento This specifies the first year for harvesting Felling may be delayed after this year if the third option for basal area control is also active and the required basal area has not been reached Ciclo de corta This specifies the felling cycle in years Fellings will not be done over an interval less than this period If this box is not checked no felling is done unless the basal area control option is active rea basal requirido antes del aprovechamiento This controls felling by basal area if it is checked The stand cannot be felled unless the specified basal area has been reached If a felling cycle has also been specified then they will act together as constraints on felling in
12. as lower classes are not created by the model with its current growth model calibration Form codes The form codes that occur in the inventory data representing trees that are acceptable for harvesting should be shown in the box labelled Codigos de forma Several values may be entered but they must be numeric not text codes and they must be separated in the list by semi colons Any form codes other than those listed will be deemed to mark a tree as defective either due to damage or bad form Figure 10 Settings for sensitivity analysis Opciones general Ioj x Volumen Factor multiplicador a modelo base Crecimiento di metrico Raz n de mortalidad Ingresos Dafio de aprovechamiento Competencia Guardar como formato Anular 2 Colour output Graphs will be printed in colour as they appear on the screen if the box labelled Imprimir en color is checked This generally works better than black and white printing with laser and HP Deskjet compatible printers In the present version the patterns selected by default for black and white printing are little confusing When graphs are copied to the clipboard this is always done in colour as a metafile image This can be edited with various picture editors including that in Microsoft Word Saving options Checking the box labelled Guardar como formato will save the options into the file DEFAULTS SN2 in the user directory when the Acceptar button is presse
13. di metricas cm A o 0 Cantidad Grupos 10 20 20 30 30 40 40 50 50 60 60 70 70 80 80 90 90 c Comercial 687 242 182 242 364 2 Gavilan o i C Vol m3 ha Nocomercial 0 61 0 30 Palmas C AB m2 ha Potencial i 1 21 1 82 Total 13 03 7 88 3 33 3 03 3 B4 B Forma Aceptable C Mala C Ambas Mostrar nombres cient ficos Escribir cuadro a archivo Totales Selecci n 21 21 Acumulado Neha Ferio ll The table shows data for trees of acceptable form only defective trees only or both depending on the value selected under Forma The boxes labelled Totales provide a calculator function that is useful for analysing harvesting and treatment specifications If a group of cells are selected with the mouse as shown in the example then the totals for this group will be displayed in the box labelled Selecci n Clicking on the Selecci n field adds this total to any value already in the Acumulado field and clears the Selecci n field for a new value The Acumulado field itself can be cleared by double clicking it These functions are useful for totalling values for particular species within a range of diameter classes 30 The table initially shows summaries by species groups Double clicking the mouse on a line expands the table for that group to show the data species by species Double clicking again causes the table to revert to its original form showing group summaries The stand table data can be saved to a
14. icon should cure this if the files do exist If the Localizer screen appears repeatedly after login then either e The wrong user directory has been given 43 e The base directory for users has not been properly specified on the D parameters for network use e The files DEFAULTS SN2 in the user directory or MODELS SN2 in the Sirena directory cannot be found Review the installation method discussed from 1 The program will normally fail with an error 53 or 76 if it is unable to find the necessary files after one or two attempts The program crashes immediately after login With versions before 2 11c this will happen with execution error 13 With later versions it will advise that the decimal point must be set to Systems which are configured to use the comma as the decimal usually give problems with a variety of software It is strongly recommended that the control panel international settings are used to set the decimal to a point the list separator to a comma and the thousands separator to a comma or a space The Spanish Mexico settings give this by default Various violation errors are reported erratically when running the program The Windows installation is probably corrupt It should be re installed from original manufacturers disks This is especially likley to be true if similar problems occur with other Windows software Virus attacks in particular can leave corrupted files even if anti virus software h
15. list in the Sirena base directory The species codes are processed as text items and may therefore be numeric hierarchic or mnemonic codes of any type CODEFORSA uses simple numeric codes If a species code is found in the inventory data which does not exist in the species list then SIRENA appends it to the internal list of the program For example if species number 999 was found in the data but not in the list it would append an entry with the following details e A code number of 999 as found e A local name of Spp lt 999 gt e A scientific name of Spp 999 e A species group of X e A growth model from the predefined default from the mopeLs sn2 file This internal list is not automatically saved to file It can be saved from the Grupos de especies dialog as discussed on page 10 The procedure suggested for dealing with missing species codes and updating the species list is as follows e Read several sets of inventory data After each set the species list should be saved via the Grupos de especies menu This will build a list of the common missing codes e Edit the species list in the user directory to insert correct local and latin names The species list can be read in to Excel and sorted by latin names to facilitate the editing of an appropriate species group code If the missing code is in fact a duplicate for an existing code then the growth model can be altered to that for the pre existing entry Otherwise t
16. scenarios and outputs These users can each be assigned a sub directory to the SIRENA directory When logging in the user types his sub directory name in the login screen A third possibility is to have the user directories located on another disk or drive The command line of SIRENA can have a parameter D d path added which gives the deafult base address The user then only types the sub directory at login For example in the network installation at CODEFORSA the following arrangment exists SIRENA base directory D APLIC GENER SIRENA2 User base directory X USUARIOS Command line D X USUARIOS User directories XNUSUARIOS DENIS X USUARIOS MAURICIO X USUARIOS RODOLFO SIRENA etc Login name icem etc DENIS MAURICIO RODOLFO SIRENA etc The command line is set by editing the properties of the program icon from the Program Manager in Windows 3 1 In Windows 95 the shortcut to the program is selected and the property keyword clicked from the menu or right mouse button When creating a SIRENA user directory the file DEFAULTS SN2 must be copied in to the directory from the SIRENA base directory This file contains details of the user s current preferences for the program Updating SIRENA 2 Updated versions can be installed by copying all the files on the update disk into the Sirena base directory This will usually include SIRENA2 EXE IDIOMA SN2
17. species groups Opciones para agrupamiento de especies Grupo Grupo m le Especies en grupo Especies en grupo Aiouea costaricensis Alnus acuminata Anacardium excelsum Anthodiscus chocoensis Balizia elegans Beilschmiedia anay Billia hippocastanum Bombacopsis quinata Brosimum alicastrum Brunellia costaricensis Caryocar costaricense Caryodaphnopsis burgerii Ceiba pentandra Anular Chimarrhis parviflora Chrysophyllum spp Cinnamomun cinnamomifoliu Media Aceptar Clarisia biflora IV Mostrar nombres cient ficos Hacer permanante los cambios Albizia caribaea Aspidosperma megalocarpon Astronium graveolens Billia colombiana Cabralea canjerana Calophyllum brasiliense Carapa guianensis Cedrela fissilis Cedrela odorata Cedrela salvadorensis gt Cedrela tonduzii Copaifera aromatica Cordia alliodora Cordia cymosa Cordia gerascanthus Cupressus lusitanica Cynometra hemitomophylla C Borrar selecci n C Anexar grupo nuevo C Suprimir grupo A group is deleted by placing it in the left hand box removing all species from it into other groups and then clicking the button Suprimir grupo The button Borrar selecci n releases any species marked as selected If the two groups in the boxes are the same the message Grupos de especies deben ser diferentes will appear Any action will be blocked until one of the groups is changed It is not possible with this screen eith
18. the comparisons line must agree with the current groups See page 23 Treatments or harvesting are not as expected Read pages 12 18 carefully The species groups must agree with you current groups the Normal option button should be checked and you should have indicated S against those species which you wish to harvest 45
