Problem Chosen
E
2023
MCM/ICM
Summary Sheet
Team Control Number
2314817
Turn off the lights, Turn on the stars!
“On the earth, even in the darkest night, the light never wholly abandons his rule.“
——— Jules Verne, Journey to the Center of the Earth.
Summary
With the rapid development of modern technology and excessive use of artificial lights, light
pollution has resulted in increasingly severe problems. Therefore, the proper design and implementation
of intervention strategies to address these problems deserve our attention. This report aims to build
a light pollution risk level evaluating model and provide credible suggestions on how to effectively
reduce the negative impacts of light pollution.
Three models are established: Model I: Ecologic-Social-Ecologic(ESE) Model; Model II: Light
Pollution Risk Assessment(LPRA) Model and Model III: Interconnected-ESE Model(I-ESE) Model.
For ESE Model (Model I), we first abstract the risk of light pollution into three subsystems :
Economic, Social and Ecological Subsystem and construct the ESE model. For economic subsystem,
we calculate the population growth by the logistics equation and measure the light pollution in economic
systems. For social subsystem, we take five types of land together with the brightness threshold into
account and obtain the proportion of light-pollution-affected land. For ecological subsystem, we set a
formula with the number of organisms together with light intensity value to calculate the number of
affected organisms.
For LPRA Model(Model II), we use Analytic Hierarchy Process(AHP) based on Google search
index to quantify the light pollution risk level(RL) of a certain region, then apply our model to 284
regions. After that, we use K-means algorithm to create a metric system for the risk levels and
constructed four ranks: Slight(0.0443 ∼ 0.1678), Moderate(0.1705 ∼ 0.2542), Serious(0.2568 ∼
0.3698), and Severe(0.3976 ∼ 0.6186). To apply our model to specific locations, we selected Ordos as
protected land, Zhaotong as rural community, Qingyuan as suburban community, and Guangzhou
as the urban community and obtained their RL scores: 0.1332, 0.1778, 0.2574 and 0.5154, while
their risk levels are slight, moderate, serious and severe respectively.
For I-ESE Model(Model III), by establishing a link between every two subsystems, we constructed
the Interconnected (I)-ESE Model on the basis of the ESE Model, and obtain quantitative relationships
between ECL and SCL, ECL and ENL using nonlinear fitting. Based on I-ESE Model, we propose
three effective intervention strategies to address the light situation: I. Reduce light intensity, II. Reduce
the impact on society, and III. Delimit dark night protection areas and their actions correspondingly.
More specifically, we select Guangzhou and Zhaotong for further discussion and determine their
optimal strategy: strategy I and II for Guangzhou and strategy II and III for Zhaotong. Then we
measure the impacts of these strategies in general and particularly in Guangzhou and Zhaotong using
I-ESE Model and visualize the impact with GE cube. The risk level in Guangzhou decreased from
0.5153 to 0.0309 while Zhaotong decreased from 0.1678 to 0.0885 after 20 years of implementing the
optimal strategy.
Finally, the sensitivity and robustness of our model are tested. When we set the cost to be raw
data, increase or reduce by 50%, the overall trend of risk level has little difference as the curve remains
similar, verifying the sensitivity and robustness of our model.
Keywords: Light Pollution, ESE Model, LPRA Model, AHP, K-means algorithm, GE Cube
更多数模资讯和学习资料,请关注b站/公众号:数学建模BOOM
b站主页:https://space.bilibili.com/350975620