Data-based analysis of climate decisions

 
How do we – individually and as a society – respond to a changing climatic environment and extreme weather events?

Our research group works to better understand the complex, dynamic, and multi-dimensional mechanisms through which climatic conditions affect human well-being and through which climate-related decisions are made, both individually and on a societal level. We analyse these systems through a variety of data-analytic methods, ranging from statistics, econometrics and machine learning to numerical modelling techniques. Using both historic data and projected data about the climate, socio-economic conditions, and development pathways, we are able to provide insight into the motivators of past climate-related phenomena as well as predict future climate-related decisions and societal developments.  

Previously, we have studied how temperature shocks affect economic productivity and electricity consumption, how extreme weather events can cause ripple effects in the global trade and food supply networks, and the factors motivating solar panel usage. 

For more details, see below.

Working Group Leader

Leonie Wenz


Cooperations

MCC, UC Berkeley, Columbia University, University of Sydney, Leiden University, University of Queensland

Data and Software

DOSE (MCC-PIK Database Of Subnational Economic output)

Research foci

Using social media data to study climate-related behaviour

Understanding the effects of the changing climate on human behaviour is essential for developing informed and effective adaptation and mitigation strategies. In our research, we employ methods from machine learning and econometrics to analyse the unprompted beliefs and feelings that millions of users express on social media platforms such as Twitter in order to understand behavioural changes in relation to environmental influences.

annika-image

One focus of our work is the analysis of conflicts and aggression in digital spaces. As the Covid-19 pandemic spread, we identified a surge of Sinophobia on social media highlighting how crisis situations can aggravate racism and discrimination online. Currently, we use Twitter data to investigate the relationship between climate variables and digital interpersonal conflicts in collaboration with the Max Plank Institute for Human Development.


Artificial intelligence & geospatial analysis for sustainable decisions

In this area, our group focuses on the use of artificial intelligence methodologies and use of geospatial data to develop insight into climate-related decisions on a societal level. So far, our work has used tree-based algorithms to better understand the factors that determine the probability of solar panel installation, as well as population relocation in the aftermath of disasters. By means of satellite imagery and population data, we have quantified the number of people worldwide without access to infrastructure via roads (Sustainable Development Goal 9.1) and assessed the trade-off between closing these access gaps and achieving ambitious climate change mitigation targets (SDG-13). We are also applying machine learning methodologies to better understand the predictors of global seasonal temperature variability. Towards these ends, we work extensively with large, geospatial datasets and time series data.

kelsey-image


The economic costs of climate change

max-image

Assessments of the economic costs of climate change are a vital tool for guiding climate policy and achieving mitigation, but such assessments lack a comprehensive empirical basis. By employing state-of-the-art statistical methods (e.g. from econometrics, pattern recognition, detection/attribution) to historical data, we aim to uncover new links between climate and society with which we can assess the future costs of climate change. 

Furthermore, in collaboration with the Mercator Research Institute on Global Commons and Climate Change we are continuing to develop DOSE, an open-access Data-base of Sub-national Economic output. With this strong empirical foundation, we will continue to identify and quantify detailed climate impacts and to translate these into policy relevant insights. See below for a list of our related work:

See below for a list of our related work: 

  • M. Kotz, L. Wenz, A. Levermann, Footprint of greenhouse forcing in daily temperature variability
    Proceedings of the National Academy of Sciences 118 (2021). https://doi.org/10.1073/pnas.2103294118
  • M. Kotz, L. Wenz, A. Stechemesser, M. Kalkuhl, A. Levermann. Day-to-day temperature variability reduces economic growth. Nature Climate Change (2021). https://doi.org/10.1038/s41558-020-00985-5/
  • M. Kalkuhl & L. Wenz. The impact of climate conditions on economic production. Evidence from a global panel of regions. Journal of Environmental Economics and Management 102360 (2020). https://doi.org/10.1016/j.jeem.2020.102360
  • F. Ueckerdt, K. Frieler, S. Lange, L. Wenz, G. Luderer, A. Levermann. The economically optimal warming limit of the planet. Earth System Dynamics 10 (2019). https://doi.org/10.5194/esd-10-741-2019
  • L. Wenz, A. Levermann, M. Auffhammer. North-South polarization of European electricity consumption under future warming. Proceedings of the National Academy of Sciences 114 (2017). DOI: 10.1073/pnas.1704339114.
  • Leonie Wenz, Matthias Kalkuhl, & Maximilian Kotz. (2021). DOSE - The MCC-PIK Database Of Subnational Economic output (Version 1) [Data set]. Zenodo. [doi.org] http://doi.org/10.5281/zenodo.4681306


The international trade network & climate change

timothe-image

International supply chains interconnect suppliers and consumers throughout the world economy. We use Multi Regional Input Output (MRIO) tables, large data sets capturing interdependencies in the global trade network, to explore these interconnections and their role in propagating climate damages. We develop algorithmic methods to advance MRIO techniques on spatial (Wenz et al., 2015) and prospective dimensions (Beaufils & Wenz, 2021). We work in close cooperation with the working group Numerical analysis of global economic impacts to investigate how the structure of the international trade network shapes our resilience toward extreme weather events (Wenz & Levermann 2016, Bren D'Amour et al., 2016).

  • T. Beaufils & L. Wenz, A scenario-based method for projecting multi-regional input-output tables
    Economic Systems Research (2021). https://doi.org/10.1080/09535314.2021.1952404
  • C. Otto, S. N. Willner, L. Wenz , K. Frieler, A. Levermann. Modeling loss-propagation in the global supply network: The dynamic agent-based model acclimate. Journal of Economic Dynamics and Control 83 (2017). DOI: 10.1016/j.jedc.2017.08.001.
  • L. Wenz & A. Levermann. Enhanced economic connectivity to foster heat stress-related losses. Science Advances 2 (2016). DOI: 10.1126/sciadv.1501026.
  • C. Bren d'Amour, L. Wenz, M. Kalkuhl, J.C. Steckel, F. Creutzig. Teleconnected food supply shocks. Environmental Research Letters 11 (2016). DOI: 10.1088/1748-9326/11/3/035007.
  • L. Wenz, S.N. Wilner, A. Radebach, R. Bierkandt, J.C. Steckel, A. Levermann, Regional and sectoral disaggregation of multi-regional input-output tables - a flexible algorithm, Economic Systems Research 27 (2015), DOI: 10.1080/09535314.2014.987731

Corona-virus related work

The working group also investigates the question of whether the COVID-19 pandemic and possible political, individual and economic responses to it will render us more or less vulnerable to future weather extremes and how we can steer that development in a favourable way such that our societies’ resilience to extreme events – of any kind – is enhanced.