4.0 Re-examination of the function and performance of oxidation ponds treating dairy shed wastewater

4.1 Anaerobic Ponds

It was widely reported by Regional Councils' officers interviewed for this report that premature "sludging up" and excessive surface "crusting" were common problems encountered with anaerobic ponds. Although no data could be presented to support it, there is a common perception by Council officers that ponds in this state have diminished function resulting in a poorer quality effluent carrying over to the second pond, which being then overloaded also produces a poorer quality effluent,

Increase in herd size beyond what a pond-system was designed for was identified as being linked to faster sludging-up and crusting. But in other instances. early sludging-up and/or crusting are found to occur in spite of correct sizing of the pond to herd size.

An officer for the Northland Regional Council felt that excessive crusting was clearly related to grazing of low food-value pasture.

Theoretically higher loading tales should result in faster sludging-up and crusting. One situation where this may be expected to occur is where the design loading rate is too high with respect to the prevailing temperature in the pond - the presumption being that slower digestion of solids at lower temperature leads to faster solids accumulation. In this regard, the approach taken in the AWM for adjusting design loading rates to temperature is to divide NZ into three, latitude-defined regions that are meant to broadly reflect significant temperature differences in ponds. This approach has been questioned as being in need of refinement or replacement.

A second cause of overloading arises from the use of the average values for DSW production given in Table 2.2 of the AWM rather than actual values, which may be higher, for calculating load and sizing for ponds. The range around these average values is broad. Differences in average time spent by cows on the milking shed holding pad will be a significant source of variation in the per cow waste load from dairy sheds. How a farmer handles his cows also is a contributing factor to the amount of manure left on the holding pad. Nitrogen load in DSW is influenced by variation in pasture nitrogen content due to seasonal influences and nitrogen fertiliser application.

There can exist major differences between the average values For dairy shed waste parameters in Table 2.2 of the AWM - values which are generally used as the basis of two-pond system sizing and performance expectations -and values For dairy shed waste production calculated from Table 2.3 in the AWM. This is illustrated in Table 2 of this report, where organic load (BOD5) as calculated from Table 2.3 is 53% greater than in Table 2.2. (See also Table I of this report.) A more spectacular discrepancy between Table 2.2 and Table 2.3 is in the nitrogen load (Table 2). Another calculation method applied to five Southland farms For determining N and P load in DSW (Table 3) supports the use of Table 2.3 for load calculations as opposed to using Table 2.2 average values. Values calculated from the American Society of Agricultural Engineers standard reference for farm waste production (ASAE. 1991) are also included in Table 2; these values too support the use of Table 2.3 values together with information on time spent on the holding pad For calculating waste load as probably being a more accurate approach to load estimation than using Table 2.2 averages.

Table 2

Loading of several parameters in diary shed effluent based on NZAEI Agricultural Waste Manual Table 2.2 and calculated from Table 2.3 of the Manual and from the ASAE Standard D384.1 (1991). Values are in g and calculated for a 500kg cow spending a total of 180 mm. per day at the milking shed.


From NZAEI Table 2.2 Calculated from NZAEI Table 2.3 Calculated from
ASAE D384.1

COD 330 537 688
BOD5 80 122.5 100
Total Nitrogen 10.4 30.0 28.1
Total Phosphorus 1.76 3.12 5.90
Volatile Solids 250 400 625
Total Solids 360 550 750

Table 3

Nitrogen and phosphorous in dairy shed waste from five Southland farms (calculated from data on herbage, animal and milk production, and herd management from Prue Williams) compared to values from Table 2.2 the AWM. Southland data are on the basis of a 450 kg cow spending 3 hours per day in the milking shed.


Nitrogen





Total N in shed effluent
(g cow -1 day--1)

N concentration in shed effluent (gm-3)


Southland Published Values Southland AWM Values

Average 49.3 10.4 704 208

Range 37.3 - 74..9 6.8 - 19.0 532 - 1070 100 - 325







Phosphorus





Total P in shed effluent
(g cow-1 day-1) 		P concentration in shed effluent (gm'3)


Southland Published Values Southland AWM Values

Average 5.9 1.76 84.7 35.2

Range 5.1 - 7.0 1.0 - 2.0 73.4 - 100.6 10 - ?

