Several datasets are presented on this website. You can learn about each of them below.
Dataset: 2012 (original)
The information in this dataset is generated by the Food Flow Model, a data-driven methodology to estimate spatially explicit food flows. The Food Flow Model integrates machine learning, network properties, production and consumption statistics, mass balance constraints, and linear programming. A full description of the methods used to generate county-to-county food flows in the United States is provided in the accompanying academic journal article (Lin et al, 2019).
Lin, X., Ruess, P.J., Marston, L., and Konar, M. (2019). Food flows between counties in the United States. Environmental Research Letters, 14(8), 084011. doi: 10.1088/1748-9326/ab29ae
Datasets: 2007, 2012 (improved), 2017
Data on food flows between counties in the United States from 2007 to 2017 is estimated with an improved version of the original Food Flow Model developed by Lin et al (2019). Three key improvements are introduced into the original Food Flow Model:
Systematic handling of estimator selection
Smoothing distance data
A quantitative approach to select the core nodes
These improvements aim to achieve a more easily reproducible version of the original Food Flow Model with better handling of outliers and less dependence on the user-input. A full description of the improved methods used to generate county-to-county food flows in the United States from 2007 to 2017 is provided in the journal article (Karakoc et al, 2022).
Karakoc, D. B., Wang, J., and Konar, M. (2022). Food flows between counties in the United States from 2007 to 2017. Environmental Research Letters, 17(3), 034035. doi: 10.1088/1748-9326/ac5270
Dataset: 2017 (cold chain)
A model of cold chain food flows was developed to estimate refrigerated truck transport via roadways of meat and prepared foodstuffs. This model was developed for the year 2017 at the county spatial resolution in the United States. In addition to estimating the mass flux of food between counties, this study also quantifies the carbon footprint of these cold chain food miles. This work builds upon the original Food Flow Model developed by Lin et al (2019). This study differentiates from the original Food Flow Model in two key ways:
Roadway travel distance is used instead of haversine distance
The corresponding carbon dioxide emissions of cold chain food flows was estimated by taking into account spatial heterogeneity in ambient temperature
Wang, J., Karakoc, D., and Konar, M. (2022). The carbon footprint of cold chain food flows in the United States. Environmental Research: Infrastructure and Sustainability. doi: 10.1088/2634-4505/ac676d
The cold chain food flow study is restricted to roadway transport via truck (Wang et al 2022) while the Food Flow Model (Lin et al 2019; Karakoc et al 2022) estimates food flows agnostic to transport mode. Estimating food flows via roads leads to a more dense network than does estimating food flows that are agnostic to transport mode. The flow capacities for each county-to-county link are agnostic to the transportation-mode in the Food Flow Model. The Food Flow Model estimates and fills up high-capacity links between counties first, which can be thought to represent waterways and railways, in additional to the highways. This leads to lower density network estimates when compared with cold chain food flow networks through only highways.
The following is a list of Standard Classification of Transported Goods (SCTG) food categories included in this study.
SCTG Model
SCTG
Model
1
Animals and fish (live)
2
Cereal grains (includes seed)
3
Agricultural products (excludes animal feed, cereal grains, and forage products)
4
Animal feed, eggs, honey, and other products of animal origin
5
Meat, poultry, fish, seafood, and their preparations
6
Milled grain products and preparations, and bakery products
7
Other prepared foodstuffs, fats and oils
You can download all the data used in the datasets on this website:
