Zone II • Versions EN1
Abstract: Despite the rapid development of China’s feed industry, feed resources were in an increasing shortage in recent years. Abuse of mineral additives, together with fecal pollution, has severely hindered a stable, healthy and sustainable development of the feed industry in China. It seems imperative more than ever to establish a database of feedstuffs and their mineral elements. This study surveyed the production, utilization, and distribution of over 98% major feedstuff resources in China and their mineral contents, covering 15 energy feedstuffs, 7 major plant protein feedstuffs, 6 major animal protein feedstuffs, 8 green forages and 4 major mineral feedstuffs. The dataset presented here provides scientific statistical guidance for reasonable use of mineral additives in feed formulations.
Keywords: feedstuff; mineral element; conventional ingredients; distribution; contents
|English title||A dataset of major feedstuffs in China and their mineral contents|
|Data corresponding author||Luo Qingyao (firstname.lastname@example.org)|
|Data authors||Luo Qingyao, Liao Xiudong|
|Time range||2015 – 2019|
|Data volume||1.4 GB|
|Data format||*.TXT, *.JPEG|
|Data service system||<http://www.sciencedb.cn/dataSet/handle/549>|
|Sources of funding||National Basic Research Program of China: Survey on the distributions of feedstuff resources for major livestock and poultry in China and their mineral contents (2014FY111000)|
|Dataset composition||This dataset consists of text data and image data of feedstuff samples. Text data record the geographical information, approximate nutrients and mineral contents of 4,000 feedstuff samples, with a data volume of about 0.20 GB, while image data contain 12,000 pictures of the feedstuff samples obtained during on-site sampling, with a data volume of about 1.2 GB.|
As the material base of animal husbandry development, feed takes up about 60% – 70% of the livestock and poultry breeding cost. Over the past 30 years, great achievements were made in China's feed industry. 200 million-ton feed were produced in 2015,1 of which large-scale feed enterprises were major producers whereas small-scale enterprises tended to wash out gradually. Among the feed products, compound feed took up an increasingly higher proportion, which amounted to 0.174 billion tones in 2015.1 Under rapid development, the feed industry is faced with new predicaments and problems, such as increasing shortage of feed resources, unreasonable distribution between feed enterprises and farming bases, abuse of mineral additives, serious livestock fecal pollution, and so on. These problems have seriously hindered a stable, healthy and sustainable development of the feed industry in China.
A country with unproportional population and land, China has severely insufficient energy feed and protein feed. Livestock breeding in China heavily relies on agricultural products and green forages. As the raw material of many feeds in China is unique, foreign feed databases are of limited referential value, and it is imperative to establish or improve Chinese feedstuff databases.2 Feedstuff database constitutes the basis of feed formulation and manufacturing. It results from the development and innovation of feed technology, and in turn serves as a tool of its development. Any feed database should serve its targeted object and strive to be complete, updated and accurate.3 After years of studies since China's reform and opening up, Chinese Feed Composition and Nutritional Value was published in 1983 and the Chinese Feed Database was built in 1989, laying basis for livestock and poultry feed formulation in China.4–6
Mineral additive is an important component of animal and poultry feed. Although accounting for a small proportion, minerals play an irreplaceable role in retaining the health of livestock and poultry and in improving breeding efficiency. In the Chinese Feed Ingredient and Nutrition Table published hitherto, most of the mineral contents are described by the mean value of static attributes, which is of poor pertinence and hence cannot reflect the actual ingredients of feeds manufactured across regions. In the 1980s, a survey was conducted nationwide jointly by several units including the Institute of Animal Sciences of the Chinese Academy of Agricultural Sciences and Huazhong Agricultural University, which generated a systematic description of selenium content in feed and forage grass.7–9 The content and distribution of other minerals remain unknown. Indeed, as feed producers may use different feed sources and/or have varied production levels, the use of mineral additives according to mean attribute values can hardly help achieve a desired balance; instead, it may cause nutritional dysfunction, mineral antagonism, mineral deficiency, and even intoxication.
