LDPS2 models livestock herds using parameters concerning productivity. Therefore, herds with different productivity are necessary to be separated.
Cattle production in Swaziland can be divided by breeds into a) exotic dairy cattle and b) other cattle (traditional and beef specific cattle). They are assigned to "Dairy cattle" and "Beef cattle" in the labels sheet of LDPS2, respectively. Beef cattle production is divided thoroughly into three sub-systems by land tenure; 1) Swazi Nation Land (SNL, communally grazed cattle), 2) Title Deed Land (TDL, commercial production) and 3) Government Ranches (GR).
Regarding milk production, 36,700 tons of milk was produced indigenously in 1996, of which 10,300 tons was produced by dairy breeds. The remaining part (26,400 tons) was produced by traditional breeds which are communally grazed in SNL. Beef cattle in TDL and GR are raised for beef production and do not produce milk.
Total indigenous beef production in 1996 is reported as 15,226 tons (FAOSTAT). Production by each system (SNL, TDL and GR) can be estimated with the numbers of slaughter and sales (see Exercise 2).
Exercise 2: Rename production systems and set production demands |
1. Rename production systems [Labels sheet] Change names of production systems in blue. from Dairy System 1 [D5] to Dairy breeds ( System 2 [D6] beef SNL --------to check milk production) Beef System 1-3 [D9:11] SNL, TDL and GR 2. Set (projective) production demands and distribution loss [Parameters sheet]
System Production Distribution loss Dairy Dairy breeds 10,300 0 (beef SNL 26,400 0.1*) Beef SNL 6,873** 0 TDL 7,521** 0 GR 833** 0 *: For SNL beef cattle, milk distribution loss rate is assumed as 0.1 (10 percent). **: Beef production of each system is provisionally estimated using total production and the number of cattle sales (ref. Table 2) 3. After setting the values, save the current workbook for safety. Click "File_Save As"menu of Excel and save the file with a new name, "swazi.xls". |
There is a series of agriculture census data which contains cattle numbers by age, sex and region, and some parameters concerning productivity, such as the number of birth, death and slaughters (Tables 1-3).
| SNL | TDL | GR | Total | |
| Bulls (non-dairy) | 21,273 | 3,028 | 430 | 24,731 |
| Cows (non-dairy) | 168,822 | 35,509 | 6,203 | 210,534 |
| Dairy cows | 1,459 | 2,369 | 382 | 4,210 |
| Oxen | 86,626 | 9,580 | 1,996 | 98,202 |
| 2-3 year, male | 40,523 | 5,062 | 1,749 | 47,334 |
| 2-3 year, female | 55,185 | 10,293 | 2,049 | 67,527 |
| 1-2 year, male | 28,766 | 6,363 | 1,093 | 36,222 |
| 1-2 year, female | 29,041 | 7,292 | 1,192 | 37,525 |
| 1-2 year, Dairy | 575 | 668 | 96 | 1,339 |
| <1year, male | 29,320 | 7,540 | 1,215 | 38,075 |
| <1year, female | 28,480 | 8,008 | 1,469 | 37,957 |
| Total | 490,070 | 95,712 | 17,874 | 603,656 |
source: Agriculture census (1996)
| SNL | TDL | GR | Total | |
| Births | 72,043 | 18,393 | 2,834 | 93,270 |
| Deaths | 41,556 | 5,358 | 913 | 47,827 |
| Total slaughter | 27,431 | 4,359 | 732 | 32,522 |
| home slaughter | 20,422 | 1,471 | 175 | 22,068 |
| Sales | 29,936 | 32,759 | 3,628 | 66,323 |
source: Agriculture census (1996)
| SNL | TDL | GR | Total | |
| Birth/Cows | 42.7% | 51.8% | 45.7% | 44.3% |
| <1yr/Birth | 80.2% | 84.5% | 94.7% | 81.5% |
| Death/Total | 8.5% | 5.6% | 5.1% | 7.9% |
| Slaughter/Total | 5.6% | 4.6% | 4.1% | 5.4% |
| Sales/Total | 6.1% | 34.2% | 20.3% | 11.0% |
| Bull/Cow | 12.6% | 8.5% | 6.9% | 11.7% |
derived from Tables 1 and 2
NOTE: |
The total cattle number recorded in Swaziland is thought to be accurate as it is actually counted by government officials in every August when cattle are gathered at dip tanks for tick eradication. However, there are very large errors in the distribution to the various age classes. It is suggested that many animals are mis-classified into the next younger class resulting progressively too many in the 0-1, 1-2 and 2-3 year age groups and too few in the 3-year old and older age categories (Review of Swaziland livestock development strategies, 1992). Such errors will be corrected through the modelling with LDPS2. |
As mentioned before, provisional parameters are applied to LDPS2 first and are adjusted in a number of iterations, comparing the size and composition of the cattle herd calculated by LDPS2 and those in statistics.
