Transparent by default / 默认透明 — supply chain · business transforming · product
Transparency is not a report at the end of the year. It is the setting we work in — open by default, sourced by default, checkable by default. Question #1 is rice.
透明不是年末的一份报告，而是我们工作的默认设置——默认开放、默认标注来源、默认可被核查。第一个问题：稻米。
The right question comes from no assumption. It stays open — until it becomes an action.
对的问题不来自任何假设。它一直保持开放——直到它变成一个行动。
A question is only finished when someone can do something on Monday morning. Before that it is a topic. The ladder below is the real sequence we follow, step by step, no shortcuts.
一个问题只有在“周一早上有人能动手”时才算完成。在那之前它只是个话题。下面这八级，是我们真正走的顺序，不跳级。
A note on Step 1: if the first real question is "how do we get credit?" or "how do we pass the audit?", the process falls into a hard hole early. The best Step 1 is simply: do, then check. Do it again, check again. Repeat until the results are good — and by then, validation follows naturally, because the evidence is already there.
关于第一步的说明：如果真正的第一个问题是"我们怎么拿到碳信用？"或"怎么通过审核？"，整个过程会很快掉进一个硬坑。最好的第一步很简单：做，然后检查。再做，再检查。重复，直到结果变好——到那时，核证会自然跟来，因为证据早就在那里了。
No single global law defines the phrase. In practice it is a three-axis claim, and a credible program must pass all three — a low carbon number on bad grain, or good grain from burnt straw, is not it.
目前没有一部全球法律给这个词下定义。实际操作中它是一个三轴主张，可信的项目必须三轴同时成立：碳数字低但米质差不算；米质好但秸秆焚烧也不算。
Variety purity and certified seed; milling / head-rice yield; chalkiness, amylose, aroma; moisture at delivery (≈14%); no foreign matter; food safety limits on cadmium, inorganic arsenic and pesticide residues (Codex MRLs).
品种纯度与合格种子；出米率与整米率；垩白度、直链淀粉、香气；交付水分（约14%）；无杂质；镉、无机砷与农残符合限量（Codex）。
kg CO₂e per tonne of paddy, cradle-to-farm-gate: field CH₄ (the dominant term), N₂O from fertiliser, pumping energy, and embedded inputs. Quantified with IPCC 2019 Refinement factors and declared under ISO 14067 / 14064-2.
以“每吨稻谷的千克二氧化碳当量”计，范围到农场门口：田间甲烷（主项）、化肥氧化亚氮、抽水能耗与投入品隐含排放。用IPCC 2019修订版系数量化，按ISO 14067 / 14064-2声明。
Safe alternate wetting & drying (AWD) or mid-season drainage; straw removed or aerobically composted, never burnt; right-rate nutrients; water fairness among neighbours; labour rights; and lot-level traceability so the claim follows the rice.
安全的干湿交替（AWD）或中期排水；秸秆离田或好氧堆肥，严禁焚烧；精准施肥；用水公平；劳工权益；批次级可追溯，让主张跟着这批米走。
Why it works: methane is made by microbes that only work without oxygen. Let the water fall below the root zone once or twice a season and you interrupt them. The single biggest amplifier is fresh straw ploughed in just before flooding — that is food for those microbes. The trade-off to watch: drying too hard raises N₂O and can cost yield. Never drop below −15 cm, and keep the field flooded from panicle initiation through flowering.
为什么有效：甲烷由只在无氧条件下工作的微生物产生。一季中让水位落到根区以下一到两次，就打断了它们。最大的放大因素是灌水前刚翻入的新鲜秸秆——那是它们的粮食。需要盯住的取舍：落干过度会推高氧化亚氮并减产。水位不要低于−15厘米；幼穗分化到开花期必须保持淹水。
Three different families are often confused. Quantification grammar tells you how to count. Farm standards tell you how to grow. Crediting methodologies tell you how to sell the reduction. A program needs all three, in that order.