While bringing about abundant animal and poultry products, the expansion of the animal and poultry breeding industry has also caused a large number of toxic or harmful substances from dung, urine, waste and waste water, such as ammonia, hydrogen sulfide, nitrogen, phosphorus and heavy metals, antibiotics, etc. When improperly handled, these substances may not only deteriorate the living environment of livestock and human alike, but also subjugate the livestock to diseases and cause a decline in animal product quality. The key of pollution treatment lies in source control, through which the use of mineral additives can be optimized. For lack of reliable data on the mineral content and distribution of feed resources in China, the design of mineral additives in current livestock feed formulas is sometimes blind. A common and convenient practice is to add mineral additives according to a unified nutrition standard without considering mineral content in sources, which is not only a waste of resources but also carries heavy metals to the environment.10–15 In fact, the Ministry of Agriculture is stipulating a proper range (including maximum level) of mineral contents in order to regulate the use of additives in feed production, which entails a consideration of mineral contents in feed sources.
In view of the situation above, this study surveyed the production, utilization, and distribution of over 98% major feedstuff resources in China and their mineral contents, covering 15 energy feedstuffs, 7 major plant protein feedstuffs, 6 major animal protein feedstuffs, 8 green forages and 4 major mineral feedstuffs. The dataset is expected to provide scientific guidance for reasonable use of mineral additives in feedstuff formulations.
The data acquisition process can be divided into three major steps: on-site feed sampling and sample preprocessing, analysis of major compositions and mineral contents, data processing and storage.
2.1 Sample collection and preprocessing
The study selects 40 animal and poultry feedstuffs, including 15 energy feeds, 7 major plant-protein feeds, 6 major animal-protein feeds, 8 green forages and 4 major mineral feeds, accounting for more than 98% of the animal and poultry feed in China. Among them, the 15 energy feeds are: corn grain, corn gluten meal, corn germ meal, corn DDGS, wheat grain, wheat middling, wheat bran, wheat DDGS, paddy, broken rice, rice bran, barley, cassava, soaked corn husk and defatted rice bran; the seven plant-protein feeds are: expanded soybean, soybean meal, cottonseed meal, rapeseed meal, peanut meal, linseed meal and sunflower meal; the six major animal-protein feeds are: fish meal, meat meal, hydrolyzed feather meal, intestinal membrane protein meal, plasma protein meal and blood-cell protein meal; the eight green forages are leymus chinensis, rye grass, silage corn, corn straw, wheat straw, rice straw, alfalfa and sweet potato vine; the four mineral feeds are: limestone, calcium phosphate, bone meal and shell meal.
Based on the feed production and utilization status across different regions, a feed-sampling scheme is formulated. A total of 4,000 samples of 40 feeds are collected from China’s 31 provinces (including municipalities or autonomous regions), coupled with 5,000 sets of printed barcode.
Sample collection conforms to Technical Specifications for the Collection, Description and Preservation of Feedstuff Samples formulated by the project team. Major steps and highlights of the sample collection include: feed confirmation and inspection (especially its manufacturing place), where feedstuff representativeness is considered before sampling; on-site sample information filling (on the sampling bag), registration form filling and sample barcode pasting during sampling; collection of 1 kg air-dried samples for each forage or straw feed, 2 kg air-dried samples for the other feeds, during which grain seeds are collected for grain feedstuffs; post-sampling information uploading using mobile sampling software, including geographical region, sampling location and sample images, after which the registration form is sealed and put into the sampling bag.
Pretreatment of the feed samples: the amount of samples is decreased by quartering; about 500 g air-dried feed samples are ground into fine powder, which passes through a 0.45-mm screen (40-mesh sieve) to achieve relatively homogeneous sizes for subsequent analysis of moisture, crude protein, crude fat, and crude ash, and passes through a 1-mm screen (18-mesh sieve) for crude fiber analysis. After the ground samples are well mixed, no less than 200 g samples for each feed type selected by quartering are bottled and sealed, with an information label pasted onto the bottle recording sample name, number, sampling location and date. They are then stored for subsequent analyses based on their province of manufacturing. Analysis samples are preserved at room temperature while reserved samples are preserved at -20 ℃.
