KiwiWaste holds the chemical composition of organic wastes: sewage sludge, biosolids and wastewater, composts, manures, digestates, food and green waste, crop and agro-industrial residues, paper-mill sludge, animal by-products, liquid effluents, food-processing wastes (fruit, vegetable and potato residues, olive mill and winery wastes, brewery and milling by-products, dairy sludges, abattoir wastes and paunch, fish and shell wastes, sugar and oilseed residues and their effluents) and biochars. Every value was transcribed from a published source into the KiwiSolids workbooks, and every value in the application links back to that source.
KiwiWaste 收录有机废弃物的化学成分:污水污泥、生物固体和废水,堆肥,粪肥,沼渣,食物垃圾和绿化废弃物,作物与农工业残余物,造纸污泥,动物副产品,液体污水,食品加工废弃物(水果、蔬菜和马铃薯残渣,橄榄油厂和酿酒厂废弃物,啤酒酿造和制粉副产品,乳品污泥,屠宰场废弃物和瘤胃内容物,鱼类和贝壳废弃物,制糖和油料残余物及其污水),以及生物炭。每个数值都是从已发表的来源转录到 KiwiSolids 工作簿中的,应用中的每个数值都链接回其来源。
1What is in the database数据库中有什么
| Loading… |
2Using KiwiWaste使用 KiwiWaste
The application is built around the things you look at, rather than around tasks. Pick one and every tab shows what the database knows about it.
本应用围绕你要查看的对象而不是任务来组织。选中一个对象,每个标签页都会显示数据库中关于它的信息。
- Waste type (Wastes tab): a sheet in one of the workbooks, such as Biosolids or Vermicompost. A whole group can be pooled (“All composts”); pooling uses only the solid waste types of the group, so mg/kg and mg/L are never mixed.
- Composition: the waste type shown as a periodic table. Each element with data carries its geometric mean and, beneath it, its 1-GSD range. Properties (pH, EC, CEC, BOD5, COD, dry matter, organic matter, C:N and, for biochars, ash, volatile matter, fixed carbon, H/C and O/C), nutrient forms and contaminants are listed below the table.
- Component: one element, property or contaminant in the selected waste: its frequency distribution, fitted distribution, summary statistics, the guideline limits that apply, the same component in every other waste type, and the samples behind the numbers.
- Sample: one row of a workbook, opened in full, with its source.
- Guideline and Source: the standards and the references, each linked to the samples and components they concern.
- 废弃物类型(“废弃物”标签页):某个工作簿中的一张工作表,例如 生物固体或蚯蚓堆肥。整组可以合并(“全部堆肥”);合并时只使用该组中的固体废弃物类型,因此 mg/kg 与 mg/L 绝不会混在一起。
- 组成:以元素周期表形式显示的废弃物类型。每个有数据的元素都标出其几何均值,并在其下方标出 1-GSD 范围。各项性质(pH、EC、CEC、BOD5、COD、干物质、有机质、C:N,生物炭另有灰分、挥发分、固定碳、H/C 和 O/C)、养分形态和污染物列在表格下方。
- 成分:所选废弃物中的一种元素、性质或污染物:它的频数分布、拟合分布、汇总统计量、适用的标准限值、同一成分在其他每种废弃物类型中的情况,以及这些数字背后的样本。
- 样本:工作簿中的一行,完整打开,并附其来源。
- 标准限值和来源:各项标准和参考文献,每一项都链接到与之相关的样本和成分。
The selection is written into the address, so Copy link reopens exactly what is on screen. Export CSV writes every sample in the selection with its values as stored.
当前选择会写入地址栏,因此复制链接能原样重新打开屏幕上的内容。导出 CSV 会写出当前选择中的每个样本,数值按存储原样输出。
3How the numbers are derived数字是如何得出的
What a sample is
什么是样本
Each sample is one row of a source table: sometimes a single analysis, more often a mean, median or survey statistic that an author reported (the Statistic field says which). The statistics are taken across those rows. They therefore describe the published record for a waste type, not a random sample of all such wastes, and a national median of 1942 composts counts once, as does a single pile. Read them as “what the literature reports”.
