Weaver User`s Manual
Contents
1. gabi gabi home simulations File Edit View Search Terminal Help Animal aca6 Animal ara6 Animal geo2 Animal species oril eats Fungus Fungus_X Giving life to animals DONE Saving Water volume as generalist 20x20 restrict001 WITH PREDS rep2 Snapshots Wa ter initial day 0 dat DONE Saving Fungus as generalist 20x20 restrict001 WITH PREDS rep2 Snapshots Fungus i nitial Fungus X day 0 dat DONE Running on day 0 out of 200 Running on day 1 out of 200 Running on day 2 out of 200 Running on day 3 out of 200 Running on day 4 out of 200 Running on day 5 out of 200 Figure 2 Running simulation example NOTE Simulations can be run in any up to date GNU Linux platform However be aware that output files and memory requirements may be quite demanding for your PC depending on the simulation parameters IN B Configuration file structure This section explains in detail the usage of every parameter included in the run_params configuration file which is loaded when running each instance of the program This file is written in an open standard format used to transmit data object JSON JavaScript Object Notation Reading and writing in this format is a simple and intuitive task and it is also easy to interpret manipulate and generate the files JSON syntax and basic data types can be viewed in http www json org A fragment from a run_params configuration file is shown below It is relatively e2. 1 0 probabilityDeathFromBackground 0 0 deadlyTank 0 1 assignedForGrowth 0 9 numberOflnstars 4 percentOfTimeSinceLastInstar 0 9 alphaForPredation 1 0 minRandomForEncounters 0 337 maxRandomForEncounters 0 340 minRandomForPredation 0 606 maxRandomForPredation 0 617 maxEncountersT 10 maxSearchAreaT 10 Q10phenology 0 35 Q10digestion 0 25 meanSizeHunted 20 08 gt sdSizeHunted 7 5 meanVorHunted 13 96 sdVorHunted 4 9 meanSpdHunted 40 65 sdSpdHunted 15 meanSizeHunter 20 08 sdSizeHunter 7 5 meanV orHunter 13 96 sdVorHunter 4 9 meanSpdHunter 40 65 sdSpdHunter 15 meanSearchAreaHunter 0 0 sdSearchAreaHunter 0 0 meanSizeX Size 806 4 sdSizeXSize 275 meanVorX Vor 389 6 sdVorX Vor 140 meanSpdRatio 25 5 sdSpdRatio 9 meanSizeRatio 397 2 sdSizeRatio 140 minTraitsRanges energy_tank 0 25 growth 1 35 pheno 3 body_size 0 001 assim 0 7 voracity 0 55 speed 0 1 search area 0 1 met rate 0 6 vorQ10 3 spd Q10 1 5 srchQ10 2 0 actE_met 0 33 maxTraitsRanges energy_tank 0 5 growth 1 45 pheno 11 body_size 0 003 assim 0 9 voracity 0 75 speed 0 3 search area 0 4 lu met
3. Network Browse Net lib selected containing 4 items Figure 1 Files location example Step 2 After verifying that we have the two files plus the subdirectory lib placed into the same directory a Weaver3D simulation can be run just using the stardard execution process the user has to type in the command Weaver3D without using any added standard input parameter However it is recommended to type the following command prior to the execution so the libraries will be found correctly at running time export LD_LIBRARY_PATH home lt user gt simulations lib where lt user gt refers to the user name in the computer This absolute path then corresponds with the lib subdirectory mentioned in step 1 Using this setup all the information needed is collected at running time from the run_params configuration file which is explained in detail in the section B of this document Step 3 The simulation starts running and the user can read some information about the progress on the standard output as observed in the figure 2 While the simulation runs some output files are generated in a new subdirectory which is created inside the very same running directory This new subdirectory is named as indicated in the configuration file run_params using the parameter namedoutputDirectory A new set of output files and subdirectories is then created inside the new subdirectory They are list in detail in the section C of this document
4. example shows indicates that all voxels will be initialized to a 25 of water content moisture patches type homogeneous value 25 ee L Spherical All voxels within the defined sphere are assigned the same value Next example defines a spherical water patch centered on voxel with coordinates x y Z 100 100 100 a radius of 10 voxels and a moisture of 25 moisture patches type spherical radius 10 xPos 100 yPos 100 ZPos 100 value 25 kan L Gaussian Similar to the spherical patch but with water values decreasing with radius following a 1D 2D or 3D gaussian distribution Therefore the center of the patch will have a maximum water quantity The amplitude value tells how much water will exist in the middle of the patch while the sigma value defines how fast this value decreases with radius moisture patches type gasussian radius 10 xPos 100 yPos 100 zPos 100 amplitude 25 sigma 3 Frei KO L Random Gaussian Here we let the program decide the physical distribution of numberOfPatches water patches following randomized Gaussian distributions We can decide whether the sigma values are randomized through useRandomSigma set to true or false but with a maximum of maxSigma Amplitude behaves the same way with useRandomAmplitude and a maximum of maxAmplitude Location and patch rad