2.2 Analysis of proximate nutrient composition
The project team also formulated Technical Specifications for Analysis of Proximate Nutrient Composition in Feed Samples, which provides methods for measuring the content of moisture, crude protein, crude fat, crude fiber, crude ash and nitrogen-free extract in feed samples. But proximate composition need not be determined on limestone, calcium phosphate, bone meal and shell meal.
Analysis method: according to national standard methods in China, including GB/T 6435-2014, GB/T 6432-1994, GB/T 6433-2006, GB/T 6434-2006 and GB/T 6438-2007, the content of moisture, crude protein, crude fat, crude fiber, and crude ash for each feed type is determined. Nitrogen-free extract is measured using differential method, which results from 1 minus the mass fractions of moisture, crude protein, crude fat, crude fiber and crude ash, as expressed below:
nitrogen-free extract (%) = 1 − moisture(%) − crude protein (%) − crude fat (%) − crude fiber (%) − crude ash (%)
2.3 Analysis of mineral content
Likewise, the project team developed Technical Specifications for Analysis of Mineral Element Composition in Feed Samples, which provides methods for determining the content of calcium, phosphorus, sodium, iron, selenium, lead, arsenic, chromium, and cadmium in feed samples.
Analysis method: according to the national standard methods in China, including GB/T 20195-2006, GB/T 6435-2014, GB/T 6437-2002, GB/T 13885-2003, GB/T 13079-2006, GB/T 13080-2004, GB/T 13088-2006, GB/T 13883-2008, GB/T 13082-1991, samples are prepared for determining the content of moisture, calcium, phosphorus, sodium, iron, selenium, lead, arsenic, chromium, and cadmium.
The dataset consists of text data and image data, of which the text data record the geographic location, proximate composition and mineral element content of the feed samples. Figure 1 shows the detailed information of a feed sample.
3.1 Collection location data
The collection location data cover the name and barcode of each feed sample, the address, longitude and latitude of sampling site, upload address, image name, time uploaded, etc. A minimum of one sampling entry is recorded for each sample. Image name is in the format of “[barcode].jpg”, which does not contain the file path. When there are more than one image file for a sample, the image is named "[barcode]_[number starting from 1].jpg", which gives easy management of the images. Images of all the feed samples are kept in one file folder.
3.2 Proximate composition data
Proximate composition data cover the content of water, crude protein, crude fat, crude fiber, crude ash and nitrogen-free extract. For each feed sample, there are at least two parallel analysis records and one measurement result, expressed in mass fraction (%) accurate to 0.1%.
3.3 Mineral content data
Mineral content data cover the content of calcium, phosphorus, sodium, iron, selenium, lead, arsenic, chromium, and so forth. For each feed sample, there are at least two parallel analysis records and one measurement result, expressed in mass fraction (%) accurate to 0.1%.
3.4 Image data
Image data are images of the feed samples taken by mobile phone during sampling, which reflect feed sampling location, feed resource and relevant information. The images are stored online in JPEG format.
In the process of sample collection, sampling methods were selected according to the characteristics and storage status of the feeds. For example, no less than 5 sampling points were selected for each bulk feed (e.g., heaped feed), which covered both the surface and the interior so as to enable a maximal coverage of the feed. On the feed production line, feed products were manually or mechanically sampled from a certain section at a certain time interval according to the flow speed. Samples of hay or straw stacks were usually obtained by hand from more than 5 points at different parts of the stack. The sampling points of the upper and lower layers should be 30 – 50 cm from its surface or bottom, and at least one bundle should be extracted at each point. 15 – 20 strains were collected from each opened bundle from outside to inside, which should be complete plants without much leaf shed in order to keep the stem-leaf ratio of the feed. Hay or straw was cut into small sections and put into the plastic sampling bag.