每个样本是来源表格中的一行:有时是单次分析,更常见的是作者报告的平均值、中位数或调查统计量(统计量字段注明是哪一种)。统计量是跨这些行计算的。因此,它们描述的是某一废弃物类型的已发表记录,而不是所有同类废弃物的随机样本;一个由 1942 份堆肥得出的全国中位数只算一次,单个堆体也只算一次。请把它们理解为“文献所报告的情况”。
Units
单位
Solids are on a dry-matter basis, in mg/kg DM; liquids per litre, in mg/L (organic contaminants in µg/kg, µg/L or ng/L as the workbooks give them). Values of 10 000 mg/kg or more are displayed as a percentage (1 % = 10 000 mg/kg); smaller ones stay in mg/kg. The biochar workbook gives C, H, N, O and S in % DM; the build script converts these to mg/kg so that every solid shares one scale.
固体以干物质为基础,单位为 mg/kg DM;液体按每升计,单位为 mg/L(有机污染物按工作簿所给,为 µg/kg、µg/L 或 ng/L)。10 000 mg/kg 及以上的数值显示为百分比(1 % = 10 000 mg/kg);较小的数值仍以 mg/kg 显示。生物炭工作簿以 % DM 给出 C、H、N、O 和 S;构建脚本将其换算为 mg/kg,使所有固体使用同一尺度。
Which rows are used
使用哪些行
- Rows on a fresh-matter basis (Basis = FM) are listed but left out of the statistics, as the workbooks’ own Summary tabs leave them out.
- Biochars made from contaminated feedstock (CCA-treated or painted wood) are left out unless the option on the Composition tab includes them.
- Where two waste types in a pooled group report a component in different units (pathogen counts per gram against per 4 g, for instance), only the most common unit is pooled and the Component tab says so.
- 以鲜物质为基础的行(Basis = FM)会列出,但不计入统计,与工作簿自身的 Summary 标签页的做法一致。
- 由受污染原料(CCA 处理木材或涂漆木材)制成的生物炭不计入统计,除非在“组成”标签页上选择了将其纳入的选项。
- 当合并组中的两种废弃物类型以不同单位报告同一成分时(例如病原体计数按每克与按每 4 g),只合并最常用的单位,“成分”标签页会对此加以说明。
Values that are not plain numbers
不是普通数字的数值
- Below detection (‘<0.2’): set to half the stated limit by default, or to the limit, or omitted (Component tab). ‘<DL’ with no number given cannot be placed and is counted but not used.
- Ranges (‘12–45’): ignored by default, as in the workbooks; optionally entered at their geometric midpoint.
- Approximate (‘≈5’, ‘7.7 (7.2–8.2)’): the leading number is used.
- Greater than, absent, present and other text are counted and shown on the samples, but not used.
- 低于检出限(“<0.2”):默认取所述检出限的一半,也可取检出限本身,或略去(在“成分”标签页设置)。未给出数字的“<DL”无法定值,会计数但不使用。
- 范围(“12–45”):默认忽略,与工作簿一致;也可选择以其几何中点计入。
- 近似值(“≈5”“7.7 (7.2–8.2)”):使用前面的数字。
- 大于、不存在、存在及其他文本会计数并显示在样本上,但不使用。
The statistics
统计量
Concentrations in wastes are close to log-normal, so the periodic table and the contaminant lists lead with the geometric mean (GM, the exponential of the mean of the natural logarithms) and the geometric standard deviation (GSD, the exponential of their standard deviation). The 1-GSD range, GM ÷ GSD to GM × GSD, holds about two-thirds of the values of a log-normal distribution; it is the log-scale counterpart of mean ± SD. Zeros and negative values cannot be logged and are left out of the GM and GSD (the count used is shown). Properties such as pH and EC are reported as arithmetic mean ± SD. The Component tab also gives the median, percentiles, minimum and maximum.
废弃物中的浓度接近对数正态分布,因此元素周期表和污染物列表首先给出几何均值(GM,自然对数均值的指数)和几何标准差(GSD,自然对数标准差的指数)。1-GSD 范围,即 GM ÷ GSD 至 GM × GSD,约包含对数正态分布中三分之二的数值;它是均值 ± SD 在对数尺度上的对应量。零和负值无法取对数,不计入 GM 和 GSD(会显示实际使用的个数)。pH 和 EC 等性质以算术平均值 ± SD 报告。“成分”标签页还给出中位数、百分位数、最小值和最大值。
The frequency distribution uses logarithmic bins by default (linear on request); the dashed line is the GM and the shaded band the 1-GSD range.
