Globally, approximately 500 million tons of milled rice are produced annually, with 90% originating from Asian countries. Improving post-harvest management is crucial for enhancing grain quality, productivity, and minimizing losses, particularly for smallholder farmers in Asia. Rice post-production processes (harvesting to milling) incur losses estimated at 20–30%, but quantitative data remain limited. Post-harvest management practices vary widely across Asia; mechanized operations are common in some countries (Thailand, Vietnam) but uncommon in others (Myanmar). Low mechanization in Myanmar significantly hinders rice production efficiency. Shifting to mechanized systems requires substantial investment (machinery) and may increase GHGE due to machine production and fuel consumption. However, this needs to be balanced against potentially higher post-harvest losses under traditional systems. In Myanmar's Ayeyarwaddy delta, manual harvesting and delayed threshing cause significant quantitative (shattering, losses to birds/rodents) and qualitative (discoloration, mold) losses. Southeast Asia primarily uses two harvesting methods: manual cutting/mechanical threshing, and combine harvesters. Delays in harvesting due to labor shortages or wet rice plants lead to substantial losses. Sun drying, while common, leads to quality and quantity losses, unlike mechanical dryers (flatbed dryers). Poor storage also contributes significantly to losses. While economic and technical efficiency of post-harvest systems have been studied in various Asian countries, a synthesis comparing different practices to minimize losses, costs, and environmental impact is lacking. This study in Myanmar aims to: (i) assess energy and costs of traditional post-harvest systems, (ii) assess GHGE from mechanized options, and (iii) compare GHGE, energy, and costs under both scenarios.

Existing research has quantified the economic and technical efficiency of post-harvest rice systems in various countries including Korea, Bangladesh, Pakistan, Thailand, Myanmar, and Vietnam. Studies have also quantified energy efficiency and GHGE using life cycle assessment (LCA) in several countries like the Philippines, Japan, Iran, Italy, and Canada. Research has also focused on specific processes such as farm mechanization, drying, and mechanized rice straw collection. However, there's a lack of comprehensive studies directly comparing different post-harvest management practices to determine optimal options for minimizing losses, production costs, and environmental footprint.

This study was conducted in Tar Pat Village, Maubin, Myanmar (16.617° N, 95.680° E) during the wet season (WS) 2014 and dry seasons (DS) of 2015 and 2016. The Sin Thukha rice variety (135-day growing period) was used. Best practices were identified based on energy balance, cost-benefits, and GHGE per hectare. GHGE were estimated using attributional LCA, following ISO14044 standards. The system boundary encompassed all processes from pre-harvest (cultivation) to post-harvest (milling). Primary data were collected during harvest and post-harvest, while secondary data were used for pre-harvest processes. Conversion factors for energy and GHGE were obtained from the ECOINVENT 3 database (version 3.3). Table 1 outlines the research treatments and practices across seasons. WS2014 involved comparing two farmer practices (FPs) – one with a 1-week (FP1w) and one with a 4-week (FP4w) delay in threshing after manual cutting – and one improved practice (IPR) with immediate threshing, flatbed drying, and hermetic storage. DS2015 and DS2016 compared FP (manual cutting and sun-drying) and improved practice with combine harvester (IPRc), flatbed dryer, and hermetic storage. The experiment used a completely randomized design (CRD) with 5 replications per scenario. Equipment included locally fabricated and imported threshers, a combine harvester, a locally made flatbed dryer, and hermetic storage bags. Harvest and post-harvest losses were measured, including shattering, threshing losses, discoloration, milling recovery (MR), and head rice recovery (HRR). Equations (1–3) were used for calculations. Energy efficiency was quantified using net energy value (NEV) and net energy ratio (NER) (Equations 4–5). GHGE were accounted for the entire production chain (Equation 6), considering the rice product recovery ratio (Equation 7). Table 2 provides conversion factors for energy and GHGE. Cost-benefit analysis used net income value (NIV) and net income ratio (NIR) (Equations 8–9). Table 3 details component costs. ANOVA and F-tests were used for statistical analysis. ECOINVENT 3, Cumulative Energy Demand 1.09, and IPCC 2013 GWP100a were used for conversion factors. SIMAPRO 8.5.0.0 incorporated these databases and methods.

