Yesterday I changed and cleaned all filters in processing loop 1L. This resulted in a big gulp water change of about 8 gallons. Half a liter of NeoMag was put into the 25 micron section. I will be watching both GH and KH very closely. The small-pore biofilter media (Matrix, Siporax, and MarinePure) was removed and rinsed out thoroughly in a hose stream. Just to be strictly correct about it, the efficiency is considered compromised until the biofilms repopulate. I will do the same in loop 1R this week. I have 20 gallons of undisturbed Coralife Bioballs in line, so no worries.
This week's trimming is overdue - i.e. last week's trimming is still not yet done - and the shade is depressing the O2 saturation cycle. Yesterday it peaked at 8.4 ppm. In itself, nothing concerning here, but it must be put right ASAP as the biofilters will need the boost.
Yesterday I cleaned and changed all filters in processing loop 1R, and high speed loops 2 and 3, with the accompanying large big gulp water change. All the small-pore biofilter media was removed and rinsed. 250 mL each of SeaChem Purigen and Continuum Catalytic Carbon were put into the 25 micron cartridge in loop 1R.
The program setting for CO2PRIMARY has allowed the running average pH to settle to 6.82. I am adjusting this just slightly to get back to 6.84.
The running average ORP has been fairly stable at > 530 mV since 1 March when a large charge of Purigen + Catalytic Carbon was installed. Reviewing the redox results over the last six months, ORP rose to > 400 mV after implementation of the streaming water change regimen, and it rose again to > 500 mV when I started using good quality pelleted activated carbon. Adding the high-grade media is a further improvement in organics scavenging efficiency.
Phosphate continues to persist without supplementation. Between plant consumption and water changes it remains within oligotrophy, seldom exceeding 1 ppm. Nitrate and iron values have been good with present plant utilization. Today I added KNO3 and micros to compensate the large water changes. After filter maintenance, I have been adding 20 ml of SeaChem Nitrogen and 10 ml of Flourish Comprehensive.
Almost certainly, the NO3 incline is due, in part, to the post-water change KNO3 boosts. However, as evidenced by the incline in PO4, which is not supplemented, recent liberal feedings of both fresh frozen and sinking green account for a considerable fraction. For now, I am reducing the daily KNO3 dose from 70 seconds (11.6 ml) to 40 seconds (6.6 ml). I will certainly have to make further adjustments, but will avoid getting on the low side of 2 ppm again.
I am not so enamored of the Hanna photometric nitrate tester as before. It's been difficult keeping tabs on NO3 and I have become suspicious that the variability has more to do with anomalous behavior in the instrument than natural fluctuations in the actual concentration. It is not a question of calibration; it is simply not consistent. Results are not repeatable, not even closely. I have been getting a lot of "check sample" error messages despite very careful adherence to instructions; wasteful of time and reagents and not very confidence inspiring. After running the same sample multiple times and getting widely disparate results, and also comparing to the simple API test as a reality check, I cannot rely on the numbers the Hanna tester is reporting. It has left me arbitrarily choosing which numbers I believe; not what is wanted in any test.
The API test gives repeatable results and is consistent. It is not precise in the sense that the photometric method is, but its accuracy is adequate for the purpose. I can easily see gradations ranging according to the following descriptors, and this will be the reporting protocol henceforth.
<< 5 ppm (significantly less than 5, but greater than 0)
< 5 ppm (close to but discernibly less than 5)
> 5 ppm (close to but discernibly greater than 5)
>> 5 ppm (significantly greater than 5)
< 10 ppm (close to but discernibly less than 10)
The Hanna photometric Checkers used for KH, iron and phosphate are precise and accurate. I like these a lot and continue having faith in them.
It's early days yet, but, tentatively, the amount of NeoMag, the rate of flow through the reactor, the dose rate of MgSO4 supplement, and the streaming water change regime seem to be finding balance. The Ca : Mg ratio does not comport with my earlier bias as to ideal proportions, but both numbers are high enough for aquarium plants so I will not be obsessing over that. The GH remains in the decidedly "soft water" category.