5. Matrices This file is generated upon completion of the simulation and its called predationOnSpecies It stores a NxN matrix where N represent the number of different animal species This matrix stores an all to all predation index with predators forming columns and preys forming rows References Moya Larafio J Verdeny Vilalta O Rowntree J Melguizo Ruiz N Montserrat M Laiolo P 2012 Climate Change and Eco Evolutionary Dynamics in Food Webs Advances in Ecological Research 47 1 Moya Larafio J Foellmer M W Pekar S Arnedo M A Bilde T Lubin Y 2013 Evolutionary ecology linking traits selective pressures and ecological functions in Spider research in the XXI century trends and perspectives ed D Penney Siri Scientific Press Manchester UK pp 122 153 Moya Larafio J Bilbao Castro J R Barrionuevo G Ruiz Lupi n D Casado L G Montserrat M Melian C J Magalh es S 2014 Eco evolutionary spatial dynamics rapid evolution and isolation explain food web persistence Advances in Ecological Research 50 75 144
6. a vector which establishes which fungus species each fungivore can feed upon Since acal is a predator this vector is empty initialPopulation 414 This is not in use in the current version Instead each population is established by allometric equations constrained by body size and instar Moya Larafo et al 2014 and only the size of the entire ecosystem in fixed at initialization ecosystemSize restrict 0 01 This is the Phi parameter which varies between 0 and 1 and establishes the degree of genetic variability in traits A low value means high genetic variability Moya Larafio et al 2012 2014 In this version one cannot establish different genetic variability for each of the traits correlationCoeficientRHO 0 1 This parameter varies between 0 and 1 and establishes the degree of genetic correlation among traits within a module In this version one cannot establish different genetic correlations for different modules Module structure positive or negative correlations among traits is identical as in Moya Larafio et al 2012 All simulations in Moya Larafio et al 2014 were run with low genetic correlations 0 1 TA 0 053 B 2 494 These are the constant and scaling coefficients for the Mass Length equation A8 in Moya Larafio et al 2014 which relates shape to body mass forDensitiesGrowth 1 4 forDensitiesEggSize 0 002 These two last parameters are necessary to calculate how many
7. animals of each instar will be present at initialization following abundance body size allometric equations Moya Larafio et al 2014 forDensitiesGrowth is the midpoint between minTraitsRanges growth and maxTraitsRanges growth see below forDensitiesEggSize is the midpoint between minTraitsRanges body_size and maxTraitsRanges body_size see below ecosystemSize 20000 Total number of individuals in the simulation at initialization How this number is scattered across species and instars is established by allometric abundance size contraints Moya Laraiio et al 2014 minCondition 0 01 Lower limit for how body condition affects flexible traits in simulations upper limit is always 1 The body condition state variable the energy tank e divided by the structural body size B Moya Larafio et al 2012 is interpolated from the maximum range min deadlyTank max up_tank where deadlyTank is the energy threshold below which an animal dies e B 0 1 and where up tank is the upper tank threshold max e B calculated from the mass increase necessary to molt to the next instar and this changed into the 1 minCondition scale Note how this allows animals in better condition to have lower c coefficients in equations A4 A6 Moya Larafo et al 2014 maxVoracityT 0 100 Va in equation A7 Moya Laraiio et al 2014 keepForSurv 3 0 coefficient by which the minimum energy available for survival is multipli
8. maxEncountersT 10 Upper limit of encounters with predators in one day This is used to establish energy and assimilation efficiency losses as well as tuning activity anti predatory behaviour 14 Above this number of encounters the effects apply according to the limits of the linear interpolation see Moya Larafio 2012 2014 for details maxSearchAreaT 10 A a in equation A7 Moya Laraiio et al 2014 Q10phenology 0 35 Developmental time and therefore birth date phenology is affected by this parameter depending on temperature using linear interpolation see Moya Larafio 2012 for further details Currently not in use Q10digestion 0 25 meanSizeHunted 20 08 sdSizeHunted 7 5 meanVorHunted 13 96 sdVorHunted 4 9 meanSpdHunted 40 65 sdSpdHunted 15 meanSizeHunter 20 08 sdSizeHunter 7 5 meanV orHunter 13 96 sdVorHunter 4 9 meanSpdHunter 40 65 sdSpdHunter 15 meanSearchAreaHunter 0 0 sdSearchAreaHunter 0 0 meanSizeX Size 806 4 sdSizeXSize 275 meanVorX Vor 389 6 sdVorX Vor 140 meanSpdRatio 25 5 sdSpdRatio 9 meanSizeRatio 397 2 sdSizeRatio 140 means and standard deviations to normalize traits and trait products to include them in equations A9 and A10 Moya Larafo et al 2014 minTraitsRanges energy tank 0 25 growth 1 35 pheno 3 b
