Davide Balula

Some farmed, others mined

Climate Impact Report

Davide Balula, Farmed Painting (Sagg creek, Thin Rock 2), 2020-2022
Courtesy the artist and galerie grank elbaz, Paris

galerie frank elbaz
66 rue de Turenne - 75003 Paris
March 26th - May 7th 2022

TABLE OF CONTENT

INTRODUCTION
GUIDELINES
ESTIMATED CARBON EMISSION
WASTE REPORT
SUPPORTING PEOPLE
ACTION
CREDITS

INTRODUCTION

We are facing a Climate Emergency. At the time of this writing, the Intergovernmental Panel on Climate Change (IPCC) just released their 6th assessment report. For the first time, the organization expressed serious concerns about social justice issues and the inevitable displacement due to Global Warming. Mentioned also in this report, is the fact that our window for effective action needed to limit the irreversible damage is closing sooner than anticipated. The IPCC report also mentions the need for a systemic change, and we believe change on such a level needs individual action to support and encourage this cultural shift. In short, the more we wait, the harder this transition will be. The later we start the harder it will be to adapt. 


Davide Balula has been encouraging our gallery to produce Climate Impact Reports for our shows and offered assistance in drafting our internal policy as a result.


By making this Climate Impact Report public, we hope to open conversations on this subject within the art sector. 


GUIDELINES

Since 2019, Davide Balula has been sharing the CO2e estimations on his labels and invoices (inviting clients to contribute to a strategic climate fund, land preservation etc).

His technical manuals acknowledge lifecycle information and his studio tries to hire local assistance as much as possible in order to reduce transportation and shipping. Sourcing materials locally is preferred when possible as the majority of objects used in his sculptures can usually be replaced/reused.
Read the artist’s statement: 
https://balula.org/c/o2e.html

galerie frank elbaz has been making efforts to limit waste (refusing and repurposing) and is reusing material (packing, etc) as much as possible. The gallery is currently working on their own guidelines.


ESTIMATED CARBON EQUIVALENT EMISSIONS

Shipping (2,540 kg CO2e)*:   
 Local L.I. - NY:
 800 kg CO2e
 
 Intl. NYC - Paris Air Freight:
 1,740 kg CO2e
 
 Travel (1, 712 kg CO2e):
  
 1 person NYC-GVA-NYC:
 1,710 kg CO2e
 
 1 train ride GVA-Paris-GVA:
 2 kg CO2e
 
 On site Energy (33.36 kg CO2e)**:
  
 Gallery electricity used for 1 month:
 33 kg CO2e
 
 Artworks using electricity for 1 month:
  0.36 kg CO2e
 
 Life-cycle Assessment (45 kg CO2e)***:
  
 Sculptures, electronic Labels:
 21 kg CO2e
 
 Sculptures, objects:
 0.05 kg CO2e
 
 Paintings:
 24 kg CO2e
 

* Local transportation of works benefited from consolidated shipping along with other items from the area, so it is hard to estimate. We will use arbitrary number of 1 kg CO2e (high estimate). Consolidation for air freight however is more common practice and many online calculators already factor this in.

** France used an average of 30g of co2e/kWh from March to April 2019
Source: https://www.rte-france.com/eco2mix/les-emissions-de-co2-par-kwh-produit-en-france#

*** Screens and objects have been reused. Paintings were produced using renewable living matter (fungi, lichen, sediments, and other organic and mineral matter).

Calculator used:
Gallery Climate Coalition (GCC):

https://galleryclimatecoalition.org/carbon-calculator/
Sustainability Tools in Cultural Heritage (STiTCH):
https://stich.culturalheritage.org/


Fig.1. Carbon equivalent emissions estimated for the duration of the exhibition



NOTES on carbon calculation related to Artificial Intelligence (NLP): 

Training an AI is hard to estimate. Once an A.I is pretrained, it will be used by numerous users and serve multiple requests with much less intensive processing power than originally needed. This is even more true with popular pretrained models. But we need to acknowledge that training an A.I. requires very high computing power, and therefore, a high amount of electricity. The energy needed for processing a request after the A.I. is trained also depends on the type of machine used, and emission depends on the location and source of energy distributed through the power grid (local/virtual machine, cloud/data center etc.).

Davide Balula and collaborators used the pre-trained transformer from OpenAi and applied further training via Google TPUs (more efficient and less energy intensive than GPUs).
OpenAi GTP2 is estimated to have produced around 652 kg CO2e for its training. The number is high, but again, we need to consider that this corresponds to the development phase. We refined the training of our model using virtual machines from Google - meaning our transformer greatly benefited from the original model already and much less training was needed. Google claims to run on renewable energy since 2017 and support environmental and social initiatives. (A “carbon neutral” tag on certain endeavors does not necessarily prevent them from being problematic.)

