Why timing slips in 2026
Late picking in 2026 is rarely just “the harvest is late”. It is usually a mix of phenology, weather, and orchard access. In key Italian origins, hazelnuts typically fall and are collected from late August through September, with later cultivars and areas stretching into early October. Week-to-week climate variation can shift that window, even inside the same PGI area.
Regional signals matter because “Italy” is not one harvest curve. In Piedmont for Tonda Gentile Trilobata and the Nocciola Piemonte PGI context, buyers should watch nut drop percentage and ground moisture, because wet soil slows collection and increases time on the orchard floor. In Campania, including Tonda di Giffoni areas, harvest can run later, and medium-late zones can extend into mid-October. In Sicily, including Nebrodi, September to October is common, and late rains can push collection later while increasing exposure time on wet ground.
The most useful question for procurement is: is it truly late, or just staggered? Hazelnuts do not always ripen in one clean wave. Trials in Central Italy show nut drop can run from late August to end September, and some cultivars can extend into early October. That spread can be normal, but it becomes a commercial problem when wet ground and limited access add extra days before nuts reach drying.
Supplier reporting should focus on field indicators, not calendar dates. Ask for the percentage of husk separation, the percentage of nuts naturally fallen, the kernel moisture trend, and whether machinery can operate. If alleys are muddy and sweepers or vacuums cannot run, harvest pauses and deliveries slide.
Delays also compress the post-harvest chain. Dehusking, cleaning, and drying capacity that works in a normal season can become the constraint in a late one. That is when you see specification drift in moisture, defects, mold, and missed delivery windows for EU confectionery and ingredient plants.
Moisture management in late harvest
Moisture targets are not negotiable in B2B trade, even when harvest is late. Common trade specifications for shelled hazelnuts often sit around 3 to 6 percent moisture, and drying practice frequently targets about 6 percent moisture with water activity around 0.70 to 0.71 to suppress fungal growth.
Late harvest creates a moisture trap because nuts pick up water from the ground. PGI Piedmont references this absorption effect, and it is exactly why “a few extra days” can change the stabilization workload. Buyers should require incoming moisture and water activity certificates of analysis by lot, both pre- and post-drying, so you can see whether the supplier is catching wet lots early.
Operationally, the best control is speed and segregation. Pre-cleaning to remove soil and leaf litter should happen fast, and drying should start immediately. Wet lots should be separated into dedicated dryer runs, because mixing them with drier intake can slow the whole batch and keep water activity high for longer.
One checklist is worth using here, because it is easy to audit in a delayed season:
- Define a maximum time from pickup to dryer start as a supplier KPI, often 12 to 24 hours depending on the local setup
- Require documented dryer logs for time, temperature, and airflow by lot
- Require water activity re-testing after tempering, not only right after drying
- Require a clear rule for “wet lot” segregation and labeling through storage and packing
Capacity planning ties directly into moisture risk. If a plant is set up for 20 t/day of in-shell intake, a late harvest can push intake pressure far above that for a short period. Without added dryer capacity or longer operating hours, nuts sit warm and wet while they wait. That is when water activity stays elevated and mold and aflatoxin risk rises.
Buyers often ask if higher incoming moisture is acceptable if the final product meets spec. Generally yes, but only if the supplier can prove drying speed and control. The risk is not the starting number, it is the time spent at high water activity before stabilization.
Kernel quality impacts
Late harvest shows up in kernel performance, not just in lab moisture. Industrial buyers often see shifts in kernel size distribution and counts, more shrivel and serious shriveling type defects, and lower yield at shelling. Many origins market quality on high shelling yield and roasting aroma, so these changes hit both cost-in-use and sensory outcomes.
Blanching and skin removal can also become less consistent in late, wet conditions. Buyers of Piedmont-style kernels value high natural peelability, and staining can increase when nuts spend longer on damp ground. In 2026 contracts, it is reasonable to add an acceptance clause for blanching performance, such as a maximum residual skin percentage after an agreed roast and blanch protocol, because that is what affects throughput and rework.
Roasting performance is another place where late harvest can surprise you. Uneven moisture across lots can drive uneven browning, and drying time and temperature influence oxidative stability. When harvest is delayed, ask for roast test reports per lot, focused on color and flavor outcomes and rancidity-related markers, so you can catch problems before they reach paste or inclusions.
Defect language should be agreed up front, especially when supply is tight. U.S. import and grade frameworks explicitly use defect tolerances and caps on mold and insect injury within defect allowance, and the practical takeaway for EU buyers is the same: define defect caps and the sampling method before the first truck arrives. If you wait until after delivery, you end up negotiating under pressure.