三类东西常被混为一谈：量化规则告诉你怎么算；农场标准告诉你怎么种；碳信用方法学告诉你怎么把减排量卖出去。一个项目按这个顺序全都需要。
IPCC 2019 Refinement, Vol.4 Ch.5: baseline daily CH₄ emission factor for continuously flooded, un-amended fields, adjusted by scaling factors for water regime and organic amendments. This is the arithmetic everyone else borrows.
ISO 14064-2 (project GHG), ISO 14067 (product footprint), ISO 14040/44 (LCA), plus the GHG Protocol Land Sector & Scope 3 rules for buyers.
IPCC 2019修订版第4卷第5章：持续淹水、未施有机物基准田的日甲烷排放因子，再按水分管理与有机物投入的调整系数修正——这是所有其他体系借用的算术。ISO 14064-2（项目层级）、ISO 14067（产品碳足迹）、ISO 14040/44（生命周期）、以及采购方使用的GHG Protocol 土地部门与范围三规则。
The Sustainable Rice Platform (SRP) Standard for Sustainable Rice Cultivation — a requirement set plus performance indicators covering water, nutrients, residue, food safety, labour and yield. Its Assurance Scheme has tiers: self-assessment, second-party, and third-party verified, with chain of custody for the label.
Third-party audits are delivered by approved certification bodies such as GLOBALG.A.P.-recognised auditors and Control Union.
可持续稻米平台（SRP）《可持续水稻种植标准》：要求条款＋绩效指标，覆盖水、养分、秸秆、食品安全、劳工与产量。其保证体系分三层：自评、第二方、第三方核证，并带标签的监管链。第三方审核由获批认证机构执行，例如GLOBALG.A.P.认可审核员、Control Union。
UNFCCC CDM AMS-III.AU “adjusted water management practice in rice cultivation” (registry); Gold Standard’s water-management methodology; Verra VCS VM0042 improved agricultural land management; the Climate Action Reserve Rice Cultivation Protocol and ACR in the US; Japan’s J-Credit methodology for extended mid-season drainage.
联合国CDM AMS-III.AU“水稻调整水分管理”；Gold Standard 水分管理方法学；Verra VCS VM0042 农地管理改良；美国Climate Action Reserve《水稻种植议定书》与ACR；日本J-Credit 延长中期落干方法学。
Our read of public policy and program evidence, region by region. Each line carries a confidence tag, because a status without a tag is an opinion pretending to be data. ◆ HIGH = documented policy or protocol. ◇ VERIFY = direction is clear, the delivered hectares are not — go and count.
这是我们基于公开政策与项目证据的判读，逐国逐区列出。每行都带置信标签——没有标签的“状态”只是伪装成数据的意见。◆ HIGH＝有明文政策或议定书；◇ VERIFY＝方向明确但落地面积待核实，请去数一遍。
Targets are anchored to what is already signed: the Global Methane Pledge (−30% methane by 2030 against 2020) and IPCC 1.5°C pathways (roughly −50% methane by mid-century). Everything below is what that means for a rice field.
以下节点锚定已经签署的东西：全球甲烷承诺（2030年较2020年减少30%）与IPCC 1.5℃路径（本世纪中叶甲烷约减半）。下面写的是：这些数字落到一块稻田上意味着什么。
Rice is roughly 8–11% of human-caused methane. Verified low-methane practice still covers only low single digits of global harvested area ◇ VERIFY. The scarce asset is not technology — it is a trustworthy water record.
水稻约占人类活动甲烷排放的8–11%。经核证的低甲烷做法仍只覆盖全球收获面积的个位数低位。稀缺的不是技术，而是一份可信的水情记录。
Observation pipes and water logs become standard issue. Straw off-field becomes the default, not the exception. First premium contracts with an emission-intensity clause. Group certification replaces one-farm-at-a-time audits.
观测管与水位台账成为标配；秸秆离田成为默认而非例外；出现首批带排放强度条款的溢价合同；集体认证取代一户一审。
Irrigation systems are rebuilt for intermittency — gates you can actually close, drainage that actually drains. Shorter-duration and low-exudation varieties are in farmers’ hands, not in trials.
灌溉系统按“可间断供水”重建：闸门真能关、排水真能排。短生育期、低根系分泌品种进入农户手中，而不再停留在试验田。
Lot-level traceability from field to mill to shelf. Laggard regions financed by transition funds rather than punished by trade barriers — because a farmer without a pump cannot choose.
从田块到米厂到货架的批次级追溯。滞后地区由转型资金支持，而不是被贸易壁垒惩罚——没有水泵的农户，本无从选择。
Rice methane at least halved from 2020 with yield and farm income higher, water use lower, arsenic in grain lower. AWD is not a project any more. It is simply how rice is grown, and nobody remembers being asked.
稻田甲烷较2020年至少减半，同时产量与农户收入更高、用水更少、米中砷更低。干湿交替不再是“项目”，它就是种稻的方式——没人记得曾被要求过。
Our read of where the outcome actually comes from. It is not one huge cheque. It is timing decisions, repeated on time, made possible by cheap physical control.
我们的判读：结果主要来自哪里。它不是一张大额支票，而是许多次准时做出的时间决策——前提是有便宜可靠的物理控制手段。
When to let it dry, when to re-flood, when straw goes in. Free to decide, costly to get wrong.何时落干、何时复灌、秸秆何时入田。决策不花钱，做错很贵。
Field levelling, repaired bunds, working inlet/outlet gates, drainage, a pump you are allowed to switch off, a 30 cm observation pipe.平地、修田埂、能用的进出水口、排水通畅、允许关掉的水泵，以及一根30厘米观测管。
Farmer field schools, water-user groups, and one trusted local demonstrator per village who others can walk over and look at.农民田间学校、用水户小组，以及每个村一位可信的本地示范户——别人走过去就能看见。
MRV platforms, balers, dryers, mills, aggregation. Necessary — but capital without water control buys nothing.监测平台、打捆机、烘干机、米厂、聚合服务。必要——但没有配水控制，资本什么也买不到。
Written to be read standing in a field, in boots, before transplanting. Adjust to local variety, soil and irrigation schedule — and skip AWD entirely on saline soils or fields you cannot drain.
这份内容是为“穿着靴子站在田里、插秧前”读的。请按本地品种、土壤与灌溉排班调整；盐渍土或无法排水的田块不要做干湿交替。
Record plot size and GPS, variety, seed rate, transplant date, fertiliser rate and dates, pump hours, yield, and cost — for one full season before changing anything.
Why: without a before, there is no after. No claim, no premium, no credit, and no way to know if you gained.
为什么：没有“之前”，就没有“之后”。无基准就无法主张、无溢价、无信用，也无法知道自己是否受益。记录：地块面积与GPS、品种、播量、插秧日期、施肥量与日期、抽水小时数、产量、成本——先完整记一季，再改动作。
Aim for under ±5 cm variation across the plot; laser levelling if a service exists, a simple drag board and water-mirror check if not.
Why: on an uneven field you must flood 10 cm deep just to cover the high spots — and deep water is exactly what makes methane. Levelling also evens out germination and saves fertiliser.
为什么：田面不平，你必须灌到10厘米才能淹住高处——而深水正是产甲烷的条件。平地还能让出苗均匀、省肥。目标：田内高差小于±5厘米；有条件用激光平地，没有就用拖板加“水面镜”检查。
Seal cracks and rat holes, raise bunds to 20 cm, fit a closable inlet and a real outlet at the low corner.
Why: AWD is a decision about water — and you cannot decide anything you cannot hold or release. A leaking bund silently re-floods your field and erases the whole practice.
为什么：干湿交替是关于水的决策——留不住、放不掉，就无从决策。漏水的田埂会悄悄把田重新灌满，让整套做法归零。做法：封裂缝与鼠洞、田埂加高到20厘米、进水口可关闭、低角设真正的排水口。
A 30 cm PVC pipe, 10–15 cm diameter, drilled with 5 mm holes every 3–5 cm; bury 20 cm into the soil leaving 10 cm above, in a representative spot you pass every day. Mark −15 cm inside with paint.
Why: you cannot manage what you cannot see. The pipe turns an invisible water table into a number a child can read — and into audit evidence.
为什么：看不见就管不了。观测管把看不见的地下水位变成一个孩子都读得懂的数字，也变成审核证据。做法：30厘米PVC管，直径10–15厘米，每3–5厘米钻5毫米孔；埋入土中20厘米、露出10厘米，装在你每天经过的代表性位置，管内用漆标出−15厘米。
Keep 3–5 cm of water for the first 2 weeks after transplanting (or until weeds are controlled). Then let the water fall; when it reaches −15 cm in the pipe, re-flood to +5 cm. Repeat. Keep the field flooded from panicle initiation through flowering. Drain finally 10–14 days before harvest.
Why: each dry phase stops methane production; the flooded flowering window protects yield; the final drainage lets machines in and keeps grain quality.
为什么：每次落干都会中断产甲烷；开花期保持淹水以保产；收前落干便于机械进田并保证米质。做法：插秧后前2周（或杂草控制住前）保持3–5厘米水层；之后让水自然下落，管内降到−15厘米即复灌到+5厘米，如此循环；幼穗分化至开花期保持淹水；收获前10–14天最后一次排干。
Bale and remove, or compost aerobically in a heap off-field. If you must incorporate, do it early with at least 30 days of dry soil before flooding. Never burn.
Why: fresh straw ploughed in just before flooding is the single largest methane amplifier — you are feeding the microbes. Burning trades methane for smoke, lung disease and lost nutrients.
为什么：灌水前刚翻入的新鲜秸秆是最大的甲烷放大器——等于给产甲烷菌喂饭。焚烧是把甲烷换成烟尘、呼吸道疾病与养分流失。做法：打捆离田，或田外好氧堆肥；必须还田则提早翻入，并保证灌水前有至少30天旱地分解期；严禁焚烧。
Split nitrogen into 3 applications matched to growth stage; use a leaf colour chart or soil test rather than habit; deep-place or incorporate urea instead of broadcasting onto standing water.
Why: drier soil is more likely to release N₂O, a gas ~273× stronger than CO₂. Excess nitrogen is how a methane win becomes a nitrous oxide loss — and it is also the most wasted cash in the field.
为什么：土壤偏干时更易释放氧化亚氮，其增温潜势约为二氧化碳的273倍。氮肥过量会把甲烷的收益变成氧化亚氮的损失，同时也是田里最浪费的一笔现金。做法：按生育期分3次施氮；用叶色卡或土测代替经验；尿素深施或翻入，不要撒在水面上。
Use certified seed of a variety the buyer actually wants; cut seed rate to the recommended level; consider shorter-duration or direct-seeded systems where water is scarce.
Why: quality is half the claim. A shorter season means fewer flooded days, which means less methane — and a variety with milling and eating quality is what earns the premium that pays for all of the above.
为什么：“优质”占主张的一半。生育期越短，淹水天数越少，甲烷越少；而具备出米率与食味品质的品种，才是支付上述一切成本的溢价来源。做法：使用买方真正需要的品种的合格种子；播量降到推荐水平；缺水地区考虑短生育期或直播体系。
One notebook or one phone app per plot: weekly pipe reading with date, every flood and drain event, inputs, pump hours, harvest weight and moisture. Photograph the pipe with a visible date. Keep three seasons.
Why: the record is the asset. It is what an auditor verifies, what a buyer pays for, and what turns a good habit into a certified, sellable claim. A false entry costs everything — transparency by default, including the bad seasons.
为什么：记录本身就是资产。审核员核查它、买方为它付钱，它把好习惯变成可核证、可销售的主张。一条假记录会让全部作废——默认透明，包括不好的那些季节。做法：每地块一本或一个手机应用：每周带日期的水位读数、每次灌排、投入品、抽水小时、收获重量与水分；拍下带日期的观测管照片；保留三季。
For the auditor with muddy boots. Sampling rule of thumb for group certification: at least √n of members, minimum five, with ≥10% unannounced. Interview farmers away from the group leader.
写给鞋上带泥的审核员。集体认证抽样经验法则：不少于成员数的平方根、最少5户，其中≥10%不通知抽查。访谈农户时请与组长分开进行。
An indicative scenario for 2030, written so you can disagree with it line by line. Replace any line with your own baseline — and if your number differs, publish why. That is what transparent by default means.
