1 of 35

Resource Intensity for Menu Items:�How Much Land is Required to provide for each Dish?

Yoko Yamakata,

Eiji Yamasue,

Benjamin McLellan,

Kiyoharu Aizawa

CEA2018 @Stockholmsmässan

2 of 35

Definition of “Resource” in this presentation

  • Not

Computational resources

      • CPU, Memories, Networking, etc.

Human resources

Economical resources

​

  • But

Natural (mineral) resources.

​

3 of 35

Weight of our life in terms of mining activity

Tracing back in the lifecycle, we ultimately reach “mining activities”.

​

How large is the amount of the earth that we need in whole life?

Transportation

Services

Food

Crops

Lives

Vehicle

Incineration

Waste

Summation

Landfilling

metal

fuel

fuel

Fer.

chemical

Chems.

detoxification.

4 of 35

(Primary Resource) Mining activities

4

http://tabisite.com/gallery_af/morocco/krbg.shtml

1 t usable phosphorous ore (30% P2O5)

→ 8 t of unusable rocks (mining waste)

　　 = Hidden flow = Mining activity

Kobayashi, H. and R. Sago (2001) Japanese Journal of Farm Work Research, 36: 141–151

1 g of gold= 106 g of Mining activity

5 of 35

What the mining activities cause.

Nickel mine

Kagu, national (wingless) bird

Rare earth element

Panda

6 of 35

Food production and mining activities (ex. Beef)

6

Resource intensity (Total Material Requirement, TMR) �= 49.3 kg-TMR/kg-beef

250 kg of beef

550 kg

Maize 7 ton

Silage 11 ton

Soya bean 450 kg

Field (0.6 ha)

heating

fuel

fuel

15.8 kg-P

as fertilizer

P-ore

mining

mining

7 of 35

Specific TMR: Livestock and dairy farming

Specific TMR of meat and egg production (kg-TMR/kg)

Cow

Pig

Hen

Egg

Milk

Food itself

Energy

Material

Specific TMR: Total Material Requirement (TMR), mining activities, to obtain a unit of target material.

8 of 35

Sushi, Beef steak, and chicken curry (500 g)

g-TMR/serving

2900 g-TMR/serving

12800 g-TMR/serving

2100 g-TMR/serving

Beef

Fish

Chicken

Rice

Rice

Rice

9 of 35

Issue of representativeness

Minestrone

Google Image Search

10 of 35

Previous research: statistic analysis of recipe

Findings and Issues

​

First attempt in the world.

​

Statistic data seem to follow normal distribution.

​

Washoku (Japanese dishes) oriented.

​

Other nations’ dishes should be analyzed.

Our previous research

Cookpad.com (Japan)

User participation-type website for �cooking recipes in Japan

1.7 million recipes

Recipe data are provided to researchers.

11 of 35

Purpose

  • Even using TMR, however, the evaluation of resource intensity involves the so-called “representativeness” issue.
  • Most families have their own recipes, meaning that a certain meal may have many variants. This means that statistical analysis is indispensable for the treatment of recipes – particularly from large datasets.
  • Thus, the purpose of the present study is to combine the concept of resource intensity through TMR and the statistically processed recipe data obtained on the Web in terms of global perspectives, and to extract their characteristics.

​

​

12 of 35

Target: Food.com (currently Genius Kitchen)

URL

  • http://www.geniuskitchen.com/
  • No data officially provided.

​

Data Collection

  • Data collection by crawling method using a robot.

​

Period

  • From April to June, 2017.

​

Data obtained

  • 500,231recipes.

13 of 35

Data Processing (information processing)

・

・

・

・

・

・

・

・

200 g

500,231 recipes.

19,305 are filtered for analysis

14 of 35

Data Processing (quantifying mining activity)

800 g-TMR

・

・

・

・

・

・

・

・

Our own database�

~800 materials�

http://www.ritsumei.ac.jp/~yamasue/tmr/database_en.html

Database of specific TMR

15 of 35

Specific TMR determination

800 g-TMR

200 g

4 g-TMR/g

Specific TMR: Total Material Requirement (TMR), mining activities, to obtain a unit of target material.