9. ERGENERALIST savelntermidiateVolumes false savelntermidiateV olumesPeriodicity 100000 encountersMatrixFilename encountersMatrix predationsMatrixFilename predationsMatrix nodesMatrixFilename nodesMatrix predationEventsOnOtherSpeciesFilename predationOnSpecies C Output files list Each time a simulation is run a bunch of output files can be generated The user decides where to store files and which ones to keep or discard through JSON directives Next there is a description of the most important files that can be obtained from a Weaver simulation O animal_constitutive_traits txt The information related to each animal traits that is born is kept in this file This is used to study genetic variability of populations and its evolution Currently 13 traits are stored for each animal energy growth pheno body assim vor speed search met vorQ10 spdQ10 srchQ10 e_met Other information is also present in this file including its unique identifier id the species the animal belongs to species generation based on parents ones g numb prt1 and g numb prt2 and parents unique identifiers ID_prt1 and ID_prt2 L extendedDailySummary txt A daily summary is kept detailing for each species the number of individuals that fall within a given phenological state Basal resources are also detailed i e fungus biomass The fields for this data file are simulation day day fungus biomass as scient
10. Weaver User s Manual This document presents all the information needed related to Weaver s usage This manual is divided into some sections each one explaining different aspects of the program the first section explains step by step how to run a simulation the second one describes the configuration file structure and the third section lists the main output files A Running a simulation step by step The user has to follow the next steps in order to run a full instance of a simulation Step 1 The executable file Weaver3D must be placed in any directory along with the configuration file run_params It is strictly needed that both files are placed in the same directory so the program Weaver3D will begin by searching for the file named run_params just inside the same directory from where the execution was called In adition that one same directory must also contain the subdirectory lib including the four files provided with the binaries Thus the files location must be similar to the one depicted in the figure 1 gabi gabi home simulations ls run_params gabi gabi home simulations ls lib libboost filesystem so 1 54 0 Libboost_filesystem so 1 55 0 libboost system so 1 55 0 Igabi gabi home simulations simulations Devices FB Home simulations lib Q Search L Datos a Computer ka ri din BB Home run params Weaver3D Ed Desktop EF Documents Downloads W Music m Pictures Videos File System Trash
11. asy to notice how the data and components are distributed inside the file A detailed explanation of each parameter is written after this whole example world soil dimensions depth 10 length 50 width 50 cellSize 1 moisture patches type sphere radius 5 xPos 15 yPos 15 zPos 5 yalue 87 5 add more patches if needed nutrients minC 0 maxC 23 minN O maxN 23 minP 0 maxP 23 temperature 18 maxK 9 6 mink 0 1 timelapseForChemostatEffect 1 thresholdForChemostatEffect 0 05 increaseForChemostatEffect 1 0 Io qs life animals name acal huntingMode active hunting genetics NumberOfLoci 20 NumberOfAlleles 10 NumberOfTraits 13 traitsPerModule 3 k edibleAnimalSpecies geol litl opil ara1 ara2 ara3 ara4 aca1 aca2 aca3 aca4 col1 col2 col3 enc1 enc2 enc3 oril ori2 ori3 edibleFungusSpecies initialPopulation 414 restrict 0 01 correlationCoeficientRHO 0 1 A 0 053 B 2 494 forDensitiesGrowth 1 4 forDensitiesEggSize 0 002 ecosystemSize 20000 minCondition 0 01 maxVoracityT 0 100 keepForSurv 3 0 maxReproductionEvents 5 assignedForReproduction 0 9 forControlingDryBodyMass