(1) Production cycles (years in each class) and mortality rates
Newborns are normally kept for one year as young stocks, and then transferred to replacement stocks or slaughter (or draught) stocks.
| Dairy | Beef (SNL) | Beef (TDL/GR) | |||||||
| Y | M | Y | M | Y | M | ||||
| Young stock | 0-1 | 0.05 | 0-1 | 0.10 | 0-1 | 0.08/ 0.06 |
|||
| Replacement stock for breeder | 1-3 | 0.03 | 1-3 | 0.05 | 1-3 | 0.05/ 0.04 |
|||
| first mating | 3 | 3 | 3 | ||||||
| Breeding (reproduction) stock | 3-8 | 0.04 | 3-12 | 0.05 | 3-9 | 0.05 | |||
| slaughter of breeders | 8 | 12 | 9 | ||||||
| Slaughter (or draught) stock | (1-2)* | 0.03 | 1-9 | 0.07 | 1-4/5.5 | 0.05/ 0.04 |
|||
| slaughter of meat stock | (2)* | 9 | 4/5.5 | ||||||
| Males in the system ? | No | Yes | Yes | ||||||
| Are young males slaughtered at birth? | Yes | No | No | ||||||
Y: Years after birth
M: Mortality rates
*: Almost female youngs are retained for milk production.
(2) Parameters for milk production and reproduction.
The number of cows (Ncow) is calculated as follows, with 1) milk production demand (Pm), 2) Distribution loss (Ld), 3) fertility rate (Rf), 4) milk yield per cow (Y) and 5) fraction of milking cows (Fm);
Ncow = Pm ÷ (1-Ld) ÷ (Rf × Y × Fm)
For example, if Ld = 0.1 (10 percent), Rf = 0.8 (80 percent), Y = 0.5 (tons per cow per year) and Fm = 1 (all cows are milking), 90 tons of milk (Pm) can be produced by 250 cows (Ncow) in one year. LDPS2 does such calculation automatically.
The user needs to adjust these parameters, comparing the number of cows calculated by LDPS2 and that in statistics. When Rf and Y increase, Ncow decreases.
The number of calves (Ncalves) is calculated automatically at the same time. In LDPS2, Ncalves means the number of birth in one year. It is calculated as follows, with 1) Ncow, 2) Rf and 3) prolificacy rate (Rp);
Ncalves = Ncow × Rf × Rp
In addition to these parameters, production cycles and mortality rates also affect the size and composition of herds. The user needs to adjust all of these parameters, comparing the results of LDPS2 and statistics.
NOTES: |
Regarding dairy cattle in Swaziland, 4,210 cows produced 10,300 tons of milk in 1996 (Tables 1 and 2). Then, Rf and Y were estimated as 0.8 and 3.0, respectively. 10,300 tons ÷ (1-0) ÷ (3.0 tons × 0.8 × 1) = 4,292 cows The number of calves in the dairy system is calculated as 1,923 heads [Results!S44]. This includes all new born females and a part of males, as males which are not necessary to sustain the herd are slaughtered at birth [Parameters!C31 and 32]. We have decided to model traditional SNL cattle production with the beef calculation routine, however, parameters concerning milk production can be calculated with the same equation. The parameters are estimated as follows: 26,400 tons ÷ (1-0.1) ÷ (0.8 tons × 0.43 × 0.505) = 168,854 cows |
Exercise 3: (Calculation of Dairy cattle herds) |
Check the numbers of cows and youngs in the results sheet [R38] after changing parameters, such as fertility rate [C7], milk yield [C10], fraction of females milked [C11], distribution loss [C5], etc. |
(3) Parameters for meat production
A) Dairy cattle
Once the herd size and composition have been modeled, the number of slaughters is automatically calculated. Meat production is calculated as follows;
(the number of slaughtered animals) × (average carcass weight).
If calculated meat production is very different from that in statistics, the user is recommended to go back to the former steps and check parameters.