Although excessive build up of sludge and crust to the point of pond dysfunction can result in poor effluent quality from the pond, this need not occur, and when it does it cannot be said to be caused by overloading. The cause is a failure to recognise the need for more frequent desludging of the pond when faced with an accumulation rate of sludge and crust that is faster than anticipated.

However, poor effluent quality can be linked directly to overloading if the strength of the influent DSW is above average. The percentage removal does not vary much within the range of loading to which DSW anaerobic ponds are subject. Therefore if a higher strength DSW is the result of more waste being carried in a normal volume waste stream, then not only will the concentration of BOD in the effluent be higher, but the quantity of BOD leaving the pond will be greater, giving a greater BOD loading to the "aerobic" pond. If, however, higher strength DSW is a result of a normal waste load but less water being used in flushing, then although the concentration of BOD in the pond effluent will be higher, the quantity of BOD leaving the pond (and entering the aerobic pond) will not be higher.

Similar arguments apply to anaerobic pond effluent that is of less than average BOD concentration. Lower than average waste load carried in an average amount of wastewater and an average wasteload in an above-average amount of wastewater will both give low effluent BOD concentrations, but the former will give a higher BOD loading to the aerobic pond than the latter.

systems

4.2 Aerobic Ponds

In a natural-aeration facultative pond that is correctly loaded to achieve 80 to 95 percent reduction in BOD5 in the effluent, there should be sufficient oxygen input from algae during daylight to keep the top 30 to 60 cm of the pond aerobic around midday. If the aerobic zone is shallower than this during daylight, i.e., the anaerobic zone extends upward to within 30cm of the pond surface, the pond is overloaded and BOD removal efficiency is reduced. Overloading results from either the actual load being higher than the pond was designed for or the oxygen supply by natural aeration being less than anticipated.

To assess whether a facultative pond is overloaded, one can measure the dissolved oxygen profile in the water column. This was done during the middle of the clay for nine of the 20 two-pond systems visited throughout NZ for this investigation. A YSI DO meter was used. In six of the nine "aerobic" ponds, zero dissolved oxygen was measured at the shallowest depth below surface that measurements could be practically taken, i.e., 1-2 cm below the surface. Three of the nine ponds had saturated or super-saturated levels of dissolved oxygen near the surface (c. 1-2cm depth), diminishing to zero dissolved oxygen at 10-12cm depth below surface.

We suggest that the DO profiles measured in this investigation are probably typical of DSW "aerobic" ponds. This being the case, the "aerobic" pond of DSW two-pond systems can be generally regarded as overloaded in terms of being a natural-aeration facultative pond. Indeed, ponds with zero dissolved oxygen at 1-2cm depth at midday should be regarded as anaerobic ponds.

The conclusion that DSW "aerobic" ponds are generally overloaded shows in the low BOD removal efficiency. Although there is not much data available where BOD5 has been recorded for the effluent from both ponds in two-pond systems, combined data from several sources show an average of 47% reduction in BOD5 concentration in passage through the "aerobic" pond for 22 ponds out of a data set of 30, with a range of 6% to 88% reduction (unpublished data). Even allowing for elevation in BOD, concentration as a result of evaporation (say 20% loss of water over the pond), average BOD5 removal efficiency for this set of ponds would still be only 59% - much less than the 80% or more removal indicated in the AWM. The remaining eight ponds in this data set showed a gain in BOD, through the second pond!

Pond overloading can arise in two ways:

1. the actual loading being higher than the design loading of 8.4 g BOD5/m2.day, and

2. the design loading rate being too high (i.e., natural aeration is less than presumed in setting the loading rate).

It is apparent from the discussion above on load calculations that the first form of overloading must be occurring in some systems. It would appear from the following discussion that lack of aeration is also a factor in overloading.

In regard to natural-aeration aerobic ponds, Eckenfeldler (1980) points out that with highly coloured or turbid wastewaters light penetration in the pond is minimal, and supply of oxygen by algal photosynthesis is curtailed. Inasmuch as estimated oxygen availability is the basis of the design loading rate of BOB, to facultative ponds, the implication is that with coloured and turbid wastewaters, the design BOB, loading rate should be reduced.