Table 4 presents grain yields, losses, and by-products. Yields ranged from 2600–3100 kg/ha (wet season) and 4200–5800 kg/ha (dry season). Harvesting losses were significantly different between wet and dry seasons; combine harvesting significantly reduced losses (0.9–1.7% vs. 4.0–9.3%). Stacking rice significantly affected discoloration and HRR. Discoloration was significantly higher in FP (6–8% vs. 3.8%), and HRR was lower (17–23% vs. 47%). In the dry seasons, improved practices reduced losses to 3.8–6.3% compared to 10–17% for FP. Improved practices significantly increased HRR (6–8%). Table 5 shows energy inputs, outputs, NEV, and NER. Total input energy from production and processing comprised 16–29% (wet season) and 25–43% (dry season) of total inputs. Dry season energy inputs were 30–50% higher than wet season inputs due to higher cultivation energy, yield, and processing needs. NEV was not significantly different between practices but was three times higher in the dry season due to higher yield and lower losses. NER was also not significantly different. Table 6 shows GHGE. There was no significant difference between FP and IPR. Wet season GHGE (5.3–5.7 Mg CO₂-eq/ha) was almost double the dry season due to higher cultivation emissions and grain loss. Cultivation contributed 70–90% of total GHGE. Grain losses increased GHGE by 30–50%. Table 7 shows cost components, NIV, and NIR. Total input costs were higher in the dry season due to higher yield. NIV was significantly higher in the dry season (748–963 $US/ha) compared to the wet season (almost no net profit). NIR for IPR with a combine harvester in the dry season was 30–40% higher than FP. Figure 3 shows the linear relationship between harvesting loss and NIV/NIR. Reduced losses significantly increased NIV and NIR. Figure 4 illustrates the energy balance, GHGE, and cost-benefits of different practices.

The findings demonstrate that improved post-harvest practices significantly reduced losses and increased profits for smallholder farmers. The study's post-harvest losses (up to 50% with FP) were substantially higher than the FAO's reported range of up to 30%. The LCA results, comparable to other studies after accounting for post-harvest losses, highlight the significant impact of losses on profitability, energy efficiency, and GHGE. The study's findings on energy consumption and GHGE for drying and milling were comparable to those from Taiwan, with minor differences likely attributable to variations in practices, germplasm, and technologies. Effective management of mechanized systems is crucial for optimal outcomes. Careful timing of harvest, drying within 24 hours, proper storage, and calibrated milling equipment were essential for minimizing losses. The cost analysis, while specific to the study area, reveals that the cost difference between IPR and FP is largely due to post-harvest losses. The use of global databases for energy efficiency and GHGE calculations minimizes local context influence, except labor, fuel, and equipment costs. Limitations include the use of secondary data for cultivation, the limited size of experimental plots (affecting harvesting loss estimates – addressed via sensitivity analysis), accounting of rice straw energy but not in-field emission estimates, and the use of general regional data for conversion factors.

Improved post-harvest practices significantly increased net income (30–50%) without increasing life-cycle energy or GHGE. Combine harvesting substantially reduced harvesting losses (3–7%), while improved drying and storage reduced discoloration (3–4%) and increased HRR (20–30%). Dry season practices yielded higher NEV and NIV (30–50%) and lower GHGE (40–60%) than wet season practices. Future research should replicate this study in other rice-producing countries to strengthen recommendations for improved post-harvest rice management.

The study's limitations include the use of secondary data for the cultivation stage, the relatively small size of experimental plots, the use of previously published data for some greenhouse gas emission estimates, and the reliance on broader regional data from global databases for some conversion factors. While a sensitivity analysis partially addressed the plot size issue, these factors could influence the generalizability of the findings.