Functional questions from confectionery customers tend to surface late, when it is hardest to fix. If you supply paste, gianduja, nibs, or granella, add functional specs that connect to processing behavior. Fat percentage, peroxide value, and free fatty acids are common anchors, and you can also align on refining behavior and sensory hold time for paste customers.
Mould and aflatoxin risk
Compliance is the hard stop in a late season. EU maximum levels are commonly cited at Aflatoxin B1 5 µg/kg and total aflatoxins 10 µg/kg for hazelnuts intended for final consumer or ingredient use, but contracts should always cite the applicable regulation and the exact product category, because limits depend on placement and use.
The mechanism is straightforward. Aflatoxin risk rises when nuts spend longer wet, meaning higher water activity for longer. Guidance commonly recommends drying rapidly to water activity below 0.70 to reduce aflatoxin formation potential, which aligns with the water activity targets many processors already use for stability.
Field hygiene is part of prevention, not a nice-to-have. Buyers can require orchard floor mowing and cleaning, removal of “mummies,” multiple fast collection passes, and strict segregation of visibly moldy or insect-damaged fractions before drying. That segregation point matters because cross-contamination can spread risk through shared handling and storage.
Rain breaks are where late seasons go wrong. A Turkish multi-year study reported that extended drying during rain increased water activity above 0.80 and increased aflatoxin risk. That is a strong justification for contingency dryers, covered drying areas, and clear stop rules when weather makes open handling unsafe.
Testing plans should match the 2026 risk profile. Move from annual or spot testing to lot-based testing with defined sampling across large lots, rapid screening followed by confirmatory methods, and re-testing after storage if shipments are held. It is also worth building RASFF awareness into your internal escalation process, because hazelnuts are periodically flagged for aflatoxin in EU alerts and the reputational cost of a border issue is high.
Contract timing under delay
Fixed delivery dates are fragile in delayed seasons. A better approach is to specify delivery windows with a floating mechanism, such as a week-number range, and define triggers based on harvest start declaration or lot readiness once final moisture and water activity are achieved. That reduces disputes and aligns delivery promises with the real constraint, which is stabilization.
Delay-sensitive specs should be explicit. Moisture and water activity at dispatch, maximum mold percentage, insect damage, foreign matter, and aflatoxin compliance should sit in the main body of the contract, not in an email chain. Where you reference common trade specs, attach test methods and the sampling plan as an appendix so both sides know what “pass” means.
Penalties and price clauses need careful wording. Step-down pricing for late delivery only makes sense when delay is within supplier control. Otherwise, define force majeure or changed circumstances handling and a renegotiation path, since contract guidance in agriculture highlights force majeure and quality shortfall handling as key clauses.
Partial shipments can keep factories running, but they need rules. If a confectionery buyer has Q4 production commitments and the harvest slips by weeks, the contract can allow early lots to ship first while the remainder follows later. At the same time, you should restrict mixing of lots produced under different drying regimes without buyer approval, because that is how you get inconsistent roast and paste behavior.
Dispute-proofing is cheap compared to a rejected container. Require pre-shipment certificates of analysis, retained reference samples, and an agreed third-party lab for arbitration. Clarify the Incoterms point where quality risk transfers, especially if delays force longer on-farm storage before dispatch.
Logistics and capacity planning
Dryer capacity is the bottleneck in late years, not the orchard. Even well-run orchards can fail specs if nuts wait 48 to 72 hours before drying because intake queues build up. Buyers should ask for the supplier’s daily drying capacity in t/day and the peak-week plan, including what happens when multiple origins deliver at once.
Labour and machinery constraints also tighten in compressed harvest windows. Multiple collection passes, more sorting, and more handling require more people at the same time other crops compete for labor. Stress-test supplier plans for machine uptime, spare parts, and backup contractors for sweeping and vacuuming, because a few days of downtime can cascade into quality loss.
Transport is another weak link when fields are wet. Wet conditions reduce truck access and can increase foreign matter, which adds cleaning time and slows intake. Contingencies like staging hubs with covered storage and pre-booked freight can reduce harvest-week rate spikes and prevent nuts sitting in uncontrolled conditions.
Contingency sourcing should be planned before the season starts. Dual-origin coverage, such as early versus late Italian regions, can reduce the risk of a single bottleneck, and any blend strategy must stay within your agreed size, count, and roast profile requirements. Pre-approving alternates also helps R&D avoid restarting under time pressure.
A weekly KPI dashboard from suppliers is one of the simplest ways to stay ahead of 2026. Ask for harvested percentage, lots dried to spec at or below 6 percent moisture and around 0.70 water activity, aflatoxin test turnaround time, stocks by grade, and confirmed ETAs. That gives procurement room to adjust promotions, factory schedules, and inventory decisions before the crunch hits.