这是一个2030年的示意性情景，故意逐行写出来，方便你逐行反驳。用你自己的基准替换任意一行——若结果不同，请公布原因。这就是“默认透明”。
59Mt CO₂e / yr by 2030
And on the 2050 view — half the addressable pool — the same arithmetic gives roughly 295 Mt CO₂e per year, from a practice whose main equipment is a plastic pipe with holes in it.
若看到2050年——可作为对象的一半——同一算式给出约每年2.95亿吨二氧化碳当量；而这套做法的主要设备，是一根钻了孔的塑料管。
A gate you can close and a schedule you can rely on. Everything else is downstream of this.
一个关得住的闸门、一份靠得住的排班。其余一切都在它下游。
Below that, MRV and audit cost more than the value created. Groups, not heroes.
低于这个规模，监测与审核成本超过创造的价值。要靠组织，而不是靠英雄。
An emission-intensity condition in a real purchase contract moves more hectares than a campaign.
真实采购合同里的一条排放强度条款，比一场宣传推动的面积更多。
Satellite radar for the area, sampling for the truth, phone logs for the daily record.
面积靠卫星雷达、真值靠抽样、日常靠手机台账。
If the premium stops at the trader, the pipe stops being read by week six.
若溢价止步于贸易商，第六周起就没人读观测管了。
Extension curricula, subsidy conditions, seed programs, credit and insurance terms, school lessons. Then it needs no campaign.
推广教材、补贴条件、种子项目、信贷与保险条款、学校课程。到那时就不需要宣传了。
Yes — it belongs at the top of the agenda, and not as a chapter about rice. A village is the smallest unit where water, land, labour, market and behaviour all sit in the same room. Ten years is long enough for a practice to outlive a project, and short enough that the same people are still there to see it.
是的——它应该排在议程最前面，而且不该只是“关于稻米的一章”。村庄是水、土地、劳动力、市场与行为唯一同时在场的最小单元。十年，足够让一种做法活得比项目更久，也短到当年那些人还在，能亲眼看见。
Map every plot, water source and pump. Measure yield, cost, water hours and straw fate for one season. Publish it on one board in the village where anyone can correct it.
测绘每一块田、每个水源与水泵；完整记录一季的产量、成本、抽水小时与秸秆去向；把结果公示在村里一块板上，任何人都可以来改错。
10–20 volunteer farmers, pipes in the ground, weekly readings, and one side-by-side plot so the difference can be walked to, not argued about.
10–20户自愿农户、观测管入土、每周读数，并设一块对照田——差别是走过去看的，不是吵出来的。
Form a water-user group with a written rotation. Fix gates, level fields, clear drains. Buy or share a baler, or build the compost yard.
成立用水户小组并写明轮灌规则；修闸门、平田、清排水沟；购置或共享打捆机，或建设堆肥场。
Group certification, one verified low-emission lot sold with a premium, and an MRV data package the buyer can audit. Reinvest part of the premium in the next 100 hectares.
完成集体认证；以溢价售出首个核证低排放批次；交付买方可审核的MRV数据包；将部分溢价再投入下一个100公顷。
Dry-season rotation, credit and insurance linked to practice, the checklist inside the village by-law, the water rule taught in the local school, and every new farmer onboarded as a matter of course. The project ends; the practice does not.
旱季轮作；信贷与保险与做法挂钩；检查清单写进村规民约；用水规则进入本地学校课堂；每位新农户都按惯例完成上手培训。项目结束，做法不结束。
What’s next is not another plan. It is question #2: who controls the water — and are they allowed to turn it off?
下一步不是再写一份计划，而是第二个问题：水由谁掌控——他们被允许关掉它吗？
Primary bodies and registries used above. Program hectares, national targets and 2035–2050 figures move fast — always open the source before quoting a number, and re-date it when you do.
上文使用的主要机构与登记处。项目面积、国家目标以及2035–2050的数字变化很快——引用任何数字前请先打开来源，并注明查阅日期。