Data consistency and accuracy were checked during analyses of the proximate composition and mineral content of the feed samples. If an operator performed two parallel measurements within a short time period on the same sample in the same laboratory, using the same approach and equipment, the difference of the two results (absolute difference and relative deviation) should meet our measurement requirements. Blank samples and substances were checked by the same method for each batch. Standards referred included National Standards for Crude Protein of Whole Wheat Flour GBW(E)100126, National Standards for Crude Fat of Whole Wheat Flour GBW(E)100127, National Standards for Crude Fiber of Whole Wheat Flour GBW(E)100128, Test results of the substances should fall in the normal ranges as prescribed by the above-mentioned standards. When the parallel measurements proved consistent results and repeatable data, an arithmetic mean of the two measurements was taken as the final result, expressed in the mass fraction (%) accurate to 0.1%. The results can be expressed on either air-dried or dried condition.
In the process of data collation and processing, we checked the source of the samples, the test standards adopted, and the consistency of the two parallel measurement results. Problematic data that did not meet our consistency and accuracy requirements were removed.
The 40 feedstuffs included in this dataset are collected from different villages and towns of 31 provinces, municipalities or autonomous regions in China. The feedstuffs can be classified according to their type or origin for further analysis of their proximate composition, major mineral elements and their content.
For certain feed samples, feedstuff varieties and corresponding processing technologies are also recorded, which can be used for relevant data analysis.
As the research work draws to an end, data are provided in batches for online sharing depending on our work progress. Sampling location data are shared online through Animal Science Data Sub-Center and China Feed Database.
Chinese Feed Information Network. Total feed production exceeded 200 million tons in 2015 in China, available at: <http://www.feedtrade.com.cn/sbm/forecast/2087273.html>.
Wang K. Strengthening the construction of feed database in China. Chinese Animal Husbandry and Veterinary News, June 12, 2005.
Li D. The construction of feed database is the basis of improving feed utilization. Veterinary Orientation 10 (2015): 8 – 9.
Zhang Z. China feed database for 20 years. in Advances in animal nutrition research— The 8th national representative & the 10th symposium proceedings, Animal nutrition branch of China, Chinese association of animal science and veterinary medicine, Hangzhou, 2008: 292 – 2994.
Lin C & Miao Z. Basic completion of Chinese feed database. Chinese Journal of Animal Nutrition 1 (1989): 69 – 70.
Zhao Z. Investigation on the distribution of selenium content in trace elements in China. Journal of Inner Mongolia Agricultural University,03 (1986): 131.
Liu H, Qi D, Zhang J et al. Investigation of the content of trace element selenium in poultry feed in Hubei province, in The 10th Academic Symposium Proceedings, Animal nutrition branch of China, Chinese Association of Animal Science and Veterinary Medicine, Hangzhou, 2008.
Yang Y & Zhang G. A Survey Report on Feed Resources in Shandong Province. Feed Research Institute, Shandong Academy of Agricultural Sciences, 1989: 567.
Li X, Lin G & Huang G. Substitution of feed protein materials in 2014. Feed China (2015): 32 – 33.
Wang J, Liu B, Ma T et al. The design of dynamic feed database for feed enterprises. Feed Industry 35 (2014): 1 – 5.
Ru Y. Significance and method of establishing interactive feed database. China Feed 8 (2006): 18 – 21.
Ru Y, Hughes R, Choct M et al. Variation in nutritive value of commercial broiler diets. Asian Australasian Journal of Animal Science 16 (2003): 830 – 836.
Selby E. The effects of processing and grain particle size of wheat based diets on the performance of liquid fed weaned pigs between 7 and 15 kg live weight. Master’s Dissertation, The University of New England, 1999.
1. Luo Q & Liao X. A dataset of major feedstuffs in China and their mineral contents. Science Data Bank. DOI: 10.11922/sciencedb.549
How to cite this article
Luo Q & Liao X. A dataset of major feedstuffs in China and their mineral contents. China Scientific Data 3 (2018). DOI: 10.11922/csdata.2017.0016.zh