19. the originating data its full directory path on the system and the field names for the plot number diameter species code and form defect code Given this information the following Foxpro commands will extract the data SET DEFAULT TO userdir USE invpath invfile INDEX ON plot TO TEMP FOR forest id COPY TO outfile FBs DELIMITED FIELDS plot species diam form where USOFTIF Les aes citas The full path of the user s directory for SIRENA work 175171272 1 EM PN EN NER This is the full path of the inventory data file On a network the user must have access rights to this path PAVING os oan ticle cue This is the name of the DBF file containing the inventory or sample plot data DIOE chis atu bib tis This is the field name containing the sample plot number IOTGsL seco aet This is the field name containing the forest name or number TTE M S This is a number or character string denoting the forest farm or sample to be extracted If forest is a text field it must be enclosed in quotes If forest is numeric it should not be A FoxPro error message results if forest and id are not of compatible types o tfile MIDI PRENNE The name of the output file with an extension FBS added BOCES cada oe DARE The field containing the species codes Gams us eed ots The field containing tree diameters TOW siet The field containing form or defect codes This method creates a temporary file TEMP IDX in the user s directory which indexes
20. to justify curvilinear relationships The functions are similar to those used in SIRENA 1 and have been reported elsewhere Maginnis 1997 Supplying coefficients to the model The coefficients for the above models are supplied in a file called MoDELS sN2 which should be in the same directory as the SIRENA program This is a simple file of comma separated values which can be amended by text editor A listing is shown in Figure 26 The values in this file are as follows e The coefficients intercept and slope for the linear basal area increment function e The coefficients for the linear mortality function e The coefficients of the negative exponential recruitment function the time lag in years and the mean diameter of recruits e The number of pioneer groups and the model numbers applicable e The area and tree functions for logging damage and the avoidance diameter for skid trails e The linear coefficients of the crown class function 40 e For each species model the model number diameter coefficients maximum diameter and mortality coefficients as set out in the table in Figure 21 e A final default model here number 36 which is assigned to all species not found in the species list but occurring in the inventory data This will will be seen to replicate model 21 which is for trees of medium size and growth rate see Figure 19 Figure 26 Listing of the MODELS SN2 coefficients file 0 3854 0 0106 3 6292
21. 0414 0 0263 101 0 104 0 118 38 0 0471 0 0695 0 0657 220 0 168 0 090 38 0 0409 0 1689 0 1293 318 0 657 0 167 72 0 0352 0 0186 0 0765 L2 rp 00 1001 5 C ho IP O OOO O7 Or Oo OOO 20 9 YN NNW ON GS o8 41 Conclusion The application of SIRENA SIRENA 2 like its predecessor is designed to supplement forest management plans by allowing the sustainability of proposed harvesting regimes to be tested over time It gives due weight to the influence of logging damage and reflects the influence of stand density variations on growth and mortality It also shows how different policies for retaining and selecting species can influence the mixture that is harvested at each successive cycle The model s accuracy is uncertain and it will require more and longer measured series of permanent plots to test it thoroughly However tree growth is not especially sensitive and over moderate periods of up to thirty years its predictions are unlikeley to greatly in error Beyond this species composition may change as a result of different patterns of recruitment and the model s ultimate accuracy is unlikley to be very great until more data is available Certain aspects of the model are not essentially dynamic and predictive but are simple data processing functions The effects of selective harvesting or treatment on species composition for example or the various short term graphs available of volume basal area and size class distributi
22. Figure 26 Listing of the MODELS SN2 coefficients file sse 41 iv Introduction SIRENA 2 is a simulation model designed to assist the management of natural tropical forests It is based on a preceding model SIRENA 1 which was developed during 1996 for CODEFORSA under the UK ODA Integrated Forest Management Project The scientific concepts around which the model are based are discussed in Alder 1995 This document is a user s guide to the model written for the forest engineers of CODEFORSA and for other potential users of the model SIRENA 2 is based on the analysis of tree growth patterns from data supplied by CODEFORSA and from plots established and maintained by Portico SA The data is applicable in the northern zone of Costa Rica covering drier forests principally in CODEFORSA s managment area and wetter forests under management by Portico The wetter forests are characterised by a high frequency of Carapa guianensis whilst the drier ones have Vochysia ferruginea as an indicator species These two species do not co occur with any frequency Pentaclethra macroloba is very common in both types of forest Data from secondary forests was also supplied by COSAFORMA for use with the model but due to limited time this has not been analysed substantially and the model will probably not reflect accurately regrowth from abandoned farmlands with its present parameters Installation Initial installation Initial installati
23. IDIOMB SN2 MODELS SN2 Obtaining updates on the Internet Updates and original installation files can be obtained by E mailing the author at D ALDER MAIL ZYNET CO UK Problems encountered during installation and operation can also be advised to this address Starting Sirena 2 The login screen When the program is started by clicking on its Icon a screen should appear as shown below Figure I The login screen E d f LJ lt SIRENA 2 10a Login Nombre de us rio The shows the characteristic background screen of Sirena 2 with a picture of forest canopy The window controls and text fonts for the main menu differ between Windows 95 and Windows 3 1 The login box offers the following options Nombre de usuario The user should enter a name corresponding to their user directory The base for the user directory is shown at the bottom left of the dialog screen OK ee m Clicking this will start the program Cancelar osdcioosud ei The program will close immediately The localization dialog If the user s base directory has not been correctly specified with the D parameter on the command line see page 2 then the login screen shown in Figure 1 will not appear immediately Instead a box will appear as shown below for selection of a directory The user can navigate this until the correct user directory is found and then double click on it and close the dialog In some simple cases the program may then operate correctl
24. User s Guide for SIRENA II A simulation model for the management of natural tropical forests This document and the SIRENA model have been written by Denis Alder for CODEFORSA Comisi n para Desarollo Forestal de San Carlos under a consultancy supported by the UK Overseas Development Administration Revised 12 Jun 1997 Contents Intro du CUO ocio oe eio ioo 0e i DL Pe Seo Po Eden De Ee Pe HE sa ceseecdesuccezeececabeeee 1 MINS CaN ATI OM Meter 1 Initial installatlon 2 hee e HE a he terim td 1 Setting Up the user dITectoby inp De e p ende perta Pax p eee e ERR e eae Pd Regen 2 Updating SIRENA 2 5 ipee d ire deed etaient pisvesd adveniens dide 2 Obtaining updates on the Internet ssssssssssseseeeenen eene nennen nennen 3 Kidd gum 3 Tie loginiicreen a dee ht og Wt bg ula leet abutettatel ettet baa HER ebd aede ette 3 thelocalizatiom CIAL OG Cm 4 The menu sysltem oet t tego tlg utu eh eiut et aient a iiber docete as 4 Defining the forest basis for a simulation eeeeeeeeeeeeeeeeeee 5 The Base forestal menu rei eire et ete eee ede eee Dee eh ee d de Renner d a ep v rud 5 The processing of species COCES cccceceeeecee cece cece eeceeeaaeceeeeeeesecaaaeceeeeeeeseccaeaeeeeeeesetensseneeeeess 7 Creating inventory basis files 0 0 eter ee ieee eters eter erties een aeee ee enne 8 Defining new inventory or sampling types
25. ario should also be given a code in a format suitable to serve as a file name This is typed in the Archivo window which only allows letters and numbers up to a limit of 8 characters to be typed The scenario is then saved by pressing the Guardar button If there is an existing scenario with the same file name it will be overwritten without warning if not then a new file is created Checking the box marked Auto guardar does not save the scenario immediately but will cause it to be re saved automatically in exit from the Aprovechamiento or Tratamiento screens This is done without warning regarding the overwriting of the existing values Existing scenarios can be loaded by clicking on them from the list This immediately loads them in to memory If the harvesting and treatment regimes have been edited without saving them as a scenario a warning message appears Escenario actuel no guardado Sobre escribir At this point the load can be cancelled and the current scenario saved first if required It is possible to load only the treatment or only the harvesting component of a scenario by checking one of the option buttons at the right of the screen The default for manually loaded scenarios is to load both components If only a partial loading is requested then the part not loaded remains at its current value For example if under the Cargar options Tratar is selected and a scenario is loaded by clicking on it then the current harvesti