It is easily noticed that the colour and turbidity of water in the aerobic pond of DSW two-pond systems are generally very much greater than in sewage facultative ponds, and that this colour and turbidity do not appear to be caused by algae.

Chlorophyll a is the principle photosynthetic pigment of green plants (including green algae). When a plant is dead or moribund, chlorophyll a is degraded to phaeophytin a. The chlorophyll a content of pond water maybe used as a surrogate measure of the amount of live algae in the water. Effluent and near-surface water samples (in the case of ponds not discharging) from many of the ponds visited in this investigation were analysed for chlorophyll a and phaeophytin a. Pigment was extracted with 90% ethanol and concentrations of chlorophyll a and phaeophytin a were determined spectrophotometrically from absorbance at 665 nm measured before and after acidification of the extract, with corrections for turbidity as measured at 750 nm.

It should be noted first of all that very high levels of plant pigment were found in the effluent from anaerobic ponds (Table 4). The plant pigment was mostly phaeophytin a, with a small proportion of chlorophyll a. As anaerobic ponds do not support an algal population, it must be presumed that the pigment in the ponds is from the pigment-rich manure in the DSW, the manure coming of course from cows on a diet of green pasture.

Algae in the "aerobic" pond must compete for light with the manure-derived pigment entering the pond in high concentration in the anaerobic pond effluent. It appears from Table 4 that algae have varying success in this competition. At one extreme were ponds where the concentration of degraded pigment was still high, the proportion of undegraded pigment was low (high percent degradation), and virtually no algae were evident under the microscope. Zero DO was measured in the near-surface water of these ponds. Attenuation of light (photosynthetic Apparent Radiation) with depth underwater was measured with a Licor light meter and underwater PAR sensor in one of the zero DO aerobic" ponds. Extinction of PAR occurred at 4-5cm depth; i.e., the zone of potential photosynthetic production of oxygen extended only to this depth.

At the other extreme were ponds where streaking "clouds" of algae were observed just under the surface. (Visibility in these ponds and in "no algae" ponds typically extends only to about 4-7cm depth below surface.) Microscopic examination of near-surface water from three of these ponds (done by Mr Steve Moore of the Taranaki Regional Council) revealed abundant green algae mostly motile Euglenoid species. The percentage degradation of pigment in these ponds was much lower than in the "no algae" ponds. It was in these ponds that DO was present in the top 10 cm. The amount of degraded pigment (phaeophytin a) competitively absorbing light in samples from these ponds was not necessarily less than in samples from "no algae" ponds.

One pond that was observed on two consecutive days showed algal clouds and a dark green-brown colour one afternoon, but there were no algal clouds visible and the colour was a light brown the following morning. Samples were not taken on the first visit; samples of near-surface water taken on the second visit showed no evidence of algae (pond #3 in Table 4).

There is no evidence from the data (Table 4) that "aerobic" ponds with algae achieve better treatment in terms of net removal of BOD5, suspended solids, or ammonia than ponds without algae.

Table 4:

Ponds systems data for several two-pond systems in New Zealand, revealing relationships of treatment efficiency and pond function parameters (DO, pigment concentrations) on single occasions of sampling


BOD5
(g/m3) 		% Removal over 2nd Pond SS
(g/m3) 		NH4-N (g/m3) Chla
(mg/m3) 		Phaeoa Phaeo % off total pigment DO pH Visible Algae Pondsize
Waikato Ponds









1. Anaerobic 180
617 123 251 1,269 83
7.4 - correct

Aerobic

61 66 187 22 1,460 478 25 yes 7.9 yes 80% oversize
2. Anaerobic 140
7.9.4 206 334 972 74 . 7.4 - 25% oversize

Aerobic

210 -50 720 69 5,370 1,615 23 yes 8.1 yes 18% oversize
3. Anaerobic 140
391 188 251 2,584 91 - 6.8
34% undersize

Aerobic

58 59 105 63 100 343 78 no 7.7 no 30% oversize
Taranaki Ponds









1. Anaerobic 170
946 217 nil 3,196 100
7,6

Aerobic

94 45 399 48 1,754 1,339 43
8.0 yes 35% oversize
2. Anaerobic 120
655 130 nil 2,268 100
7.6 - correct

