LANDIS-II Biomass Succession Extension (v1.0) User's Guide

Transcription

1 LANDIS-II Biomass Succession Extension (v1.0) User's Guide Robert M. Scheller James B. Domingo University of Wisconsin-Madison Last Revised: February 22, 2006

2 Table of Contents Introduction Biomass Shade Calculation Cohort Reproduction Cohort Growth and Ageing Cohort Senescence and Mortality Dead Biomass Decay Initializing Biomass Interactions with Age-Only Disturbances References Acknowledgments Input File Rules Input File Parameters Seeding Algorithm Shade Table Biomass Succession Species Parameters Establishment Probabilities Maximum Growth Rates Leaf Litter Decay Rates Age-Cohort Disturbance Link Page 2 of 10

3 Introduction This document is a member of the set of documents that describes how to use the Biomass Succession Extension to the LANDIS-II model. The purpose of this document is to describe the Biomass Succession Extension for Ecologists who want to learn how to implement the model as computer software. Therefore, this document omits general, conceptual details about the Biomass Succession Extension or the LANDIS-II model. For that information, please see the references below. The Biomass Succession Extension generally follows the methods of the Age-Only Succession: Cohorts are reproduced, they age, and die. However, in addition to cohort age and species information, biomass (kg/ha) is added. Therefore, cohorts must have an initial cohort density and this density changes over time. The Biomass Succession Extension also changes the calculation of shade. Lastly, the Biomass Succession Extension tracks dead biomass change over time, divided into two pools: woody and leaf litter. Information about the Age-Only Succession Extension can be found in a separate document. 1.1 Biomass Shade Calculation Site shade is calculated based on a percentage of the maximum biomass possible for each ecoregion. 1.2 Cohort Reproduction Cohort reproduction is the addition of a cohort, aged 1 year, and is given an initial biomass. (-1.6 * Bsum / Bmax) initial biomass = x B max x e where B max is the maximum biomass possible for each ecoregion and BBsum is the current total biomass for the site (not including other new cohorts). Initial biomass must be 1 (kg / ha); if < 1, initial biomass is set equal to 1. Note: this initial cohort will be grouped ( binned ) appropriately into a larger cohort (e.g., 1 10) at the next successional time step. Page 3 of 10

4 1.3 Cohort Growth and Ageing Cohort net growth is based on the principles outlined in Scheller and Mladenoff (2004). Cohort net growth takes into consideration the age of the cohort, species, ecoregion, and competition. Cohort net growth is gross growth minus development-related mortality. Cohort ageing is simply the addition of the time step to each existing cohort. 1.4 Cohort Senescence and Mortality As a cohort nears its longevity, there will be additional mortality and a loss of biomass. This is called age-related mortality. The biomass will decline to near zero at the maximum life span. Cohorts are not randomly killed as in Age-Only Succession. If a cohort exceeds the longevity for that species, then the cohort dies. 1.5 Dead Biomass Decay When a cohort dies or either development-related or age-related mortality occurs, this biomass is added to one of the two dead biomass pools: woody and leaf. There is a mean decay rate for each pool at each site, determined by using an average of the input decay rates and the current pool decay rate, weighted by biomass. Disturbances can alter the dead biomass pools. They can add dead biomass (e.g., wind) and/or remove dead biomass (e.g., fire will add some woody dead biomass and remove all leaf dead biomass). 1.6 Initializing Biomass At the beginning of a scenario, the initial communities must be given appropriate living biomass values and dead biomass estimated for the site. The Biomass Succession extension iterates the number of time steps equal to the maximum cohort age for each site. Each cohort undergoes growth and mortality for the number of years equal to its initial cohort age. Beginning with the oldest cohort, cohorts are added according to their age. Therefore, biomass initialization includes competition between cohorts. Cohort biomass initialization does not account for disturbances that would likely happen prior to initialization and therefore overestimates initial live biomass and underestimates initial dead biomass quantities. Page 4 of 10

5 1.7 Interactions with Age-Only Disturbances Biomass Succession was written to allow disturbances that operate on any cohort type to interact with the two dead biomass pools. For example, a User is able to run the Base Fire or Base Wind extensions with Biomass Succession. Although neither extension is biomass aware, a simple interface was created that enables the cohort s biomass if killed by the disturbance - to be allocated to dead biomass pools. The interface allows a User to indicate a) whether and how much non-woody or woody live biomass is tranferred to their respective dead pools by a disturbance type and b) whether and how much of the non-woody or woody dead biomass pools are removed by a disturbance type. For example, if a fire kills a cohort, we would expect that all of its non-woody and some of the woody biomass to be volatalized immediately and this biomass would not enter a dead biomass pool. In addition, we would expect some of the existing woody dead biomass pool to be volatalized during a fire and perhaps all of the existing nonwoody biomass pool (i.e., the forest floor) to be volatalized. This interface does not allow dynamic changes in the transfer rates into and out of the dead pools. Rather, the interface was designed to allow existing age-cohort disturbances to be used with Biomass Succession. 1.8 References Scheller, R. M. and Mladenoff, D. J. A forest growth and biomass module for a landscape simulation model, LANDIS: Design, validation, and application. Ecological Modelling. 2004; 180(1): Acknowledgments Funding for the development of LANDIS-II has been provided by the North Central Research Station (Rhinelander, Wisconsin) of the U.S. Forest Service. Valuable contributions to the development of the model and extensions were made by Brian R. Sturtevant, Eric J. Gustafson, and David J. Mladenoff. Page 5 of 10

6 2 Input File Rules The input rules for age-only succession are identical to those of the LANDIS-II Core Model. Please see the LANDIS-II Model User s Guide for further instruction. 3 Input File Parameters The first parameter is the title of the input file: LandisData "Biomass Succession" The second parameter is the time step in years. For example: Timestep Seeding Algorithm The third parameter is the seeding algorithm to be used. Currently, there are three options as described in the LANDIS-II Description document. The three options are random, universal, and WardDispersal. SeedingAlgorithm WardDispersal 3.2 Shade Table Site shade is calculated based on a percentage of the maximum biomass possible for each ecoregion. A table PercentFullSunTable provides these data. For every shade class (1-5 must be present) and for every active ecoregion, a percent of the maximum biomass for the ecoregion must be given. The maximum biomass for an ecoregion is the maximum growth rate of all species X 30 (equation 2, Scheller and Mladenoff 2004). Example: PercentFullSunTable >> Shade Percent Max Biomass >> Class by Ecoregions >> eco1 eco2 1 25% 20% 2 35% 30% 3 45% 40% 4 60% 50% 5 95% 80% Page 6 of 10

7 3.3 Biomass Succession Species Parameters For each species, the Biomass Succession extension requires three (3) additional parameters (Table 1). Table 1. Biomass Succession species parameters. Parameter Data Type Units species name string leaf longevity number > 0 years woody biomass decay rate 0 number 1 mortality curve parameter 5 number 25 The mortality curve shape parameter determines how quickly agerelated mortality begins. If the mortality parameter = 5, then agerelated mortality will begin at 10% of life span. If the mortality parameter = 25, then age-related mortality will begin at 85% of life span. Here is an example of the BiomassParameters table: BiomassParameters >> Species Leaf WoodyBiomass Mortality >> Longevity DecayRate ShapeParam >> abiebals acerrubr acersacc Establishment Probabilities Recall that sites with similar abiotic conditions are grouped into a single land type or ecoregion. Each ecoregion requires a probability of species establishment (PEst) for each species (Table 2). These data are contained within a table. First, the table is identified with the table name: EstablishProbabilities. Following the table name is a row containing ecoregion names. These names must match the names provided in the Ecoregion Input File (see LANDIS-II Model Users Guide). The ecoregion names need not be in any order nor do all of the ecoregion names need to be present. If an ecoregion is not listed, the default PEst is 0.0 for all species. An ecoregion may have no establishment probabilities for any species Page 7 of 10