26. as cleaned the system Access denied errors occur when saving species lists or scenarios Under a network this will probably occur because you are working in a network directory to which you only have read access Modify the installation setup for directories The D parameter on the command line can be used to direct your work by default to a local disk or user directory see page 2 If it is a stand alone system then you may be running two copies of SIRENA simultaneously or may have the required file opened by another program such as Excel or Word Sirena2 runs excruciatingly slowly Sirena 2 should not be run on an SX machine as this lacks a co processor and does arithmetic about 10 times slower than DX machines Simulation is very arithmetic intensive The lowest standard processor useable is a DX2 66 A Pentium is recommended The model crashes with a division by zero message when running simulations If any of the update frequencies on the General Options screen Figure 9 are zero then the program will crash when a simulation is started 44 The program runs but produces ridiculous values There could be a number of reasons The mopeLs sn2 file could have been modified Check it carefully against Figure 26 The sensitivity parameters could be zero negative or extreme values Check the screen shown in Figure 10 All the values should be 1 The comparisons graph is always zero The species groups listed for
27. ay be borne in mind when devising group names The graphic outputs use a built in scheme of colouring patterns for up to 6 groups Thereafter the system resorts to default colours supplied by the graphics system These may not be very ideal and may not be the same for different groups between graphs It is therefore better if possible to limit the number of groups to not more than 6 Both these restrictions apply to SIRENA 2 10 and will probably be relaxed on later versions Managing species groups in SIRENA The menu Grupos de especies gives access to a dialog screen for moving species between species groups and for creating and deleting groups This screen is shown in Figure 5 below Species can be moved between groups by clicking on them with the mouse to select them in either the left or right box and then clicking on either of the arrows in the centre to move them to the alternate box The group displayed in either box is selected from the pull down lists at the top A new group is created by over typing the group name in the right hand box and then clicking the button marked Anexar grupo nuevo If no species have been selected in the left hand box an empty group will be created If species were selected in the left hand box they will appear directly in the new group This was not the case with SIRENA 1 which worked only with pre defined groups and gave no access to individual species 10 Figure 5 Screen for managing
28. ble to compile increment and mortality data and fit regressions In the event it was found that tree size was not a significant factor in tree increment regressions once canopy position and defect status had been accounted for Increment statistics were therefore compiled as a simple table of mean increments for lower canopy trees crown classes 1 3 upper canopy trees crown classes 4 5 and defective trees malformed damaged or trees which died over the period Mortality statistics were compiled on a similar basis The ceofficients are shown by species groups in the table below Figure 21 Diameter increment and mortality coefficients by species groups Species Total em increment cm yr T iim Mortality Rate group trees Canopy Defective Canopy Defective 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 This table shows the total number of trees for each species group Not all trees are useable for all analyses Dead trees cannot be used in increment studies for example The Dmax figure was compiled as either twice the 9596 order statistic of the diameter distribution or the actual largest diameter whichever was smaller This was designed to give an estimate that avoided excessive values due to eccentrically large individuals 36 The basal area increment model It was found in initial trials of the model that simulated increment tended to be excessive after heavy disturbance and regeneration of large numb
29. clature of groups prior to creating them Specification of harvesting operations The Aprovechamiento dialog Selecting the menu bar Aprovechamiento brings up the dialog screen shown in Figure 6 Figure 6 Screen to define harvesting options w Opciones de aprovechamiento Control de aprovechamiento y Opciones especiales IV A o del primer aprovechamiento 10 C Sin aprovechamiento Iv Ciclo de corta a os 10 7 Homal rea basal requirido antes del aprovechamiento m2 ha D C Mode de pausa Limite de n mero de rboles a cortar n ha iii v Limite de rea basal a cortar m2 ha C Cambiar a escenario Selecci n de rboles Grupo Diam Min Remanente Cortar remanente especificado por NZ C N ha Ordenar la tt prioridad J de grupos Anular This screen allows various options relating to harvesting to be specified The options are divided into three functional boxes Control de aprovechamiento This includes options which control the decision of whether to harvest or not and general limits on the number of trees or the basal area to be removed 12 Selecci n de rboles This covers the felling limits and retention specifications by species groups and determines the order of priority when felling species groups Opciones especiales This provides a fast way of selecting a no harvesting option standard operation
30. d The existing file of this name will be renamed as DEFAULTS 01 02 etc each time the file is saved These backup copies can be deleted from time to time They are intended as a safeguard as inappropriate values for some of these general options can make the program unrunnable Sensitivity analysis The second tab of the Opciones general screen labelled Sensibilidad reveals several parameters which can be varied to test the sensitivity of the model or to adjust its calibration to different sites or situations see Figure 10 Each value is a multiplier for the basic model referred to For example replacing the value of 1 in the Crecimiento diametrico with a value of 1 1 will increase all diameter increments by 10 subject to other constraints in the model These factors should not be set to values other than 1 for normal use The only field that may be applicable for general forest management is that for logging damage Dario de aprovechamiento which may be adjusted to test the implications of more or less damaging regimes However it should be appreciated that coefficients outside the range 0 7 to 1 3 implying a 30 reduction or increase in typical damage are unlikely to be realistic in any context Large values for these sensitivity factors may make the program unrunnable 22 Figure 11 Options for volume definitions amp Opciones general w 0002268 Comercial o 2222 Volume options Selectin
31. dicating both a minimum period between fellings and a minimum basal area that must be reached If no felling cycle is specified then felling will be done whenever the stand reaches the specified basal area L mite de n mero de rboles a cortar This places a general limit on the number of trees per hectare that may be removed during felling It can be given as a fraction if required representing whole trees over a wider area For example 2 5 ha indicates 25 trees on 10 ha L mite de rea basal a cortar This is an alternative to limiting felling by tree numbers It specifies the maximum basal area to be removed in m2 ha If both tree numbers and basal area constraints are given together the program will apply whichever 13 constraint is most severe in a particular harvest However this is likely to be confusing to interpret and is not recommended Specifications for the selection of trees The group of options shown under Selecci n de rboles in Figure 6 determine the specifications for felling At least one line in this table must be checked with S in the Cortar column for any harvesting to be done Specifications for felling trees include the species group to be affected by a given rule the minimum diameter for trees to be felled and the number or percentage of trees to be retained For each group in the table that is to be felled the Cortar box should be clicked with the mouse to show si The retention can be spec
32. ed but the specifications will be exchanged The current line in the table can be deleted by pressing the Ctrl Del key A new line will be inserted by pressing Ctri Ins A single cell is deleted by pressing Del The deleted text is stored in a memory and will reappear in the active cell when ns is pressed This can be used to rapidly enter the same figure in several cells It is entered in the first cell and Del pressed to place it in the memory Pressing ns on each of the required cells will then insert the figure The last column header Cortar can be checked or cleared by clicking with the mouse or by pressing the spacebar 14 Special options There are four special options available as alternatives Sin aprovechamiento When this is checked no harvesting is done This is a quick way of establishing a baseline for management without having to reset all the individual options Normal The controls and specifications are applied throughout the simulation Mode de pausa The simulation stops after each harvesting has been done It can be restarted when required by pressing Ejecutar under the Ejecuci n menu This option provides a means of examining the graphs and tables which do not have a time base before they change as the simulation proceeds While the program is stopped the details of harvesting silviculture etc may be modified to take effect as the program is re started Cambiar a escenario This mode requires tha