频数分布默认使用对数分组(可选线性分组);虚线为 GM,阴影带为 1-GSD 范围。
Fitted distribution. Four distributions are fitted to the values by maximum likelihood: the normal, the log-normal (normal on ln x), the generalised normal (exponential power, with density proportional to exp(−|(x − μ)/α|β); β = 2 is the normal, β < 2 has heavier tails and a sharper peak, β > 2 is flatter) and the log-generalised normal. The one with the lowest small-sample Akaike information criterion (AICc) is shown; the likelihoods of the log models include the Jacobian (−Σ ln x) so all four are compared on the original scale. The three-parameter models are tried only from eight values and are chosen only when they improve the AICc by more than 2. The log models need every value above zero, so a component with zeros is fitted on the linear scale. The chart shows the fitted cumulative probability as a curve, the samples at (rank − ½)/n so the fit can be judged by eye, and the fit’s 1st–99th percentile shaded; the same fit is drawn over the histogram as the count it expects in each bin. The statistics table gives the fitted parameters, median and 5th–95th percentile, the AICc of every candidate and the Kolmogorov–Smirnov distance, and the guideline table adds the fitted probability of exceeding each limit. With fewer than three values, or all values equal, nothing is fitted and the empirical cumulative frequency is shown instead.
拟合分布。用最大似然法对数值拟合四种分布:正态分布、对数正态分布(ln x 服从正态分布)、广义正态分布(指数幂分布,密度与 exp(−|(x − μ)/α|β) 成正比;β = 2 即正态分布,β < 2 尾部更重、峰更尖,β > 2 则更平坦)以及对数广义正态分布。显示小样本赤池信息准则(AICc)最低的那一种;对数模型的似然包含雅可比项(−Σ ln x),因此四种分布都在原始尺度上比较。三参数模型只在有至少八个数值时才尝试,并且只有当它们使 AICc 改善超过 2 时才会选用。对数模型要求每个数值都大于零,因此含零值的成分在线性尺度上拟合。图表以曲线显示拟合的累积概率,把样本绘于 (rank − ½)/n 处,以便目测判断拟合效果,并以阴影显示拟合分布的第 1–99 百分位数区间;同一拟合也以每个分组的期望计数绘于直方图上。统计表给出拟合参数、中位数和第 5–95 百分位数、每个候选模型的 AICc 以及 Kolmogorov–Smirnov 距离;标准限值表还会加上超过每个限值的拟合概率。数值少于三个或全部相等时,不进行拟合,改为显示经验累积频率。
4Guidelines标准限值
Guidelines come from the Standards & Limits tabs of the compost and biochar workbooks (EU Fertilising Products Regulation, AS 4454, NZS 4454, BioGro, the NZ Biosolids Guidelines Grade A, US Composting Council STA, Washington WAC, EBC and IBI), plus five transcribed into the build script: NZ Biosolids Guidelines Grade b, US EPA 40 CFR 503 Table 1 ceiling concentrations, the EU Sewage Sludge Directive 86/278/EEC, and the FAO recommended maximum concentrations for irrigation water. Each is marked with the waste groups and matrix it was written for. A waste is compared by default only with the guidelines written for it; the option to offer the others is on the Composition tab.
标准限值取自堆肥和生物炭工作簿的 Standards & Limits 标签页(EU Fertilising Products Regulation、AS 4454、NZS 4454、BioGro、NZ Biosolids Guidelines Grade A、US Composting Council STA、Washington WAC、EBC 和 IBI),另有五项转录在构建脚本中:NZ Biosolids Guidelines Grade b、US EPA 40 CFR 503 Table 1 上限浓度、EU Sewage Sludge Directive 86/278/EEC,以及 FAO 灌溉水推荐最高浓度。每项都标明了其所针对的废弃物组和基质。默认情况下,一种废弃物只与为其制定的标准限值比较;提供其他标准限值的选项在“组成”标签页上。
On the periodic table a chosen guideline shades each element green where both the GM and GM × GSD are below its limit, amber where the GM is below but the upper end of the range is above, and red, with a bar, where the GM exceeds it. On the Component tab the share of samples beyond each limit is counted from the values used in the statistics.