8 because it represents a region of the landscape which does not have forested sites (for example, bodies of water, urban areas). Next, there is one row per species. Each row begins with the species name followed by the PEst for each ecoregion name listed above, in the order given by the list of ecoregion names, above. Not all species need to be listed nor do they need to be in order. Species not listed will have a default PEst of 0.0 for all ecoregions. Here is an example of the EstablishProabilities table: >> Species Ecoregions >> Eco3 Eco4 Eco10 Eco19 pinurigi acersacc poputrem Table 2. Species by ecoregion parameters. Parameter Data Type Units species name string establishment probability species 0 number 1 maximum growth rates 0 number kg / year / ha 10,000 leaf litter decay rate 0 number Maximum Growth Rates Each ecoregion also requires a maximum growth rate for each species (Table 2). These data are contained within a table. First, the table is identified with the table name: MaxGrowthRates Following the table name is a row containing ecoregion names. These names must match the names provided in the Ecoregion Input File (see LANDIS-II Model Users Guide). The ecoregion names need not be in any order nor do all of the ecoregion names need to be present. If an ecoregion is not listed, the default maximum growth rate is zero (0) for all species. An ecoregion may have no growth rates for any species Page 8 of 10

9 because it represents a region of the landscape which does not have forested sites (for example, bodies of water, urban areas). Next, there is one row per species. Each row begins with the species name followed by the maximum growth rate for each ecoregion name listed above, in the order given by the list of ecoregion names, above. Not all species need to be listed nor do they need to be in order. Species not listed will have a default maximum growth rate of zero (0.0) for all ecoregions. Here is an example of the MaxGrowthRates table: MaxGrowthRates >> Species Ecoregions >> eco1 eco2 abiebals acerrubr acersacc betualle Leaf Litter Decay Rates Finally, each ecoregion requires a leaf litter decay rate for each species (Table 2). These data are contained within a table. First, the table is identified with the table name: LeafLitterDecayRates Following the table name is a row containing ecoregion names. These names must match the names provided in the Ecoregion Input File (see LANDIS-II Model Users Guide). The ecoregion names need not be in any order nor do all of the ecoregion names need to be present. If an ecoregion is not listed, the default decay rate is zero (0.0) for all species. An ecoregion may have no decay rates for any species because it represents a region of the landscape which does not have forested sites (for example, bodies of water, urban areas). Next, there is one row per species. Each row begins with the species name followed by the leaf litter decay rate for each ecoregion name listed above, in the order given by the list of ecoregion names, above. Not all species need to be listed nor do they need to be in order. Species not listed will have a default decay rate of zero (0.0) for all ecoregions. Page 9 of 10

10 Here is an example of the LeafLitterDecayRates table: LeafLitterDecayRates >> Species Ecoregions >> eco1 eco2 abiebals acerrubr acersacc betualle betupapy Age-Cohort Disturbance Link The table that enables age-cohort-disturbances (those that are not inherently biomass aware, such as Base Wind, Base Fire, Base BDA) to work correctly with Biomass Succession dead pools is indicated in the optional parameter, AgeOnlyDisturbanceParms. A file name is expected, e.g., AgeOnlyDisturbanceParms biomass_age-disturbances.txt Within this file, a table is expected with a form similar to other LANDIS-II input files. First, is the LandisData parameter indicating the file type expected: LandisData "Age-only Disturbances - Biomass Parameters" Next, is a table listing the Live Biomass and Dead Pool Reductions expected for each disturbance type. Data must be entered as integers with the % indicator. Notice that default disturbance behavior can be defined using the {default} keyword. >> Live Biomass Reductions Dead Pool Reductions >> Disturbance Woody% Non-woody % Woody Non-woody >> fire 33% 100% 8% 100% wind 0% 0% 0% 0% harvest 85% 0% 0% 0% (default) 15% 0% 0% 0% Page 10 of 10