After an AI is trained, it is hard to determine the exact power used to generate a single output as it takes only a couple seconds to generate. The machines (laptop) used to display, select and archive these outputs locally should also be accounted for, but multitasking, emails, browsing etc makes the task complicated to isolate. We decided not to account for the training part as it has served millions of users already. The computing power needed to finalize the project is negligible and would anyways be too complicated to estimate. 

We also do not account for the Life-cycle Assessment of the servers/cloud/machines used for similar non exclusive usage. Such high level computing power requires regular updating of software, and hardware upgrade. 

We do acknowledge however the mining practices and land disturbances required to extract minerals and other matter from the soil. All the other numerous extracted materials (plastics, metals, coal, crude oil etc) needed to make calculations, cooling, data centers infrastructure, cables, etc. 

Hence the title Farmed/Mined. See https://anatomyof.ai/ for more information on this subject.

The LCD screens for the A.I. Generated Instructions works display the estimated carbon emissions emitted from the work in real time and accounts for the Life-cycle assessment of the objects and the energy used for each work.

WASTE REPORT

 Waste Category
 Examples of Items
Reuse:

to be reused for the same purpose as the original use

  • Plastic packaging (Plastic sheets were cleaned and
    kept with works to be reused.)

  • Protective foam

  • Gloves, cotton, washed

 
Repurpose:

to be kept, sold, or donated and used for a different purpose in the future

  • Crate will be reused with other works for local storage,
    and potentially other intl shipping.

  • Original packaging saved and modified for storage of electronic works.

  • LCD screens were reused from previous shows, and reprogrammed
    to create new work.

  • Gasoline will be donated after the show.

  • Sand will be returned to a park nearby.

 
Storage: 

items sent to storage, but without a clear plan for immediate reuse or repurpose

  • Empty crates will be reused with other works for local storage, 

  • and potentially other intl shipping.


Refuse:

item was not used at all and therefore potential waste was avoided

  • Various foams and plastics

 
Recycle:

items placed in the recycling bin

  • Usual plastics for packing, i.e. food.
    Note: acknowledged problematic recycling programs.

  • For more on plastic and recycling, see: here

 

Landfill

items sent to a landfill

 
  • Plastic tape

  • Left over cotton canvas, bamboo rayon left over from
    paintings were not composted. Amount to a small grocery bag.

  • Potato chips

  • Flower petals

 
 

Consumable:

“consumed” during the exhibition

  • Alcohol, food

 

SUPPORTING PEOPLE

Preparing this CIR allowed us to engage in conversations around climate justice, racism, harassment and other forms of inequality.

COLLECTIVE ACTION

This Climate Impact Report will be accessible online (artists website, gallery website, artistscommit.com ).
This CIR will be used to draft internal climate conscious internal guidelines. 

In the meantime, galleries are exploring incentives to create other CIRs and maintain this as common practice. Davide Balula offered to be available to help the preparation of other CIR for the gallery.

Collaborators and partners were asked about their policies and none had one. Shipping companies often feel defensive on the subject. We are hoping to see more transparency and open conversations on the subject.

ADDITIONAL NOTES/CLOSING THOUGHTS

We tried to engage vendors with climate conversations, and offered schedule flexibility for consolidation of items to transport.

One conversation in particular with a Logistic/shipper company is worth mentioning:

The company raised the argument for “clean fuel” of airfreight vs dirty fuel of maritime shipping, which notoriously uses a lower quality and more “polluting fuel”. 

However, freight capacities differ widely between the two. See Gallery Climate Coalition: Cadogan Tate case study Maritime vs Maritime. Other articles abound online about maritime vs freight, regulations etc each coming to different conclusions. “Is it be better to drive solo in a Tesla with its own complicated life-cycle, short lifespan, related to questionable mineral extraction practices, charging on fossil fuel powered grid etc. but offering a major shift of the industry towards renewable energy… or ride an old bus with dirtier fuel and directly responsible for increased acidification and accelerating extinction of certain species?” When you are in a situation where you can’t avoid either, every answer seems to be the wrong one.

CREDITS

This report was prepared by Clara Berthiaux (galerie frank elbaz) and Davide Balula (artist / artists commit) from a template available at Artistscommit.com

Link to Project: Some farmed, other mined

Link to Image: image