Aerobic

29 76 221 12 3,616 521 13 yes 8.2 yes correct
3. Anaerobic 130 711 251 nil 3,918 100
7.6 - correct

Aerobic

100 23 416 109 1,718 1,736 50 no 8.2 no correct
4 .Anaerobic 110
999 217 nil 3,608 100 - 7.7
correct

Aerobic

72 35 323 103 nil 1,031 100 no 7.9 no correct
5. Anaerobic 210
533 172 788 7,718 91 - 7.0 - correct

Aerobic

86 59 422 113 430 2,903 87 no 7.8 no correct

Table 4 Ctn'd:

Ponds systems data for several two-pond systems in New Zealand, revealing relationships of treatment efficiency and pond function parameters (Do, pigment concentrations) on single occasions of sampling

BOD5
(g/m3)		 % removal over 2nd Pond
 SS (g/m3) NH4-N (g/m3) Chla (mg/m3) Phaeoa (mg/m3) Phaeoa % off total pigment DO PH visible Algae Pond Size
Southland & Otago Ponds








1. Anaerobic 208 503 180 208 623 6969 92
8.0
30% undersize
Aerobic . 72 65 325 120 315 2997 90 no 8.0 no correct
2. Anaerobic 277
550 202 nil 8,660 100 - 7.8
correct
Aerobic 220
 21 540 231 nil 6,805 100 no 7.8 - correct
Northland Ponds









1. Aerobic 152
581 22.3 8,363 1,926 19 yes - yes 15% undersize
2. Aerobic 44
306 40 974 572 37 yes - yes 15% undersize I
2. Anaerobic 238
742 208 nil 644 100 -
 -
18% undersize
Anaerobic 1
140 37 599 146 nil 180 100 -
 - no 25% undersize
* Anaerobic 2
 125 16,48 518 85 nil 155 100 - -
 yes 45% undersize

Algae clouds were patchily distributed in the pond; sample was from outlet where algal clouds were absent.

Table 5

BOD5, suspended solids, and plant pigment concentrations in unsettled and in settled effluent (supernatant at 5 cm depth after 60 minutes settling) From anaerobic and aerobic ponds

BOD5
(g/m3)		 % Removal by settling Sus. Solids % Removal by settling Chla
(g/m3)		 Phaeoa
(g/m3)
Waikato Ponds
Farm A
Anaerobic Pond

Unsettled 180
617
251 1269

Settled 170 5.5 577 6.5

Aerobic







Unsettled 61
187
1460 478

Settled 39 36 128 32 372 195
Farm B






Anaerobic Pond






Unsettled 150
794
334 972

Settled 140 6.7 701 12 -
Aerobic Pond





   Settled     210

720

5370
1615

Unsettled
140
 33
 346
52
 845
 289
Farm C






Anaerobic Pond





Unsettled    140

391

251
 2584
Settled
 120
14
310
21


Aerobic Pond 1
      Unsettled
110

239

191
1115
Settled 110 225 5.9
Aerobic Pond 2 (duckweed cover)





Unsettled 58 105 100 343
     Settled 50 14  101 3.8

Table 5 Cont'd

BOD5, suspended solids, and plant pigment concentrations in unsettled and in settled effluent (supernatant at 5 cm depth after 60 minutes settling) from anaerobic and aerobic ponds


BOD
(g/m3)		 % Removal by settling .Susp. Solids % Removal by settling Chla
(g/m3)		 Phaeoa
(g/m3)
Southland Ponds




Farm A





Anaerobic Pond




Unsettled

182 406
nil 7011

Settled

 115 35


Aerobic





Unsettled







Settled





Farm B
Anaerobic Pond




Unsettled

277 550
 nil 8660

Settled

213 23


Aerobic Pond




Unsettled 

     220
540
nil
7011

Settled

 174
 21










5.0 Concepts regarding the function and performance of DSW two-pond systems, and how they may be improved

The analogy mentioned above between anaerobic ponds for manure wastewater and septic ranks for sewage is worth pursuing in regard to two points. It is known that septic tanks in warm climates perform better than in cold climates in regard to anaerobic digestion Paradoxically, warm climate septic tanks perform poorly in regard to achieving reduction of BOD and suspended solids concentrations in the effluent. This phenomenon is caused by the greater gas mixing resulting from greater anaerobic activity keeping more solids in suspension in the effluent. A corollary phenomenon is a slower rate of accumulation of sludge solids in warm-climate septic tanks.

