A simple population model - UMDseminar/data/y10fall/...− VENSIM – simulation software − used...
Transcript of A simple population model - UMDseminar/data/y10fall/...− VENSIM – simulation software − used...
A simple population model
Harald Rybka, Jason Wong, Jorge Rivas RISE (Research Internship of Science and Engineering) Intern
Johannes Gutenberg University, Mainz and University of Maryland AOSC, 2 September 2010
1. Introduction to VENSIM 2. Limits to Growth – World3 model
3. Basic New Model / Structure 4. Improved Basic Model
results – global
5. Improvements and Problems
OUTLINE
9/2/2010 2
− VENSIM – simulation software − used for analyzing and simulating dynamic
feedback modeling − identifying leverage points and causal loops − similar to “STELLA”
Results: − use “SyntheSim” mode − different effects can be simulated with
sensitivity simulations
INTRODUCTION TO VENSIM
9/2/2010 3
INTRODUCTION TO VENSIM ─ AN EXAMPLE
9/2/2010 4
source sink
stock flow flow
constant / auxiliary variable “SyntheSim”
sensi[vity run average life[me: 5 – 25 years average life[me: 20 years
average life[me: 6 years
─ created by Donella Meadows, Jørgen Randers, Dennis Meadows (authors of the book: Limits to Growth, 1972, 2004)
─ computer simulation of interactions between: ─ the population system ─ the industrial system ─ the food system (agriculture and food production) ─ the non-renewable resources system and ─ the pollution system
─ ten different scenarios present results from 1900 till 2100
─ results varies from stabilization to collapse in the future ─ depending on several variables (e.g. nonrenewable resources)
LIMITS TO GROWTH – WORLD3 MODEL
9/2/2010 5
WORLD3 MODEL – STANDARD RUN: agrees with obs. 1970-‐2000 (Turner, 09)
9/2/2010 6
popula[on food industrial output persistent pollu[on nonrenewable resources
business-as-usual
population collapses because of - decrease of food and health services - increase costs of nonrenewable resources
WORLD3 MODEL – SCENARIO 2
9/2/2010 7
doubled initial nonrenewable resources
popula[on food industrial output persistent pollu[on nonrenewable resources
nevertheless: population collapses about 20 years later
WORLD3 MODEL – OPTIMISTIC (STABILIZED) SCENARIO
9/2/2010 8
popula[on food industrial output persistent pollu[on nonrenewable resources
Compared to Standard Run: ─ doubled nonrenewable resources ─ increase of land yield ─ improved pollution control ─ reduced land erosion
stabilized population with the right policies
WORLD3 – TYPES OF POSSIBLE SOLUTIONS
9/2/2010 9
STABLE – no overshoot
UNSTABLE - overshoot
carrying capacity
population
continuous growth sigmoid approach to equilibrium
overshoot & oscillation overshoot & collapse
BASIC NEW MODEL
9/2/2010 10
simulate World population via fertility and life expectancy
assumption: carrying capacity of the world is unlimited!!!
BASIC NEW MODEL -‐ RESULT
9/2/2010 11
year 2010 ~ population of 7 billion
next we will add a capacity limit
World Population
United Nations demographic data Basic New Model
Now food production is limited!
BASIC NEW MODEL WITH LIMITS
9/2/2010 12
limiting factors
9/2/2010 13
BASIC NEW MODEL WITH LIMITS -‐ RESULTS
population approaches 8.9 billion people
Setup: food production = food consumption
for a population of 8 billion
consumption per capita is a constant
population
notes:
overproduction has no positive effect on life expectancy
life expectancy collapses after year 2030 because of malnutrition
life expectancy in years
before agriculture was proportional to the population
now we add instead of agriculture a new variable: pollution
BASIC NEW MODEL – NEW VARIABLE: POLLUTION
9/2/2010 14
Basic Model popula[on system
collapse or oscillation depends on pollution impact on life expectancy
COLLAPSE OR OSCILLATION?
9/2/2010 15
POPULATION
Improvements: ─ demographic system divided in age-cohorts
→ more details about demographic characteristics
─ more complex agriculture system → based on FAO (Food and Agriculture Organization) data
─ many variables show a sigmoid trend → production of meat and cereals, increasing arable land, land yield
─ calculation of life expectancy restricted to food consumption → cutting-off connection between life expectancy and population
─ different scenarios can be simulated
BASIC MODEL WITH AGE-‐COHORTS AND FAO AGRICULTURE DATA
9/2/2010 16
BASIC MODEL – DEMOGRAPHIC SYSTEM
9/2/2010 17
no direct connection between life expectancy and population anymore!
BASIC MODEL – AGRICULTURE SYSTEM
9/2/2010 18
→ independent parameters
assumptions:
- total food production increases ~ constantly
- max. life expectancy 90 years
- fertility rate does not increase in the future
nevertheless: → population decreases WHY?
BASIC MODEL – OPTIMISTIC SCENARIO
9/2/2010 19
population
total food production
9/2/2010 20
BASIC MODEL – OPTIMISTIC SCENARIO
reason for decreasing population:
→ age distribution change
higher mortality rate for older people
→ increasing overall mortality rate
age 0 to 14 age 15 to 44 age 45 to 64 age 65 plus
fer[lity rate mortality rate
DEPENDENCY RATIO -‐ dependent/working popula[on
9/2/2010 21
popula[on dependency ra[o
dependency ratio = (population 0 to 14 + population 65 plus) / population 15 to 64
small variation of dependency ratio (0.65 – 0.75)
BASIC MODEL – PESSIMISTIC SCENARIO
9/2/2010 22
population
cereals yield
total food production
cropland
“Erosion, Urbanization” “Green
Revolution”
“Running out of oil”
Required changes to simulate regions:
- change variables into arrays → no need to make several models
- most regions are not self-sufficient → need to include imports and exports
- add migration rate → dividing migrants in age-cohorts
FUTURE IMPROVEMENTS – INTRODUCE REGIONAL POPULATION MODELS
9/2/2010 23
SCHEMATIC OF THE REGIONAL MODELS
9/2/2010 24
9/2/2010 25
SCHEMATIC OF THE REGIONAL MODELS
fraction of migrants
─ migration should be related to economic and demographic status
(not much data for age-distribution of migrants)
─ imports and exports should be related to economics and food requirement
→ Most important: we have not yet included an economic subsystem
ISSUES WE HAVE TO DEAL WITH
9/2/2010 26
WE NEED TO ADD FEEDBACKS SUCH AS
9/2/2010 27
Popula[on
Urbaniza[on
Land Use Cropland
Food Produc[on
Popula[on
Food Requirement
Intensifica[on of Agricultural Produc[on
Soil Deple[on
Land Degrada[on
Food Produc[on
─ display real data for comparisons ─ add new systems like economy, emissions, nonrenewable
resources, etc.
9/2/2010 28
OTHER FUTURE DEVELOPMENTS
Popula[on System
Agriculture System
Economic System
Nonrenewable Resources
Emissions
─ add interactions between regions (A) and (B) such as
─ add indicators for population Quality of Life
─ Human Development Index ─ Ecological Footprint
OTHER FUTURE DEVELOPMENTS
9/2/2010 29
emissions region (A)
total emissions
cereal’s yield region
(B)
food produc[on region (B)
…”We still see our research as an effort to identify different possible futures. We are not trying to predict the future. We are sketching alternative scenarios for humanity as we move toward 2100.”…
Donnella Meadows, Limits to Growth – The 30-year Update
9/2/2010 30
…IN THE END