The research, published in Cleaner Engineering and Technology, arrives at a moment when agriculture’s climate accounting is under intensifying scrutiny. Farming contributes an estimated 10 to 15 percent of global greenhouse gas emissions, and China, the world’s largest carbon emitter, has pledged to peak its emissions by 2030 and achieve carbon neutrality by 2060. Yet most carbon footprint studies of fruit production have examined only isolated stages, such as planting or transport, leaving the cumulative impact of processing and packaging largely unmeasured. The team behind the new work argues that this fragmentation has prevented the emergence of genuinely comprehensive low-carbon strategies, and they set out to close the gap by treating the whole industrial chain, from orchard to processed product, as a single accounting and optimization problem.

Qiongzhong County offered an ideal case. Its tropical monsoon climate, with an average annual temperature of 22.8 degrees Celsius and rainfall of 2,444 millimeters, suits citrus cultivation well, and the Qiongzhong-Lvcheng orange carries significant economic weight in the region. Because the local industry is still young, it currently relies on just a handful of product pathways: bulk fruit sold as harvested, standardized graded fruit, and small volumes of green orange wine and green orange vinegar brewed from secondary fruit that cannot be sold whole. The researchers extended this real-world chain hypothetically to include concentrated juice and not-from-concentrate juice, drawing processing data from comparable citrus operations in the literature, and defined six functional units, each based on one tonne of harvested oranges flowing to a different final product.

The carbon accounting revealed stark differences among the pathways. Planting one tonne of oranges for bulk sale generates 165.1 kilograms of CO2-equivalent, while graded fruit comes in at 180.1 kilograms. Diverting the same tonne to concentrated juice raises the footprint to 282.1 kilograms, to NFC juice 188.7 kilograms, to green orange wine 385.1 kilograms, and to green orange vinegar a striking 796.8 kilograms. The economics tell a different story. Graded fruit earns a profit of about 18,060 yuan per tonne of input, green orange wine roughly 20,270 yuan, and green orange vinegar the highest at about 22,790 yuan. Concentrated juice is the outlier in every sense: it carries a high carbon footprint and a negative profit of roughly 2,660 yuan per tonne of oranges, because the energy-hungry concentration and refrigeration stages consume far more value than the tiny 0.063-tonne juice yield recovers.

Dissecting the emissions further, the study found that fertilizers and pesticides dominate the planting stage, contributing 53 percent and 43 percent of its footprint respectively. Organic fertilizer alone accounts for 48 percent, not because it is carbon-intensive per kilogram, but because nearly two tonnes of it are applied per tonne of fruit. Insecticides, with an emission factor of 16.61 kilograms of CO2-equivalent per kilogram, contribute 38 percent, a burden driven by the orange’s acute vulnerability to Huanglong disease and red spider mites, which force heavy pesticide applications. For the processed products, packaging emerges as a surprisingly powerful lever: glass bottles account for 60.8 percent of green orange vinegar’s total footprint and 18.9 percent of green orange wine’s, while concentration, cooling, and refrigeration dominate the concentrated juice pathway.

With these per-product profiles established, the team turned to optimization. They employed NSGA-II, a widely used evolutionary algorithm, to search for the allocation of harvested oranges across the six products that best balances three competing objectives: maximizing economic profit, minimizing carbon footprint, and limiting industrial instability, measured through the coefficient of variation of the product mix. Two constraints grounded the simulation in reality. At least 20 percent of the harvest must flow to deep processing, reflecting the share of secondary fruit that cannot be sold whole, and no product’s share may deviate from the current structure by more than 30 percent, acknowledging that consumer demand and processing capacity cannot shift overnight. The algorithm, run with a population of 300 over 400 generations, produced a Pareto front of non-dominated solutions from which four representative scenarios were distilled.

The scenario results are where the study’s headline finding emerges. Under the economic development scenario, profit rises by 40.7 percent as oranges shift toward graded fruit, green orange wine, and green orange vinegar, but total emissions climb 63.7 percent. The low-carbon development scenario, which channels roughly two-thirds of the harvest into bulk and graded fruit and the remainder into green orange wine and NFC juice, raises profit by 10.0 percent while holding the carbon increase to just 7.9 percent. The stabilized development scenario spreads the harvest almost evenly across all products, cutting the industrial instability index from 1.42 to 0.10, though it sacrifices 5.1 percent of profit and raises emissions by 63.9 percent. The comprehensive development scenario, concentrating on the three best-performing products, lifts profit 31.8 percent with a 19.4 percent carbon increase, but leaves instability relatively high at 1.18.

For the researchers, the low-carbon pathway stands out as the most immediately actionable. It demonstrates that a modest, targeted expansion of deep processing, one that fully utilizes the roughly 20 percent of fruit currently sold off cheaply as secondary produce, can simultaneously raise incomes, diversify the product portfolio, and keep emissions nearly flat. The current industry, by contrast, is highly exposed: with 49.5 percent of oranges sold in bulk and 44.5 percent as graded fruit, and only 6 percent processed, the sector’s total profit of 128 million yuan rests on a fragile, single-structure foundation that leaves it vulnerable to fresh-fruit market swings. Sensitivity and Monte Carlo analyses, involving 10,000 simulated iterations, confirmed that the ranking of products by carbon footprint remains robust even under wide uncertainty in the input data.

The implications extend well beyond one county’s citrus groves. The planting stage accounts for more than 85 percent of the footprint of primary products, a pattern consistent with citrus studies in Spain and apple studies in China, and the dominance of fertilizers and pesticides echoes findings in other intensively managed crops. The team’s recommendations, improving organic fertilizer efficiency, strengthening physical pest control, recycling CO2 from fermentation and distillation, shifting processing energy to renewables, and adopting lighter, recyclable packaging, are transferable to most fruit and vegetable value chains. For policymakers in Qiongzhong, the study points toward subsidies that make low-carbon production visibly valuable, standardized cultivation practices, and a deliberately diversified processing sector. For the broader challenge of reconciling agricultural livelihoods with climate targets, it offers a concrete demonstration that the trade-off between profit and emissions, while real, is far more manageable than a single-minded focus on either goal would suggest.

Subject of Research: Whole-chain carbon footprint accounting and multi-objective optimization of the Qiongzhong-Lvcheng orange industrial chain in Hainan, China

Article Title: Value-chain optimization enhances economic resilience and profits with moderate carbon footprint increase: Qiongzhong-Lvcheng orange case

Article References: Gong, C., Yu, Q., He, K., Dong, X., Guo, Q., Cai, Y., He, L., Varbanov, P. S., & Wang, X.-C. (2026). Value-chain optimization enhances economic resilience and profits with moderate carbon footprint increase: Qiongzhong-Lvcheng orange case. Cleaner Engineering and Technology, 34, Article 101318. https://doi.org/10.1016/j.clet.2026.101318

Image Credits: AI Generated

DOI: 10.1016/j.clet.2026.101318

Keywords: carbon footprint, life cycle assessment, value chain optimization, citrus production, NSGA-II, agricultural emissions, economic resilience, green orange wine, Qiongzhong-Lvcheng orange, sustainable agriculture, multi-objective optimization, Hainan