在元素周期表上,所选标准限值按以下规则为每个元素着色:GM 和 GM × GSD 都低于其限值时为绿色;GM 低于限值但范围上端高于限值时为琥珀色;GM 超过限值时为红色,并加一条横杠。在“成分”标签页上,超过每个限值的样本比例是根据统计中所用的数值计算的。
5Adding data添加数据
KiwiWaste is built from the workbooks, so the usual way to add data is to add it to them and rebuild.
KiwiWaste 由工作簿构建,因此添加数据的常规做法是把数据加入工作簿,然后重新构建。
- Add rows to the right tab of the right workbook in
000 Kiwidata/KiwiSolids/, with the tab’s column conventions (README tab of each workbook): mg/kg DM, text such as ‘<0.5’ or ‘12–45’ as printed, the full reference in Source of Information and the DOI in Source URL. A new tab becomes a new waste type. A new workbook named<Name>_Chemical_Composition_Comprehensive.xlsxbecomes a new group. - Rebuild from the site root:
python3 kiwiwaste/scripts/build_data.py
It needs nothing but Python 3. It reports the counts, and lists any column heading it does not recognise; those are kept as text on each sample until a line is added toPARAMSin the script. - New group only: add its
data/samples-<group>.jsto the list of scripts at the foot ofkiwiwaste/index.html(and of this page), and give it a name and description inGROUPSin the script. - Reload the page.
- 添加行:在
000 Kiwidata/KiwiSolids/中相应工作簿的相应标签页里添加行,并遵循该标签页的列约定(见各工作簿的 README 标签页):单位为 mg/kg DM,像“<0.5”或“12–45”这样的文本按原样录入,完整参考文献填入 Source of Information,DOI 填入 Source URL。新标签页会成为新的废弃物类型。命名为<Name>_Chemical_Composition_Comprehensive.xlsx的新工作簿会成为新的组。 - 重新构建:在网站根目录运行
python3 kiwiwaste/scripts/build_data.py
它只需要 Python 3。它会报告计数,并列出所有无法识别的列标题;在脚本的PARAMS中为其添加一行之前,这些列会作为文本保存在每个样本上。 - 仅限新组:把它的
data/samples-<group>.js加入kiwiwaste/index.html(以及本页)底部的脚本列表,并在脚本的GROUPS中为其设置名称和说明。 - 重新加载页面。
For a handful of samples typed in by hand, or to try a dataset before it goes into a workbook, use kiwiwaste/data/user-data.js. The build never overwrites it, and the comment at its head shows the format: a waste-type id, a name, a source key and the values keyed by component id (Zn, Cd, ph, ec, cec, bod, cod, pfas…). The same objects can be added from the browser console with KiwiWaste.db.addSamples([...]) followed by KiwiWaste.redraw().
如果只是手工录入少量样本,或想在数据集进入工作簿之前先试用,请使用 kiwiwaste/data/user-data.js。构建过程从不覆盖该文件,文件开头的注释说明了格式:废弃物类型 id、名称、来源键,以及以成分 id 为键的数值(Zn、Cd、ph、ec、cec、bod、cod、pfas…)。同样的对象也可以在浏览器控制台中用 KiwiWaste.db.addSamples([...]) 添加,再运行 KiwiWaste.redraw()。
New guidelines go in GUIDELINES_EXTRA in the build script, or as a row in a workbook’s Standards & Limits tab.
新的标准限值写入构建脚本中的 GUIDELINES_EXTRA,或作为一行加入工作簿的 Standards & Limits 标签页。
6Sources and terms来源与条款
The Sources tab lists every reference with its DOI and the samples taken from it; the extraction notes (which table, which page, which corrections) are in 000 Kiwidata/KiwiSolids/sources/. The compiled values may be reused with attribution to the original publications and to KiwiScience. Guideline values remain the property of the issuing bodies.
“来源”标签页列出每篇参考文献及其 DOI 和取自该文献的样本;提取说明(哪张表、哪一页、做了哪些更正)位于 000 Kiwidata/KiwiSolids/sources/。汇编的数值可以重复使用,但须注明原始出版物和 KiwiScience。标准限值仍归发布机构所有。
KiwiWaste is a reference compilation for research and teaching. It is not a substitute for analysing the waste you intend to use.
KiwiWaste 是供研究和教学使用的参考汇编,不能替代对你打算使用的废弃物所做的分析。