33. epting the COSAFORMA plots which was 12 68 cm relative to a minimum measurement diameter of 10 cm The function shown is based on a 4 8 year lag between the initial basal area and the accumulated recruitment In the model this lag period is used to input to the function basal area at an earlier date when calculating the current recruitment It will be seen for example that if a heavy harvesting is done recruitment does not respond until 4 years after the event This reflects the average time for trees to enter the measurement limit of 10 cm plus Recruitment is apportioned among species based on these present in the inventory data As this is read a frequency table is built up of pioneer and non pioneer species This is used to determine raltive frequencies of recruits A bias function is applied to increase the proportion of pioneers and decrease that of non pioneers at lower basal areas 38 Figure 24 The basal area recruitment model Foi 22554 2S 2228 pat E AW UA eu SEER SSE ede See SOS eee te s 1 40 4 o CODEFORSA B Portico 120 4 4 Meandecruitment m xm E 3 22 244 94 SSP BSS ue PF Be Se 20S S SSS 2 8 iia 2 x 64 x m o9 om Boe egos E LLL LL LLLLLSLQSLSLLLLLCLCLLLLLLLLLLLLC 0 80 4 Recruitment m2 ha yr 0 60 4 0 40 4 0 0 5 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 Initial basal area m2 ha Crown classification of trees A key element of the growth model is the assignation of a canop
34. er to duplicate or delete species from the list They can only be moved from one group to another The checkbox labelled Mostrar nombres cientificos will switch the naming between scientific or vernacular names However it should be noted that typically many species will not have vernacular names For these the generic name is shown enclosed in brackets Generally the scientific names are more useful in this presentation The checkbox Hacer permanente los cambios is important It causes the species list in the user s directory to be overwritten with the new groupings when the Accepter button is pressed A warning message precedes this to allow the user to change their mind With this facility each user may therefore create a unique set of groupings Ifa particular grouping system is determined as being the definitive one then the relevant 11 species list file eg CODFORSA SPL can be copied back into the Sirena base directory from Windows There are some deficiencies in this dialog screen that will probably be remedied in later revisions of SIRENA e The blocking function when the two lists are for the same group is unduly oppressive It should be relaxed to apply only to operations that move species e Itis not possible to directly rename a group Presently a new group must be created the species copied to it and the old group deleted This is rather tedious and it is therefore important to spend some time thinking out the nomen
35. ers of trees As in the earlier SIRENA 1 model it was decided to add a basal area increment function for the stand as a whole that would be used to condition individual tree increments whilst preserving the relative differences between species and stand components lower and upper canopy defectives This function is shown in Figure 22 below Figure 22 The basal area increment model VAD See PEE P20 ph sow Uf xs CODERORSA 7 9 Far SF Y E eRe Ree RCRORQ9 OR S PAS eS Portico Hn 1004 al MeanBAI 2 222222 e e o H y 0 0106x 0 3854 R 0 0951 0 80 e 0 60 4 BA increment m2 ha yr 0 40 4 O20 2 EC EN p a See age SS oe Sra E E 0 0 5 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 Initial basal area m2 ha The correlation between basal area increment and stand basal area at the start of a period is very weak The calculations here are done over the whole measurement period and are gross increments on live trees The average period was 4 83 years for all plots The data for CODEFORSA and Portico are shown with separate symbols The COSAFORMA data was not used due to the small plot size Although the correlation is low this function is useful in showing that increment is relatively constant over a range of basal areas and implies higher net per tree growth at lower stockings It is used as a moderating and conditioning function for the individual species and stand compone
36. ffectively consumed non renewably Figure 17 Example of the comparisons graph for three scenarios Volumen de arboles 50 cm dap para especies Comercial 150 100 1 Sin Aprovechamiento m3 ha 2 Aprovechar cade 10 a os 50cm 40 rem 3 m2 ha lim 50 3 Aprovechar a 25 m2 ha 50cm 60 rem 3 m2 ha lim 0 0 10 20 30 40 50 A o The title to the comparisons graph will show the volume limit and the species groups selected in the Opciones general screen The legend shows the individual scenario titles This requires that the scenario is saved via the Manejo de escenarios otherwise only run numbers are shown and the legend text is blank This graph is not well suited for monochrome reproduction due to the similarity of the lines 29 Output tables Table of species by diameter classes Clicking on the menu bar Cuadro de clases di metricas brings up a screen like the example shown in Figure 18 This displays initially a stand table for species groups It has several active functions The statistic shown in the table may be N ha volume or basal area Volumes if selected are calculated to the lower diameter limit set under the Opciones general Volumen screen Figure 18 Screen showing the table of diameter classes FASIRENA 2 11b Modelo para simulaci n de bosque tropical Cuadro de clases diametricas Iaf x sl Escenario Gr ficos Ejecuci n 8 x r Clases
37. file sc2 The forest basis file FBs The species list file sPL The date of the Sirena Exe file and the version number shown Missing files A message on start up that the file VB40016 DLL or VB40032 DLL is missing indicates that the program has not been installed properly It cannot be installed by simply copying the EXE files but requires a number of library files The original installation disks are needed for this purpose Missing files with extensions OCX or DLL may be reported later when particular functions are accessed This is the same problem Copies of the Windows 3 1 compatible installation disks are available from CODEFORSA This version can also run on Windows 95 The Windows 95 installation process with the 32 bit version is problematic due to a fault in the Microsoft installation Wizard for this version Please contact the author by E mail for current advice Wrong Windows version The Windows 95 version cannot be run under Windows 3 1 The file is called SIRENAQ5 EXE to distinguish it from sIRENA2 ExE which is the Windows 3 1 version The latter can however run on Windows 95 and is certainly easier to install Continued appearance of the localizer screen If the screen shown in Figure 2 page 4 appears repeatedly before the login screen then the program cannot find the files IDIOMA SN2 and IDIOMB2 SN2 which should be in the same directory as SIRENA2 ExE Setting the working directory on the program startup
38. g the third tab from the Opciones general screen labelled Volumen give access to settings for calculating and displaying volumes The general volume equation coefficients are for a logarithmic equation with the form Volume a Diam with volume in m and diameter in cm The L mite m nimo field gives the smallest diameter tree that will be included for volume calculations on the graph of standing volumes The Especies para comparaciones should be a list of species groups separated by commas not semi colons in this case to be included in the volume calculation for the lines on the graph comparing successive simulation runs The group names are case sensitive If a group is given here which is either misspelt or does not correspond to an existing group no error is reported but no volume is calculated The Diam field for the comparisons line shows the tree sizes to be included in calculating the comparisons volumes 23 Output graphs The species abundance graph The species abundance graph is displayed by clicking the line labelled Abundancia de especies under the Graficos menu An example of this graph is shown in Figure 12 below The most abundant species is listed first proceeding down the graph to the least abundant Scientific names are abbreviated to the first 5 characters of the genus and three of the species all capitalized If the use of local names is requested these are abbreviated to approximately 12 characters whe
39. he default value should be left this will probably be 36 subject to revisions of the MODELS snz2 file e If working with Excel save the list again as a CSV file to the user directory e Check that Sirena reads the list correctly by opening files and viewing the stand table to see that the species are now correctly named e From Windows or DOS copy the species list back from the user directory to overwrite the master list in the Sirena base directory Clicking the check box labelled Mostrar c digos de especies faltantes in Figure 4 gives a screen displaying missing codes after a file has been read with a count of the number of occurrences of each code Its use is not necessary for the operation of the above procedure but it does provide a quicker check than examining the species list with a text editor Creating inventory basis files The inventory files used by SIRENA are text files which are given the extension FBS Normally the data will be extracted from a database file At CODEFORSA for example the data may be held in the PLANMAN file called INV DBF or a file maintained by the new TREMA system The most direct route for creating and FBS file is to extract the data from its DBF format into a text file using FoxPro and then edit it using a word processor to add the four lines of header information that SIRENA requires A limited knowledge of FoxPro is needed to achieve the first step It is necessary to know the file containing