16 of 35

Results: Weight histogram

Weight/serving (g)

Frequency (number of data)

Average = 639 g

Mode = 100~125 g

17 of 35

Results: TMR histogram

TMR/serving (g-TMR)

Frequency (number of data)

Average

= 1572 g-TMR

Mode

= 500~600 g-TMR

18 of 35

Results: specific TMR histogram

Specific TMR (g-TMR/g)

Frequency (number of data)

Average

= 4.5 g-TMR/g

Mode = 3 ~ 4 g-TMR/g

19 of 35

Results: Specific TMR Data obtained

Specific TMR (g-TMR/g)

ex. Hot Buttered Rum (Aluminum)

ex. Chocolate Spritz Cookies (Stainless steel)

ex. Solo Baked Chicken Breast (Ordinary steel)

20 of 35

Discussion: weight/serving and S. TMR

Weigh/serving (g)

Specific TMR

Weight/serving

Specific TMR (g-TMR/g)

Specific TMR for butter (54.2 g-TMR/g) and for beef (48 g-TMR/g)

21 of 35

Future perspectives: global trend of food consumption

Annual TMR (Mt-TMR/year)

EU

India

China

USA

Japan

Africa

ASEAN

S. America

18 Gt-TMR (2015)

327 t / 960E (Komatsu)

22 of 35

How large is 18 Gt-TMR/year ?

18 Gt-TMR = 60,000,000 cars

= 1.8 million km in length

327 t / 960E (Komatsu)

・・・・

23 of 35

How long is 1.8 million kilometers?

Going around the earth 45 times.

2 round-trip from the earth to the moon.

24 of 35

Summary

  • By crawling food.com (currently Genius Kitchen, http://www.geniuskitchen.com/), 500,231recipes could be extracted and 19,305 of them were subjected to be associated with mining activities.
  • Obtained median weight per serving for various dishes are acceptable.
  • Weight and TMR per serving and the specific TMR for many dishes seem to follow normal distribution.
  • Specific TMR for dishes ranges from about 1~ 52 g-TMR/g.
  • Higher specific TMR of butter and beef strongly effect on higher mining activities (resource intensity) behind dishes.

​

25 of 35

Thank you for your attention!!

yamasue@fc.ritsumei.ac.jp

Ritsumeikan University

Eiji YAMASUE

Acknowledgement

This research is partly supported by "Environment Research and Technology Development Fund of the Ministry of the Environment, Japan”, (3K163001 and S-16) and Kakenhi (16KT0030, 18K11425) and the research grants by The SKYLARK Food Science Institute in 2016 and 2017, and cookpad.com.

Let’s eat Sushi!

Let’s consider not only direct weight but also weight of primary resources behind our life!

26 of 35

27 of 35

28 of 35

TMR for meals (g-TMR/serving)

TMR (g/serving)

Error bar: 99% confidence level

TMR from cooking energy

Staple

Main

Soup

Side

TMR of material

29 of 35

Specific TMR of meals (kg-TMR/kg)

TMR (kg/kg)

Beef Bowl

Roast beef

Specific TMR of Aluminum

Steel

Specific TMR of Stainless steel

Beef stroganoff

30 of 35

Preliminary results

31 of 35

Data processing

31

※KAKASIを使用、同音食材（「牡蠣・柿」など）や単音食材（「酢」など）は後の処理で区別可能なように別途変換

From weight to TMR

​

​

卵1g：6.3TMR

・・・

料理・材料名�Kanji ->Hiragana

​

​

大蒜：にんにく

・・・

500,231�recipe data

For two dishes

鶏むね肉：１枚

生姜：１かけ

からあげ

2にんまえ

とりむねにく: 1 piece�しょうが：1 piece

おいしいからあげ

2にんまえ

とりにく：280g

しょうが：20g

Unit conversion

​

​

1 Egg：60g

・・・

おいしいからあげ

2にんまえ

TMR：2115.2 g-TMR

32 of 35

Future perspectives

  • Research grants
    • Kakenhi (National Research Grant)
      • Existing: Scientific research (B)
        • Resource intensity regarding food production
        • 2nd year ( 3 years in toal)
      • New proposal: Scientific research (A)
        • About 500 thousand AUD/4-5 years
        • TMR (mining activities) for human’s whole life
    • CREST or PREST

​

33 of 35

Australia, Japan, China (direct weight)

Food supply (kg/capita/year)

Milk

Fruits

Cereals

Alcoholic beverages

Roots

Sweeteners

Vegetables

Bovine

Fish

Australia

Japan

China

Vegetables

Cereals

Cereals

Roots

Fruits

Fruits

Milk

Vegetables

34 of 35

Australia, Japan, China (TMR)

Food supply (kg/capita/year)

Milk

Fruits

Alcoholic beverages

Sweeteners

Bovine meat

Australia

Japan

China

Stimulants

Pig meat

Vegetables

Stimulants

Bovine meat

Pig meat

Cereals

Due to higher specific TMR of Beef production

35 of 35

Specific TMR database

Goods (~800, materials, energy source, foods, organic and inorganic compounds, solutions, etc)

Specific TMR (kg-TMR/kg)

Scope of this data is mainly production activities in Japan.