Vertical Farming

[12] “Fifteen billion trees are cut down every year” - (2015) Mapping tree density at a

global scale | Nature (as quoted in Deforestation and Forest Loss | OurWorldinData)

“The UN FAO estimates that 10 million hectares of forest are cut down each year” - (2020) Global Forest Resources Assessment (as quoted in Deforestation and Forest Loss | OurWorldinData):

- Rates of net forest loss are often cited at around 5 million hectares per year: “Between 2010 and 2020, the net loss in forests globally was 4.7 million hectares per year. However, deforestation rates were much higher.”

“At least three-quarters of this is driven by agriculture - clearing forests to grow crops, raise livestock, and produce products such as paper” - Drivers of Deforestation | OurWorldinData:

- Several estimates can be found in the literature on what share of deforestation is driven by agriculture. Differences in these estimates are often explained by differing considerations of deforestation versus forest degradation. Agriculture accounts for 70-80% of tropical deforestation - the permanent conversion of forested land to another land use. This share is slightly lower when degradation is included i.e. temporary loss of forest or permanent loss of forest quality. Noriko Hosonuma et al. (2012) estimate that 73% of tropical deforestation is driven by agriculture.

See here and here for more information

(Here farming is meant as terrestrial agriculture for food production, thus excluding fisheries, aquaculture and non-food products)

states food production and consumption accounts for 26% of global GHG emissions, detailed further in the following breakdown: 24% land use (8% for human food, 16% for livestock); 27% crop production; 30% livestock & fish farms (+1% wild fisheries); 18% supply chain. This equates to around 20% of global GHG emissions when excluding post-farming supply chain and fisheries-related emissions.

From the original abstract: “In 2015, food-system emissions amounted to 18 Gt CO₂ equivalent per year globally, representing 34% of total GHG emissions. The largest contribution came from agriculture and land use/land-use change activities (71%), with the remaining from supply chain activities: retail, transport, consumption, fuel production, waste management, industrial processes and packaging”.

OurWorldinData offers explanation of differences between estimates from [16] & [17]: [17] allocates all global deforestation (and attributed CO₂) to agriculture, while acknowledging only 80% of deforestation is actually agriculturally driven (the rest being urban expansion). This brings their estimate down closer to 23.5% of GHG emissions before supply chain, when excluding urban expansion but including fisheries-related emissions and some non-food related products. [16] provides a separate estimate for non-food agricultural products, which together would bring their total up from 26 to 33%.

In 2017, the percentage contribution of agriculture to world CO₂eq emissions from all human activities was 20%, equating to 11.1 Gt CO₂eq yr^-1. This included 6.1 Gt CO₂eq yr^-1 (11%) from crop and livestock activities from farming and 5.0 Gt CO₂eq yr^-1 (9%) from agricultural land use, largely due to deforestation and peatland degradation.

Agri-food systems accounted for 29.7% of the 53.5 Gt CO₂eq released in 2022. Farming and related land use change accounted for 48% and 19% of this faction respectively, amounting to 19.9% of global GHG emissions.

2% of all land on Earth - specifically cropland, out of all agricultural land (37%) (2021).

presents 14% of all ice-free land on Earth (2014), while [27] presents 12% of all ice-free land on Earth (2008) - possibly out of date.

cites from 2007, 12% of all lands excluding Greenland and Antarctica.

It is worth acknowledging that several vertical farm enterprises and business ventures have collapsed in recent years, and how those referenced here will fare in the coming years is unknown. Commentary has pinpointed prevailing reasons for these failures, highlighting attempts to scale too rapidly or to adopt and showcase advanced, high-cost technologies before establishing core operational stability and economic viability, prioritizing innovation over basic agricultural and financial fundamentals.

Common challenges included high energy demands, narrow crop selection, overestimated yields, and underdeveloped business strategies. Many of these farms focused on technological sophistication at the expense of simplicity, efficiency, and market fit.