NOTE: |
Meat production by the dairy breeds and its offtake (slaughter) rate are calculated as 235 tons and 8.0 percent, respectively. They do not include veal production of newborn males. |
B) Beef cattle
Regarding beef cattle, beef productions of each system were roughly estimated with the number of cattle sales (see Exercise 1). They are, therefore, less reliable than the herd size shown in the census data and have to be checked with LDPS2.
NOTE: |
We can find that the size and composition of each system calculated by LDPS are larger than those in Table 1. According to the census data and FAOSTAT, 32,522 cattle were slaughtered in 1996, producing 15,223 tons of beef. Average carcass weight is calculated as 468 kilograms, however, it seems too heavy. On the other hand, offtake rate (slaughters / total cattle) in the statistics is 5.4 percent, while that calculated by LDPS is 8.4 percent. These mean that total beef production can be over-reported, while slaughter number can be under-reported in the statistics. Comparing cattle herds modeled by LDPS and those in the statistics, beef production excluding dairy cattle meat (235 tons) is estimated as 9,750 tons. System Beef (tons) Slaughter (heads) Offtake (%) SNL 6,380 31,700 6.3 TDL 2,900 12,200 12.6 GR 490 2,100 11.3 Sub total 9,750 45,900 7.4 |
Exercise 4: (Calculation of Beef cattle herds) |
Change beef production demand [Parameters!H:J5] to 6,380 (SNL), 2,900 (TDL) and 490 (GR) tons, and check the numbers of cattle in the results sheet, comparing those in Table 1. |
(4) Growth of the herd
The modeling has not been completed at this step, however, growth of the herds can be calculated by the Herd Growth Routine of LDPS2 using estimated parameters. Growth is an important factor to analyze condition of a herd. As LDPS2 is a technical model, it shows the largest growth which is technically possible. When the shown growth is too slow or too fast, the user is recommended to change parameters. Growth rate increases when years in breeders and fertility rate increase, or mortality rates decrease.
As the Herd Growth Routine figures livestock herd using a completely different model from that used in the Demand Driven Routine, results of these routines are different. For example, newborn males are not slaughtered and draught stocks are included in other stocks in the Herd Growth Routine. (see the user's guide at page 45)
Exercise 5: |
Go to the results sheet and try the Herd Growth Routine, selecting 1) production system, 2) routine and 3) years to display. |
NOTE: |
The Herd Growth Routine showed that growth rates of beef cattle herds are around zero percent p. a. The beef herds can not expand its size because of low reproduction rates and high mortality rates. In fact, the census data showed that the national cattle herd had decreased at 6.0 percent in 1996 while it had grown at 2.5 percent in 1995. Surprisingly, the herd had grown at more than 5 percent during the early 1990s. The rapid growth in the period seems strange and the cattle numbers in 1990 - 1992 could be over-reported. |
(5) Parameters for skins (or hairs)
Skin production is calculated with parameters on the proportion of usable skin and weight of skin per animal.
(6) Parameters for liveweight and milk fat content
These parameters are used for calculation of feed energy and protein requirements. Therefore, they do not change herd size.
(7) Parameters for draught
Draught power is considered an output of adult dairy cattle, beef cattle and buffaloes only. Draught animals are regarded as a "by-product" of cattle and buffalo production systems, where surplus calves (i.e. those calves not kept for herd growth) are partly diverted into draught use, and partly diverted into slaughter stock.
In the LDPS2, four kinds of animals are available for draught use, breeders (males and females), male replacements and draught specific oxen. The user can select animals used for draught with parameters No. 48-51. LDPS2 distribute total power demand to draught specific oxen, which come from other (slaughter) stock, at first. When the demand is not satisfied by the stock, the remaining demand is distributed to male breeders, male replacements and female breeders, in turn.
It is difficult to estimate total requirements for draught (or animal) power with a set of generalized coefficients, because there are many kinds of works, techniques and other factors which affect the requirements. Then, LDPS does not estimate the requirements, but the user determine it empirically. LDPS calculates number of draught animals with the following formula:
No. of draught animals
= (Peek power requirement per month) ÷ 30 days ÷ (Average productivity per animal per day)
A unit for the requirement and productivity is also defined by the user. Hectares/day, Man-day or Animal-day, for example, will be available.