12. ed and the result substracted from the energy budget e in order to decide the total budget available for reproduction maxReproductionEvents 5 maximum number of egg batches laid per one individual in its lifetime assignedForReproduction 0 9 when the animal reaches its maximum instar this parameter defines the amount of energy needed in order to reproduce If the animal has enough energy its life state changes to reproducing forControlingDryBodyMass 1 0 this parameter changes depending on whether the mass length equation A8 in Moya Larafio 2014 has been applied based on wet 1 or dry mass 0 3 assuming a 70 water content probabilityDeathFromBackground 0 0 background mortality for reasons others than starvation or predation deadlyTank 0 1 e B threshold below which death from starvation occurs assignedForGrowth 0 9 proportion of accumulated energy which is invested in molting numberOflnstars 4 Number of instars from egg to adult percentOfTimeSinceLastInstar 0 9 time threshold beyond which if the target energy for molting has not been reached the animal molts anyway It is a fraction of the time spent in the former instar pp 131 132 in Moya Laraiio et al 2014 minRandomForEncounters 0 337 maxRandomForEncounters 0 340 minRandomForPredation 0 606 maxRandomForPredation 0 617 P values to decide encounters and predation p 134 in Moya Larafio et al 2014
13. ificName_biomass and the number of individuals at any phenological state as scientificName_X where X represents the state 0 to be born 1 active 2 starved to death 3 predated 4 reproducing 5 background death 6 natural death CI animals each day end This is the most complete set of information regarding each individual A file is kept for each day containing detailed information for them Its structure is as animals_day_X txt where X represents a given day Each file is written after a simulation day is finished and contains a summary of each animal currently participating in the simulation unique individual id id its species species spatial location x y z phenological state state instar for reproduction purposes instar the initial phenology value pheno_ini date of creation date_egg first reproduction date age_first_rep number of reproduction episodes rep_count number of offspring to date fecundity death date date_death its generation number based on parents ones g numb prt1 and g numb prt2 parents unique ids ID prt1 and ID_prt2 the number of encounters with predators this day encounters pred the accumulated number of encounters with predators along its life global_pred_encs time to finish digestion days_digest and all variables regarding genetic traits that can change values along the simulation energy growth pheno body assim vor speed search met vorQ10 spdQ10 srchQ10 e_met L
14. ius will be always randomized moisture patches type radomGaussian numberOfPatches 10 useRandomSigma true useRandomAmplitude true maxSigma 10 maxAmplitude 100 patchesRadius 35 kas In Moya Larafio et al 2014 water pockets have a spherical shape sphere with 5 cells of radius The xPos yPos zPos refer to the position coordinates of the center of the sphere in the world value is the relative humidity of the water pocket which in this version is fixed through the entire simulation These islands with water available hold the only cells in which fungi can grow Everywhere else in the world there are no basal resources growing Nutrient availability nutrients minC 0 maxC 23 minN O maxN 23 minP O maxP 23 hy Not yet in use Temperature at which the simulation occurs temperature 18 Temperature in Celsius In this version temperature is constant through the entire simulation Carrying capacity of each cell maxK 9 6 minK 0 1 Maximum and minimum fungus carrying capacity per cell In Moya Larafio et al 2014 each cell was set at 0 99 maxK at initialization Chemostat parameters timelapseForChemostatEffect 1 How many days elapse between renewal pulses of basal resources fungi In Moya Larafio et al 2014 this was every day 1 thresholdForChemostatEffect 0 05 This la
15. ody_size 0 001 assim 0 7 voracity 0 55 speed 0 1 search area 0 1 met rate 0 6 vorQ10 3 spdQ10 1 5 srchQ10 2 0 actE_met 0 33 maxTraitsRanges energy_tank 0 5 growth 1 45 pheno 11 body_size 0 003 assim 0 9 voracity 0 75 speed 0 3 search area 0 4 met rate 0 8 vorQ10 4 spdQ10 2 5 srchQ10 2 5 actE met 0 42 Standard trait ranges Ly Uy from which actual trait ranges are obtained in combination with the parameter restrict p above Moya Larafio et al 2014 p 124 minTraitLimits energy_tank 0 01 growth 1 01 pheno 1 body_size 0 00001 assim 0 1 voracity 0 4 speed 0 05 search area 0 05 met rate 0 4 vorQ10 1 spdQ10 1 srchQ10 1 actE_met 0 2 axTraitLimits energy_tank 1 growth 2 0 pheno 100 body size 394 7 assim 1 voracity 0 8 speed 0 35 search area 0 5 met rate 0 9 vorQ10 6 spdQ10 BIN srchQ10 3 actE met 0 9 b Evolutionary trait limits II K beyond which trait values cannot evolve Moya Larafio et al 2014 p 124 Any individual surpassing this threshold for any trait is locked at the limit value for the trait Output choices runDays 200 outputDirectory spheres_chemo1_distant_HYP