40. hese include graphs of species abundance size class distributions volumes by species groups basal area by functional components and a graph comparing successive simulations Ejecuci n zug au This controls the exection of a simulation starting pausing or cancelling it It also allows the current graphic to be copied to the Windows clipboard for pasting into for example a Word document and allows a graphic to be printed in high quality mode directly to a printer Figure 3 The SIRENA 2 menu system Escenario Gr ficos Ejecuci n Base forestal Abundancia de especies Ejecutar Distribuci n por clase de tamafio Pausa Grupos de especies Anular Aprovechamiento Volumen por grupos Tratamientos silviculturales Volumen aprovechados Copiar Din micas de area basal Imprimir Manejo de escenarios Comparaciones de escenarios Salir Par metros de modelo Cuadro de clases di metricas Cuadro de aprovechamiento Selections from the menu are made in the conventional way with Windows by pointing and clicking with the mouse or by using the underlined hot keys from the keyboard Not all menus are initially accessible Generally as a minimum a file of forest data must be read under the Base forestal menu before graphics can be displayed or a simulation run Defining the forest basis for a simulation The Base forestal menu The necessary first step in running SIRENA is to read in forest data that will define a forest to be simulated The s
41. ified either as a percentage of the standing trees above the size limit for that species group or as the total number of trees per hectare to be left This is done by clicking on one of the buttons to the right of the table The species group can be changed by clicking the mouse on the group name to make it active and then pressing the spacebar to cycle through the list of available groups Note that if the species grouping is changed the specifications in this table do not automatically reflect those changes and it will be necessary to adjust the group names and other details as required The order of groups in the table determines the priority of felling when a general limit has been applied If for example there is a general limit of 2 m ha then all possible trees from the first line of the table will be felled until the limiting basal area is reached If it has not been reached by felling these then the process continues with the second line and so on To achieve a proportional felling through a range of groups when there is a general limit a high Remnante is needed to ensure all trees are not taken from the group The order of species in the table can be changed using the up and down arrow buttons at the lower left This exchanges cells in a row from the active cell to the left with the row above or below If the cell selected is in the Grupo column the whole row is exchanged If it is in the Diam column the groups will remain as order
42. it is saved the file name extension will automatically be adjusted to reflect the simulation year The yield table All the figures that underlie the graphs of standing and harvested volumes can also be saved to file as the model is run This requires that a scenario name is selected beforehand using the scenario manager The menu bar Cuadro de aprovechamiento is then chosen A tick will appear beside the bar in the menu to show that this feature is active Outputs from the model will be saved into a file with the same name as the scenario code but with an extension SY2 This file can be imported into Excel in the same way as the diameter class table as noted above The file is opened in Excel and the text import Wizard will automatically format the file using its default options The file contains a heading giving the forest name the scenario title and the headings for the various data columns The rows in the file correspond to years in the simulation The data can be edited as required for presentation 31 Executive menu functions Running pausing and cancelling simulations Under the menu heading Ejecuci n will be found three groups of functions which control execution of the model exportation of outputs or the exit from the program Ejecutar runs the model with currently defined parameters If the model has not yet been started or has finished with FIN displayed in the lower right box of the status bar then it will start f
43. ither the scientific names of species or their local names where given in the species list the name of the genus is substituted in brackets where no local name is present Time controls These cover several aspects of model operation The Limite de tiempo 1s the total simulated time the model will run for Actualizar graficos determines the frequency with which graphs are updated Unir cohortos merges cohorts internally at the intervals inidcated A value of about 10 years will probably give optimal program speed Ajustar copas determines the frequency with which the crown classes of cohorts are recalculated This should not be more than 5 years and on faster computers can be 20 reduced to give smoother lines on the graphs The Ancho de cohortos is not strictly a time control but effects program speed and resolution A value of 2 cm is recommended on slower machines and cm on faster ones with the 32 bit version of the program Diameter classes The diameter classes used in output tables and graphs can be set by typing in lower bounds in the box labelled Limite minimo por clase diametrica These values must be separated by semi colons not commas There is no special limit on the number or size of classes and these values can be changed while the program is running However outputs will become congested if too many classes are chosen the program may fail if there are less than 2 classes and the lowest class should not be less than 10 cm
44. l Exoticas Gavilan Nocomercia Palmas 10 20 30 40 50 60 70 80 90 20 30 40 50 60 70 80 90 Potencia Diameter class cm Volume by species groups over time A graph of standing volume by species groups as it develops over simulated time can be shown from the menu selection Volumen por grupos An example of this graph is shown for a regime with a felling cycle of 10 years and extraction limited to 3 m ha in Figure 14 In this example only Gavilan and the Comercial group are being felled There is a decline in the stock over time but it appears to stabilise over about 50 years It is probable that a felling controlled to remove a lower basal area would be more suitable on such a short felling cycle The volumes are calculated down to the limit specified under the Volumen tab of the Opciones general screen see Figure 11 In this case volumes are shown for all trees above 40 cm diameter The lines are cumulative so the volume for an individual species should be worked out by taking the difference between its line and the preceding line 25 below The time axis will always depend on the time limit for the simulation whereas the volume axis is set automatically Figure 14 Graph of standing volumes over time by species groups Aprovechar cade 10 a os 50cm 40 rem 3 m2 ha lim Ferlo II 200 Comercial Exoticas Volumen m3 ha diam 40cm 100 We Ls ea
45. l and then re inserted to the same and other fields by pressing ns see page 14 Lines in the table can be deleted by pressing Ctri Del and blank lines inserted with Ctri ins Moving the cursor to a cell under the Grupo field and pressing the spacebar or clicking the mouse will scroll through the currently defined species groups The S D option cannot be toggled on or off S or blank if there is a figure in the adjacent raleo field Likewise data cannot be entered in the raleo field if the S D field constains S Because there is no prioritization between groups for treatment there is no control in this table to change the order of groups unlike the similar table on the Aprovechamiento screen 17 The scenario manager Concepts and definitions A scenario in the terminology of SIRENA 2 is a combination of harvesting and silvicultural treatments The scenario manager allows scenarios to be saved to file recalled or deleted from disk Each scenario has a title and a code which is also used as the file name The file name consists of the code plus an extension of sc2 The scenario manager allows a standard list of harvesting and treatment regimes to be kept on disk without needing to re enter the details each run As has been noted both harvesting and treatment specifications can include the linking of one scenrio with another to allow variable specifications which each felling cycle and to allow variable silvicultural