However, it is stressed that these failures should not be interpreted as evidence against the viability of vertical farming as a whole. Rather, they represent early-stage setbacks in an emerging field. Vertical farming remains a promising tool for enhancing climate resilience and food security, particularly in urban environments and regions facing land or water constraints. The lessons from early missteps provide an opportunity to learn and valuable insights to guide more pragmatic and sustainable approaches in future development.

See Further Videos ‘Lessons to be learned’ for example commentaries.

Recent global studies estimate that food-miles account for about 19% of total food-system emissions, or roughly 3 Gt CO₂eq annually ^[79]. Transport emissions are especially high for perishable, often-imported goods like fruits and vegetables, where freight emissions can even exceed production emissions.

While earlier U.S.-focused studies ^[81] found that transportation contributes only 11% of food-related emissions (with most emissions from production), newer global assessments indicate that international trade plays a larger role than previously believed, especially in affluent, import-heavy nations.

However, a complete shift to fully local food systems isn't feasible everywhere: only 11-28% of the global population can fulfill demand for specific crops by sourcing within a 100 km radius, while over a quarter rely on food that must be transported more than 1,000 km ^[80]. Still, this equates to a substantial volume of avoidable emissions. While global trade remains essential for food security, particularly in less self-sufficient regions, increasing the share of locally produced food - growing closer to where it's consumed, especially in wealthy countries - could avoid hundreds of megatons of CO₂ and make food systems more resilient.

Estimated density and frequency of lettuce harvest under traditional field cultivation is based on agronomic guidelines (average field spacing recommendations) for common lettuce types (leaf, butterhead and romaine):

Row spacing typically ranges from 40-80 inches per bed, with 2-6 seedlines, yielding average row spacings of approximately 37-50 cm.

In-row spacing varies by variety, ranging from 23-30cm for full-sized heads to 13-15cm for romaine hearts. These configurations yield densities of approximately 6.9-9.7 plants/m² for standard lettuce types and 14-19.6 plants/m² for romaine hearts.

Lettuce is typically harvested once every 3.5-4 months, depending on climate and variety, and does not allow for stacking or frequent replanting. However, harvests aren't consistent throughout the year - sometimes they only take ~2-2.5 months growing time in warm conditions (65-80 days when planted in summer, up to 130 days when planted in winter/early spring).

We opted for ~9 plants per square meter per 4 months as a conservative, mid-range benchmark for traditional farming systems. By contrast, vertical farming systems can achieve substantially higher productivity through vertical stacking (e.g. 10 or more tiers) and faster crop turnover (e.g. 18-30-day cycles), enabling yields of 600+ plants/m² over a comparable 4-month period.

This scale of yield is generally supported by other figures quoted across scientific literature ^[82-85] (additional industry claims may be found online). Jin et al. (2022)^A reported that each layer of vertically farmed lettuce could produce annual yields of 700kg per m² under optimal light regimes (several hundred heads per m² over a few months, depending on crop cycle and head size). Meanwhile, a new approach described by Ptak, Wasieńko & Makula (2024)^B achieved a reported yield equivalent to 597 heads per m³ in a rotating vertical farm system. Beyond efficient use of space, these examples highlight opportunities of vertical farms for further improvements in yield through greater light use efficiency. Continuous light was shown to drive near-constant photosynthesis, balanced against increased energy demand and risks of photoinhibition, heat stress and disruption of circadian processes in plants^A. In contrast, multi-chamber light/dark regimes allow each crop layer to receive sufficient daily light while conserving resources and reducing operational costs and stress on plants and equipment. Such approaches show promise in reducing energy demand while retaining the yield advantages of vertical farming.

While ‘agricultural development has improved human lives for the better’ - enabling population growth, food surpluses, and technological advancement - it’s worth acknowledging the broader health and social consequences of sedentary civilization that followed. These include reduced dietary diversity compared to forager diets, increased transmission of infectious diseases due to higher population density and proximity to domesticated animals, social stratification, and heightened risk of organized conflict. These outcomes are not inherent to agriculture itself, but rather to the societal structures that emerged alongside permanent settlement.