16. r trophic level omnivory animals feeding on more than one trophic level is allowed within the vector edibleAnimalSpecies see below genetics NumberOfLoci 20 NumberOfAlleles 10 NumberOfTraits 13 traitsPerModule 3 b NumberOfLoci is the number of loci involved in each trait To calculate the value of each trait for one individual the values of all these loci are summed up assuming perfect codominance NumberOfAlleles refers to the total number of different alleles present per locus in the entire population NumberOfTraits refers to the total number of traits with quantitative genetic basis for each particular species traitsPerModule is the number of traits involved in each module for which one can play with the degree of genetic correlation from pleiotropic effects among the traits correlationCoeficientRHO NumberOfTraits and traitsPerModule must not be modified for now as this version only supports computation for the current traits Further details can be seen in Moya Larafio et al 2012 2014 edibleAnimalSpecies geol litl opil ara1 ara2 ara3 ara4 aca1 aca2 aca3 aca4 col1 co 12 col3 enc1 enc2 enc3 oril ori2 ori3 This is a vector which establishes the directed links of each predator to its prey Since connectance is maximal in this example this predator is linked to all other species in the food web edibleFungusSpecies This is
17. rate 0 8 vorQ10 4 spdQ10 2 5 srchQ10 2 5 actE_met 0 42 minTraitLimits energy_tank 0 01 growth 1 01 pheno 1 body size 0 00001 assim 0 1 voracity 0 4 speed 0 05 search_area 0 05 met_rate 0 4 vorQ10 1 spdQ10 1 srchQ10 1 actE_met 0 2 maxTraitLimits energy_tank 1 growth 2 0 pheno 100 body size 394 7 assim 1 voracity 0 8 speed 0 35 search area 0 5 met rate 0 9 vorQ10 6 spdQ10 3 srchQ10 3 actE_met 0 9 add more species if needed fungi name Fungus_X sporeMass 4 minimumFungus 0 ACTIVATION ENERGY 0 68 NORMALIZATION_B 25 98 minHR 85 maxHR 90 In maxRScale 0 5 zeroFungi 8 patches type sphere radius 5 xPos 15 yPos 15 zPos 5 value 1 0 add more patches if needed b simulation runDays 200 outputDirectory spheres_chemo1_distant__HYPERGENERALIST savelntermidiateVolumes false savelntermidiateVolumesPeriodicity 100000 encountersMatrixFilename encountersMatrix predationsMatrixFilename predationsMatrix nodesMatrixFilename nodesMatrix predationEventsOnOtherSpeciesFilename predationOnSpecies Here we explain in detail what is the
18. st parameter works if and only if timelapseForChemostatEffect is set to 0 and then renewal occurs when basal resources are below thresholdForChemostatEffect times the initial biomass across the entire world which is the sum of the 0 99 maxK across all cells in the world increaseForChemostatEffect 1 0 This parameter determines which proportion of the initial basal resources i e sum of the 0 99 maxK across all cells in the world are added to the world each time resources are renewed i e at each pulse Description of animal parameters As an example we explain the details of acal predatory mite species number 1 for the simulation with maximum connectance 0 55 and maximum genetic variability restrict of 0 01 in Moya Larafio et al 2014 life animals name acal huntingMode active hunting hunting mode includes three types of hunting active hunting Predators include animals that actively search for their prey or sit and wait and animals that sit and wait for their prey such as web building or burrowing spiders This last feature was used in Moya Larafio et al 2013 using mini Akira the former R version of Weaver but is currently not in use Fungivores are included in the category does not hunt which forces animals to feed on basal resources only Omnivory sensu stricto i e animals 11 feeding on both other animals as well as fungi is not yet implemented Howeve
19. usage of each parameter Dimensions of the world world soil dimensions depth 10 length 50 width 50 cellSize 1 The world is defined by cells so far of arbitrary units cellSize refers to one arbitrary unit For now it should be kept at a fixed value of 1 depth length and width refers therefore to the 3D dimensions of the simulated ecosystem N Dimensions of the water pocket islands moisture patches type sphere radius 5 xPos 15 yPos 15 zPos 5 value 87 5 type sphere radius 5 xPos 35 yPos 15 zPos 5 value 87 5 type sphere radius 5 xPos 35 yPos 35 zPos 5 value 87 5 type sphere radius 5 xPos 15 yPos 35 zPos 5 value 87 5 Water pockets have also 3D dimensions and may have different shapes and gradients There exist different ways of defining water in our scenery homogeneous spherical gaussian and random_gaussian All these patches can overlap and there is no virtual limit for the number of them Just take into account that if two or more patches overlap in space the maximum value for all of them will be assigned to the affected voxel Here is an explanation about each type of patch leo L Homogeneous All voxels in the soil will be assigned this same value Next
Download Pdf Manuals
Related Search