46. ng specifications will remain unchanged but the treatment specifications will be overwritten with those from the selected scenario A scenario is deleted by selecting it and clicking the Suprimir button Once a scenario has been selected the dialog can be closed with Salir or by using the conventional Windows close control 19 Figure 9 The dialog screen for setting general options Opciones general OF x General Sensibilidad Control de modelo Diametro Forma Limite de tiempo L mite m nimo por clase diam trica Actualizar gr ficos 10 20 30 40 50 60 70 80 90 Ancho de cohorts cm Unir cohortos cada afios C digos de fama para Ajustar copas cada a os rboles aceptables 0 1 Nombres de especies Idioma Cient fico C Ingl s Imprimir en color Vulgar Espa ol Guardar como formato CERTE p sera Setting general program options The general options dialog Clicking the mouse on the menu bar Parametros de modelo produces a screen similar to that shown in Figure 9 This is a tabbed dialog with three levels the first of which is shown when the screen first appears This tab gives access to the following options Language Checking either of the approriate option buttons switches all texts and menus between the Spanish and English versions Note that titles created by the user for scenarios or data files are not translated Species names These may be e
47. nt growth models Basal area mortality A function for basal area mortality was used in a similar way to that for basal area increment This conditions individual species group and stand component mortalities from Figure 21 so that their relative values are preserved whilst total mortality agrees with the stand level function 37 Figure 23 The basal area mortality model o CODEFORSA pij Lo E xx i Por co S o Spo GA Gee cee 22682 deg he Bu e Mean BAI B5 20 4 2 Se duelo 2S SRS Se Se ES EE AR EES SESS CSS OE HES ESBS 5 y 0 0324x 0 2723 amp e R 0 2699 N o CQ Bk Lu E ee Say el Ge A ch rt G A aaa on Vis a es ys wel X a Se Be ie e ue tl es sl Seta St ae ae dE 10 2 Bo El E e E lt ea Initial basal area m2 ha As with the earlier regression the R is very low Nonetheless this function greatly improves the behaviour of the model which otherwise does not show a sufficient tendency to increasing mortality with stand density Recruitment basal area Whereas the mortality and increment functions for basal area condition individual tree models the recuitment basal area model is the primary method of predicting recruitment The data and function are shown in Figure 24 This is a more clearly defined function than the others showing a definite decline in recruitment with increasing stand basal area The number of recruits is calculated from the mean diameter of all recruitment exc
48. on These are likely to be useful under any circumstances The model also provides a very useful teaching tool for forest management It is not easy to envisage the important variables that may influence forest growth and composition or the implications of stand treatments or harvesting SIRENA allows these questions to be explored in an interactive way The strong influence of felling damage on short cycle logging is clearly shown The possibilities for manipulating species composition or for reducing accumulated defect through stand treatment can be readily studied Every effort has been made to make the model as easy to use and as interactive as possible Given the complexity of modern operating systems this is certainly a non trivial task Hopefully users will find SIRENA as friendly as its short development cycle may permit The model does emphasise the importance of permanent sample plot work It is only from this type of data that we will learn enough about natural forest dynamics to be able to manage them with any confidence for the longer term 42 Appendix Troubleshooting This Appendix summarises some common problems that will be found when installing and running SIRENA and suggests some possible solutions If all else fails please contact the author by E mail to D ALDER MAIL ZYNET CO UK and attach copies of the current files you are using especially IDIOMA SN2 IDIOMB SN2 MODELS SN2 DEFAULTS SN2 The scenario
49. on is made by a Microsoft supplied installation Wizard from a set of three disks Separate installation kits are available for Windows 3 1 or Windows 95 In each case Disk 1 of the set is placed in the A drive and the A SETUP program run from within Windows The SETUP program is automatic except with respect to the target directory This defaults to C SIRENA2 but another directory name can be typed instead if preferred If the installation is to a network a directory on a shared drive or network server should be specified The installation program copies any required DLL or other system files to the Windows system directories and creates the program Icon During this process error messages may arise The gnore option should be chosen and the installation will proceed to completion These error messages do not appear to affect the operation of the program The directory into which SIRENA 2 is installed is called in this guide the Sirena base directory Setting up the user directory SIRENA 2 can work with a variety of possible directory arrangements The simplest is to have all working files within the same directory as the SIRENA program itself In this case no command line parameter is needed and when the login screen appears it should be left blank simply clicking OK or typing the ENTER key A second arrangement is to have several possible users each with their own working files including species list with customized groupings
50. ource of this data may be any kind of forest inventory or permanent sample plot The data is organized into text files with an extension FBS using a format that is discussed in more detail on page 8 Sources of data may be of several types For example CODEFORSA uses nventario preliminar or Muestreo diagnostico data at present The size subsampling and associated species list for each type of plot may vary and are defined in a control file called iNvrvP sN2 which resides in the Sirena base directory This is a simple text file which can also be edited to add new sample types as dicussed further below When the Base forestal menu bar 1s selected a screen is displayed as shown below Figure 4 Dialog to select a forest basis file w Bosque para ser utilizado en la base de simulaci n OF x Directorio actual D ASIRENA2 FORESTS Tipo de parcela CODEFORSA Muestreo diagnostico Y Lista de bosques Fase Il Maria Mayela Varela Ulate Fase Il Mario Araya Y Roxana Rodriquez Fase Il Jorge Jimenez e Hilda Romero Anular Mostrar c digos de especies faltantes Aceptar The first field shows the current directory for forest files This can be changed by clicking the Cambiar button A directory selection box appears similar to that in Figure 2 but with the heading Buscar archivos de inventario An alternative directory can be selected after which the use is returned to the Base forestal dialog The second field sh
51. ows the current file type Clicking the down arrow at the right will expose a list of available file types Clicking on a member of this list will update the list of displayed files in the third field The third field displays available files of the selected sampling type Clicking one and pressing Acceptar will load that file The loading process takes some moments A count of plot numbers will appear in the status display at the bottom of the screen This can be used to verify that the file is properly structured In particular blank lines in the data may be interpreted as false plots and will reduce the overall estimates of stocking as a result of an overestimated area figure for the sample After loading the dialog screen will disappear The name of the selected sample will appear at the bottom left of the status display All other menu options will then be accessible The processing of species codes When an inventory file is read SIRENA first reads the species list specified by the corresponding inventory types This gives the codes used in the inventory and their corresponding local and latin names the user assigned species group and a growth model number SIRENA searches for the species list in two locations The first is the user directory This will contain any copies of the species list that have been saved by the user with their own system of species grouping If no list is found in the user directory it will then read the
52. re necessary by appending an ellipsis Options appear next to the graph which allow several aspects of its appearance to be changed The graph can show abundance in terms of numbers per hectare or basal area 9o The trees included can be set to include only those above a specified minimum diameter The number of species to be shown can be set In general with more than 15 20 species shown the text for the left axis may become too small to be readable This graph is a very useful indicator for characterising the type of forest being simulated with its predominant species If the model is run while the graph is displayed it will be seen to changed dynamically with each simulated year The year is shown in brackets on the title line Figure 12 The species abundance graph Sin Aprovechamiento Ferlo II A o 0 Otras especies 0 20 o Area basal 24 Size class distribution The graph showing size class distributions by species groups is invoked using the menu bar Distribuci n por clase de tamafio An example is shown in Figure 13 below The size classes used are set from the general options screen as discussed on page 20 The left axis can be set to show trees ha basal area ha or volume ha The graph is updated dynamically as the simulation runs and shows the current year as part of the title Figure 13 Graph of species groups by size classes Ferlo II Sin Aprovechamiento Afio 0 Comercia
53. ree increment analysis but were not used in the basal area functions due to their small size These plots were measured annually over a 4 year period from 1992 In all a total of 64 678 tree records exist in the database covering 624 species There are three distinctive types of forest e The CODEFORSA plots occur mainly in the central northern zone of Costa Rica and are characterised by somewhat lower basal areas typically 20 25 m ha the presence of Vochysia ferruginea as an indicator and the relative absence of Carapa guianensis The forests in which the plots are located are mostly smaller fragments of about 50 100 ha e The Portico plots occur further to the east in higher rainfall areas in larger continuous tracts of forest and are characterised by a high frequency of Carapa guianensis and the absence Vochysia ferruginea e In both types of forests Pentaclethra macroloba is a very common component typically constituting 30 40 of the basal area in many cases Its abundance is generally lower in the Portico forests however e The secondary forests of the COSAFORMA data set are young regrowth stands of various pioneer species A distinctive aspect is the frequency of Cordia alliodora and the relative absence of Pentaclethra Vochysia or Carapa These plots are also located in the central northern zone of Costa Rica The present version of the model is therefore limited in its applicability to these forests of the northern
54. rom year zero If the model has been paused then it will continue from the year at which it paused Pausa halts the model temporarily It can be restarted by clicking on Ejecutar again It will continue from the year in which it was paused While the model is paused harvesting and silvicultural options can be changed the scenario manager accessed and any of the output screens examined Outputs can be printed or copied to the clipboard as required It is also possible to change species groups while the model is running or to import new forest data This is definitely not recommended however and may cause the program to fail If it is desired to do this then the Anular option should be used beforehand or the model allowed to finish its run indicated by FIN on the status line Anular cancels an active simulation and returns the year to zero reloading the base line data for the forest Exporting and printing outputs The menu bars Copiar and Imprimir act on a currently displayed graph if there is one Copiar copies the graph to the Windows clipboard as a metafile This can be imported into Microsoft Word as a picture object and may be edited within Word to a limited extent A number of graphics packages can also import this format and may allow more facile editing Using Word it is simply necessary to click the Copiar bar switch to Word and press CTRL V at the point where the graph is to be pasted in to the document Double clicking on
55. s sese nennen 16 Figure 8 The scenario manager dialog eese eene eene nete ener ener 18 Figure 9 The dialog screen for setting general options sese eee 20 Figure 10 Settings for sensitivity analysis eese eene eene eerte ene trente entrent 21 Figure 11 Options for volume definitions sse 23 Figure 12 The species abundance graph eese eee eene reete trennen 24 Figure 13 Graph of species groups by size classes esses 25 Figure 14 Graph of standing volumes over time by species groups esee 26 Figure 15 Graph of harvested volumes eese eene eene rennen ener nnn 27 Figure 16 Graph of basal area dynamics esses eene rennen trennen nennen 28 Figure 17 Example of the comparisons graph for three scenarios sse 29 Figure 18 Screen showing the table of diameter classes sse 30 Figure 19 Species groups formed around key species sse eee 34 Figure 20 List of key species for Sirena 2 model groups sse 35 Figure 21 Diameter increment and mortality coefficients by species groups sse 36 Figure 22 The basal area increment model esee eene trennen enne 37 Figure 23 The basal area mortality model essere 38 Figure 24 The basal area recruitment model esee tenete eene nennen 39 Figure 25 Model for crown class classification esses 40
56. scattered over the ordination space Other species were then assigned to the groups seeded from these key species on the basis of their distance in ordination space Thus each non key species was attached to a group formed by the nearest key species Figure 19 Species groups formed around key species Mean increment cm yr 3 00 4 22 22 O CROTO SMI oocHouc 2004 za 17 O CROTOKIL 1 50 4 2 w 25 5 25 O CECRO OBT 10 O V CHY ALL E PETI ga 25 15 155 O TERMI ANA 15 1010 1 00 5 3 5 15 O TAPIR GU 5 15 i O VOCHY FER aa 18 918 8 et PRET PAN P VIROL Kos 4 ane cay my 2a san gu o QuaLe PAR piPTE PA CARAP GUI O PTERO OFF 0 50 4 vues d NU ER on b CORD ALL bl H 7 O TABEB R S 7 m d Ad SOS Ao y Vepend 7 p CEN PERD A ia E B DR GABRos RUB 40 60 80 100 120 140 160 180 200 O PENTA MAC 4 Maximum diameter cm yr 34 The results of this ordination are shown in Figure 19 The key species that seed each group are shown by a circle with their abbreviated species name The small numbers show the position of other species associated with each group The table below lists the groups with their key species and other statistics Figure 20 List of key species for Sirena 2 model groups Group Key species Trees Dmax Dinc Nsp Palms Pterocarpus officinalis Carapa guianensis Pentaclethra macroloba Dipteryx panamensis Vochysia ferruginea Vochysia allenii
57. t can be restarted from the Ejecutar menu bar The fourth button causes a new set of treatment options to be loaded as soon as the first treatment specified has been performed A more detailed account of these choices is given on page 15 Specifications for trees to be treated The table shown within the box labelled Di metros y especies on the Tratamiento screen see Figure 7 allows the selected trees to be specified Trees are selected by species group Grupo field and must fall within a range specied by minimum and maximum diameter Dmin Dmax fields Of the eligible trees only a certain percentage will normally be treated 6raleo field For commercial species it is unlikely that thinning treatment would be applied except to trees of bad form or seriously damaged trees from past logging This case can be selected by clicking the mouse on the S D field It is not possible to combine on the same line a phytosanitary specification with a percentage thinning but it is possible to enter the same group on two different lines In this case for example one could specify the thinning of all defective trees and then a percentage of the remainder on a second line If the Dmax field is left blank then all trees above the specified Dmin will be eligible for thinning If both Dmin and Dmax are blank then all trees will be eligible for thinning irrespective of size As with the Aprovechamiento screen data can be stored in a memory by pressing De
58. t the scenario manager has been used beforehand see page 18 Ifthis is the case then the pull down list will contain a series of codes which are the file names for scenarios Selecting a code will cause the program to switch to that felling regime after the first harvest in the current specification This process allows several quite different harvesting schedules to be linked Note that when a chained scenario is loaded only the harvesting specifications are updated any silvicultural treatment will remain the same Silvicultural details can be chained separately and operate in a similar fashion in that they only alter treatment rules and do not affect the harvesting rules 15 Specification of silvicultural treatments The dialog screen The dialog screen for specifying silvicultural treatments is invoked by clicking the menu bar Tratamientos silviculturales It has the same general layout and functions as the harvesting dialog with analogous options to control the timing of treatment and the affected species groups Figure 7 Dialog for silvicultural treatment options w Opciones de tratamiento i OF x r Cronograma de tratamientos Opciones especiales Tratamiento inicial antes aprovechamiento afio C Sin tratamiento v Tratamiento despu s del aprovechamiento a o U Repetir tratamiento despu s a os Pausar despu s tratamiento Di metros y especies Grupo Dmin Dmax raleo S D 40
59. text file that is readable by Excel It is necessary before doing this to assign a scenario name to the simulation using the scenario manager The check box labelled Escribir cuadro a archivo will then become accessible A default file name consisting of the scenario code followed by the year as an extension will be shown in the text box This name can be edited manually When the check box is clicked the file is written and at the bottom of the screen the full output path and file name will be shown The file contains heading information giving the forest and scenario name the statistic used N ha volume or basal area and the type of trees included acceptable defective or both as well as the current year It then lists each species with its code name and group and the stock by diameter classes This table can be imported into Excel very easily using default options The file is opened as if it were a spreadsheet A Wizard will then appear Select immediately the Finished button at the lower right to indicate acceptance of all default settings and the table will be imported correctly The table can then be formatted edited and sorted as required and saved as an Excel workbook under another name If the screen is displayed while the model is running a simulation then the displayed values will be updated automatically for growth mortality etc The model can be paused and the table saved to file as required Each time