Exercise 6: (Calculation of Beef cattle herds) |
Most of oxen in SNL are used for draught power supply for 2 - 3 months mainly in Spring. Input 87,000 heads ×30 days = 2,610,000 to "Peek animal draught power demand/month [Parameters!H51]", 1 to "Are there Draught specific oxen?[Parameters!H52]", 1 to "Average productivity/animal/day [Parameters!H59]", 60 days to number of days worked, respectively. And check the number of draught animals on the results sheet. |
Now, cattle production in 1996 has been modeled with the Demand-driven Routine of the LDPS2. Other species (sheep, goats, pigs and poultry) have already been calculated in the workbook.
1) Feed inventory
Feed energy supply from grazing lands is calculated by LDPS2 using the area size. In Swaziland, there are 10,810 square kilometers (sqkm = 100 hectares) of SNL grazing lands and 3,329 sqkm of TDL grazing lands. As their productivity is relatively low, 156,667 Livestock Standard Units (LSUsnote, see the user's guide at page 21) and 48,246 LSUs of feed energy are estimated to be supplied from SNL and TDL grazing lands, respectively (see Appendix C).
In addition to them, non-grazing resources provide 102,000 LSUs of feed energy. (Livestock Sub-sector Review and Range Survey, TCP/SWA/2353, 1994, also see Appendix D)
Exercise 7: (Calculation of Beef cattle herds) |
Change hectares (Grazing land), quantity (other resources) and energy contents in the resources sheet, and check LSUs calculated. Remind that quantity is in dry matter. |
2) Feed Utilization Matrix (FUM)
Those feed energy (in LSUs) is allocated to each production system in the "Feed Utilization Matrix (FUM)" of the resources sheet, using a series of decision rules:
Physiological criteria are straightforward in that they can be universally applied. Certain feed types are not suitable for certain animals (e.g. roughage for poultry) and so cannot be allocated to them, or only in small quantities.
Geographical criteria relate to the locational coincidence of feed availability and livestock populations. For example, SNL grazing lands supply 156,667 LSUs and they sustain all SNL cattle, 75 percent of sheep and 95 percent of goats.
Economic criteria relate to cost/price ratios of basic concentrates, or commercial feeds to the price of main livestock products. These ratios roughly indicate the profitability of feeding various feed types at given productivity levels.
Exercise 8: (Calculation of Beef cattle herds) |
Check and modify the FUM on the resources sheet. |
Note: **TO BE CHECKED BY OFFICERS IN VET. SERVICES, SWAZILAND** |
According to the FUM, grazing resources are not enough to sustain all the SNL livestock (cattle, sheep and goats). Soil erosion caused by overgrazing is reported especially in SNL grazing lands. Carrying capacity of grazing lands in Swaziland is calculated as only 205,000 LSUs which can supply only 84 percent of total beef cattle requirements. Scarcity of feed resources is one of major constraints for the livestock sector of the country to expand the production. On the other hand, there are crop residues and by-products which are not utilized as feed. For example, sugarcane tops are usually burned and not utilized for feed, while they could provide an estimated 170,000 tons of dry matter of winter maintenance ration for cattle. It is equivalent to 40,000 LSUs of energy or 16 percent of current energy requirements for beef cattle. Another possible resource is molasses of which 12,000 tons in fresh weight are used as a feed at present. There are plans to utilize a further 25,000 tons (18,800 tons in dry matter) per year in the local manufacture of a liquid product of molasses to be used by the livestock industry which will be of benefit to the local beef fattening and dairy industries. |
*****TO BE CHECKED BY OFFICERS IN VET. SERVICES, SWAZILAND**** |
I used an information on available feed resources reported in the Livestock Sub-sector Review and Range Survey, TCP/SWA/2353, 1994, by FAO (Dr. Sweet, Dr. Khumalo, Dr. Taylor and Dr. Xaba). According to the FUM, two thirds and a quarter of total energy are supplied by grazing lands and crop residues, respectively. And, about 70 percent of total energy required by SNL and TDL cattle are supplied by the grazing lands. It means, cattle are reared only by grazing or feed from grazing lands (such as hay and silage) during 70 percent of a year. (Regarding sheep and goats, I assumed that about 40 percent of energy are supplied by the grazing lands.) ****** Do you agree with this figure? If not, there may be other feed resources, I think. |
The user can save results as an Excel workbook with a new name (for example, Swazi.xls), using the "File_Save As" menu of Excel.
The simplest way to print results is by pressing the "Print results" button in the results sheet. The user also can print results using "File_Print" menu of Excel.