As crust and sludge in a septic tank accumulate, the volume of the "clear-space" liquid zone between the crust and the sludge diminishes. The hydraulic residence time in a septic tank is essentially the clear-space volume divided by the effluent flow rate through the tank. There is a minimum hydraulic retention time needed to effect good settling of influent solids in a septic tank. If sludge and crust are allowed to accumulate (and the clear-space volume diminish) beyond the point of minimum effective HRT, effluent quality rapidly deteriorates.

It seems reasonable that the phenomena described in the above two paragraphs on septic tank function also operate in DS\V anaerobic ponds, which are functionally similar. Seen this way, the treatment BOD, removal efficiency of DSW anaerobic ponds will depend to an extent on how well they function to remove settleable solids from the effluent. In DSW anaerobic pond effluent, as with septic tank effluent, BOD in the effluent is closely associated with suspended solids in the effluent (Table 5).

For septic tanks (Laak, 1980) it has been shown that a length to width ratio of 3 or more gives better settled effluent than tanks of the same volume with a smaller length to width ratio. Also, a gas-deflecting baffle to divert gas-bubble entrained sludge away from the opening of the effluent pipe results in better-settled effluent. A final factor in septic rank efficiency is sizing, with higher HRT giving better-settled effluent. This factor is implemented to best effect where the additional tank volume is added as a second chamber after a gas-baffled first chamber.

As data on the strength and amount of waste coming from dairy sheds is rarely available to relate to effluent strength and volume from anaerobic ponds, there is no good indication of the removal efficiency achieved by anaerobic ponds. The fact that BOD, concentration in anaerobic pond effluent is generally less than 450 g/m3, and presuming average DSW strength is about 1500 g BOD/m3, then BOD, removal efficiency may well be generally greater than the 70% suggested for anaerobic ponds.

However, an additional 5% removal of BOD over the anaerobic pond would result in a 15% to 25% improvement in overall treatment efficiency over the two ponds without any change in the removal efficiency over the second pond. We suggest that modifications to DSW anaerobic ponds following the principles giving better-settled septic tank effluent arc worth trialling.

In regard to designing anaerobic ponds with an accurate, planned-for maintenance (de-sludging) interval, greater effort could be put into load estimates for individual dairy sheds. Information on sludge/crust accumulation rates in relation to fibre content of the cows' diet would also he useful information; such information is used in current American practice in sizing anaerobic lagoons for animal waste (ASAE, 1991).

It is appropriate not only to regard DSW anaerobic ponds as uncovered septic tanks, but to regard the "aerobic" pond of DSW two-pond systems as functionally analogous to the second chamber in a two-chamber septic tank system. In other words, maximum efficiency in the removal of settleable solids should be the function that is targeted and designed for in both ponds of two-pond systems. For the "aerobic" pond, adoption of this concept would suggest that maximum exposure to wind mixing and sunlight, as is currently recommended far facultative ponds, is inimical to performance efficiency.

The potential value of shading and improved settling for achieving better effluent quality from the "aerobic" pond can be deduced from Table 5. Where algae are present in the effluent, they can account for a significant proportion of the effluent BOD5. In ponds where algae are not significantly present in the effluent, much of the BOD5 is settleable.

An increase in BOD removal efficiency from 60% efficiency to 70% efficiency over the "aerobic" pond, if achievable, would result in a 25% improvement in effluent quality.

The suggested improvements in pond sizing and design suggested above, combined with diligence in the regulation and inspection of ponds systems to ensure correct installation and maintenance (ref. section 9.2 below), could result in a general improvement in effluent quality from two-pond systems as schematically illustrated in Figure 1.

Figure 1

Schematic representation of the benefits expected from improvements in the design and management of DSW treatment ponds. fig1.gif (9447 bytes)

Previous Page TOC Next Page

Contact for Enquiries

MAF Information Services
Pastoral House
25 The Terrace
PO Box 2526
Wellington, NEW ZEALAND

Fax: +64 4 894 0721
Contact this person

 
Print this page Print this page