60. the graph will invoke the Word picture editor for minor embellishments The Imprimir function sends a high quality image of the graph to the system default printer These functions work only on graphs They do not print or copy either the stand table or any of the input screens The stand table can be output via Excel as discussed on page 30 above Exiting from Sirena 2 Clicking the Salir menu bar causes the program to be terminated This can also be effected by closing the main program window or with the ALT F4 key 32 Technical basis of the model Introduction It is not necessary to read this section in order to use SIRENA It is intended to provide background information on how the model works A more technical discussion of cohort modelling in general will be found in Alder 1995 The data base The basis for analysing the growth functions in the model were 9 1 ha and 27 4 ha plots established by CODEFORSA since 1991 with an average remeasurement period of about 4 years and 17 1 ha plots established by Portico SA since 1989 measured over about 7 years in all cases with 2 or 3 intermediate measurements The same data was used in the SIRENA I model but in this instance intermediate measurements have been used and the latest remeasurements for 1995 and 1996 added to the data set In addition to this COSAFORMA provided measurements from 34 plots of 600 m established in secondary forest These plots were included in the t
61. the lower canopy will tend to increase Above the standing basal areas are shown thin sections which indicate recruitment increment and basal area losses through mortality Harvesting and treatment are shown as triangular peaks on the bottom axis of the graph which are deducted from the standing basal area 27 It will be noticed that the line between upper canopy and lower canopy basal areas tends to be somewhat jagged This is a function partly of the harvesting process in this instance as trees are removed or destroyed but it is also a function of the frequency with which crown classes are recalculated The options screen Figure 9 allows the frequency for recalculating tree canopy positions to be adjusted This is quite a slow process within the model and typically can be done on a 5 year cycle However if a smoother more natural looking graph is required it can be set to an annual basis This will however slow the model down and does not much influence overall accuracy Figure 16 Graph of basal area dynamics Aprovechar cade 10 afios 50cm 40 rem 3 m2 ha lim Ferlo II Bl Aprovechamiento B Tratamiento B Dosel superior Area basal m2 ha 15 ll Dosel inferior B Defectuosos Ingresos Crecimiento Mortalidad Regardless of the adjustment frequency set crown positions are always recalculated following any disturbance of the stand due to logging or treatment Some trees will then be promoted
62. to upper canopy status to compensate for losses in effect simulating the process of gap formation Comparison of scenarios When successive simulations are run the volumes for a defined component of the forest in each trial are plotted on the comparison of scenarios graph This is invoked from the menu bar by clicking on Comparaciones de escenarios The actual volumes included in this graph are determined from the settings in the Opciones general Volumen screen discussed on page 23 This determines the lower diameter limit for the volume shown and the list of species groups to be shown It is important that the names of the species groups correspond exactly with current groups as otherwise nothing may appear and the comparisons line will simply lie along the zero axis These names are cases sensitive so that capital and small letters must correspond 28 The example shows three regimes The first is for the forest without harvesting This shows a rise in volumes over 50 cm to about 100 m3 ha after some 35 years where it stabilises The second corresponds to the preceding examples with 3 m2 ha being removed on a 10 year cycle The third is an attempt at a more sustainable method of harvesting The retention percentage is increased to 60 and felling is scheduled when the stand reaches 25 m2 ha but removals are still 3 m2 ha It shows a very similar pattern to the second regime with declining large tree volumes over time as the stock is e
63. treatments within a cycle This gives the program almost complete flexibility in the way treatments are specified The scenario manager can also be called up at any time while a simulation is running to introduce an immediate change in the management regime This can be useful for testing variable treatments and schedules in an ad hoc fashion Using the scenario manager The scenario manager is called from the menu bar manejo de escenarios The dialog screen shown in Figure 8 appears Figure 8 The scenario manager dialog amp Manejo de escenarios Iof xi Titulo de escenario seleccionado Archivo aprovechar cade 10 afios 50cm 40 rem 3 m2 ha lim A1 0D50 Escenarios en archivo Ciclo 10 anos solo Gavilan cans Tratar 50 Gavilan 30 cm C Aprovechar Ciclo 15 anos solo Comercial C Tiat Ciclo 10 afios 40 cm 30 rem Jai Aprovechar cada 10 anos arboles de 50 cm Ambos Ciclo 10 anos solo Gavilan amp Aprovechar cade 10 a os 50cm 40 rem 3 m2 ha lim Suprimir Hu Guardar A scenario is created by defining harvesting and treatment specifications using the aprovechamiento and tratamiento screens discussed on pages 12 and 16 Once this has been done the scenario manager is invoked from the menu The title box will appear initially blank A descriptive title of the regime should be typed in bearing in mind that 18 this title will appear as a heading on output graphs and tables from the model The scen
64. y It is more likely that the login screen will appear correctly but some other file access commands will fail at a later time with a system message such as Error 53 File not found If installation has been done correctly the localization dialog should not appear before the login screen Figure 2 The file selection dialog box used to locate a directory Localizar DEFAULTS SN2 x amp d DATA codforsa csv codforsa spl C document 7 C forests N O install 3 old versions CI setupwiz fin339x xls forbase frm genopt frm harvest frm harvest frs idioma sn2 x d sirena2 Cancel The menu system Once the login screen has been negotiated the program s system of menus will be acessible These appear as shown in Figure 3 The functional groups of menu choices under the main headings are as follows Escenarios This group defines and reads in the forest data to be used as a basis for simulation establishes options for harvesting and silviculture allows the definition of species groups and sets general options for program control These details can be saved as either program defaults for the general options or as user defined species lists for grouping methods or as scenarios for combinations of harvesting and silviculture Gr ficos This displays one of a number of possible charts or tables showing the state of the forest as the simulation progresses T
65. y position to each cohort in order to select the appropriate increment and mortality coefficient The crown positions are simplified into two categories Lower canopy comprising crown classes 1 3 and upper canopy including crown classes 4 and 5 The model uses the relationship shown in Figure 25 to determine the basal area which is comprised of lower canopy trees given the total stand basal area Individual cohorts are sorted by size and allocated to the upper canopy until the quota of available basal area is satisfied The smaller diameter trees are then assigned to the lower canopy Defective trees are included as part of the lower canopy in the regression and in its application within the model although they draw on their own distinct mortality and growth coefficients 39 Figure 25 Model for crown class classification 25 00 4 O CODEFORSA 4 A Portico CODEFORSA only T Sa 4 20 00 0 Both 6 o A 0 a A y 0 6434x 3 6292 R 0 603 15 00 4 10 00 4 Basal area of crown classes 0 3 0 00 5 00 10 00 15 00 20 00 25 00 30 00 35 00 40 00 Plot basal area Logging damage The logging damage functions are based on surveys carried out by CODEFORSA with ODA assistance which related the volume removed from seven farms to the area impact in terms of gaps and trails and to the percentage of severely damaged residual trees Simple linear regressions were used as the data was too sparse
66. zone of Costa Rica 33 The formation of species groups The formation of species groups is a necessary first step in analysing growth functions for any tropical forest data This is because there are a large number of species many of which are represented in the data by only a few individuals For the SIRENA 2 data methods were tried based on statistical comparisons of regressions between species Initially species that were rare were grouped by genera to achieve at least 5 or 6 individuals A logarithmic regression of basal area increment on tree basal area was then fitted Using as a discrimination factor a 95 probability that two groups differed some twenty groups were formed containing very unequal numbers of trees and combining species with no apparently similar attributes The method was not felt to be very useful as it did not combine naturally species that ought to be similar such as fast growing shorter lived species or slower growing longer lived types As with past modelling efforts with SIRENA 1 and the CAFOGROM models for Brazil a more pragmatic method was eventually used in which mean increment of upper canopy trees classes 3 5 was graphed against the largest observed individual for each species This naturally groups short lived fast growing ephemerals and larger slower growing forest trees In order to make the grouping as consistent as possible with common sense the groups were formed by identifying key species

User's Guide for SIRENA II - Denis Alder

Contents

Download Pdf Manuals

Related Search

Related Contents