Not done yet regarding nitrate testing. The possible problems with the way I have been using the photometric tester do lie in the sample prep. When I take a 250 ml (1 cup) sample from the aquarium, almost always just after the morning dosing, I strain it through a fish net to remove debris. First problem, photometric testers are going to be sensitive to suspended matter. The sample should be passed through filter paper; a net or ordinary strainer is not adequate. It would be best practice to filter the sample this way for any test. It is easy enough to contrive an apparatus for pouring the sample through a coffee filter. This was done for today's sample. There were no "check sample" errors today.
Second problem, I have recently noted (16 March entry) the sudden dip in redox that occurs every morning upon dosing. This time of day is when the chemical oxygen demand (COD) is high and redox is actually unstable. The presence of strongly reducing and oxidizing substances, and also of protein hydrolysates, will interfere with the Hanna nitrate test (explicitly stated in the directions), and is probably not an ideal condition in which to conduct any chemical test. The dosing occurs between 07:00 and 07:20, just after the last AM open drain event, six hours before the first PM ODE. As a rule, the sample is taken between 07:30 and 08:00. Tests are run between 08:00 and 10:00, as was the case today.
I ran the Hanna photometric test three times and the API reagent test three times. The results:
Hanna: 1st test 7.6 ppm, 2nd test 6.4 ppm, 3rd test 7.0 ppm.
API: 1st test 5 ppm, 2nd test << 5 ppm, 3rd test << 5 ppm.
What to make of this? Is each test result within a realistic margin of error? The photometric tests are implying that NO3 is 7 ppm, + or - 0.6. That is certainly plausible. The data are such that I can draw no other conclusion; I could get a higher level of confidence if I took an average of more tests. The reagent tests are more problematic. I asserted that the API test is consistent, but, on this occasion at least, that is not the case. If the photometric test is anything close to real, the API test is low by 2 ppm, at best. I could put this down to two different kinds of tests with differing accuracy tolerances having an expected, but not trivial, differential. But the two follow on API tests, showing significantly lower values still, puts this interpretation in doubt. There is just something really wrong here. I will admit for the record that I believe that the NO3 concentration in this aquarium at the stated time of day is closer to 7 ppm than << 5 ppm.
For several reasons, I fully expect that NO3 will be different at 15:00 or 16:00 this PM.
At 15:50 a sample was taken, filtered, and allowed to stand for one hour. The water is well along in recovering stable redox. Testing commenced at 16:50. The jungle has been cranking full tilt throughout the day. The plants have been making proteins, perhaps to the point where NO3 has been fully consumed. There have also been four open drain events before sampling, so a small amount was lost in the water changes. Yesterday and the day before, fresh frozen was fed generously. A moderate quantity of flake and some sinking green was fed at around 09:00 today, and nothing since.
Hanna photometric: 5.2 ppm.
API reagent: 1st test 0 ppm, 2nd test 0 ppm.
Fluval reagent: 0 ppm.
The Fluval test confirms the API test. Both the API and Fluval tests are capable of reliably indicating over 1 ppm, and neither confirm the Hanna test. This strongly suggests that the NO3 << 5 ppm this morning may be closer to the truth after all.
The NO3 numbers being reported all this time are somewhat suspect as to absolute accuracy. There is some work to be done here. I don't think that any errors were so severe as to actually conceal excessively high nutrient. The observation that NO3 does not appear to persist in the environment is borne out by all the tests. And the plants do not exhibit nitrogen deficiency. However, in the interests of obtaining and presenting accurate data I mean to settle on a standardized procedure and a set of acceptance criteria that can be regarded as reliable protocol.
Nitrate test interference can be caused by chloride, ferric iron, copper, ammonia, urea, amines, protein hydrolysates (amino acids, peptides), high reduction activity (which is an indicator of the presence of these), and, in the case especially of photometric testing, fine particulate matter in suspension. All testing should be expected to have such problems; any test must be assumed to present with error possibilities requiring appropriate diligence to avoid. Biologically active water such as natural habitats and aquariums are problematic with respect to testing for dissolved inorganics, because sources of interference are unpredictable, but certain to be present. The test manufacturer's directions must be strictly adhered to, and the advice as to sampling heeded.
1) Take the sample at the time when the aquarium chemistry is at its most stable and interference factors are at minimum. All supplement dosing occurs at morning twilight, just after the last open drain event of the AM streaming water change, 07:00. The next open drain event is the first of the PM streaming water change 6 hours later. This raises the supplement concentrations to maximum levels as the lights come on and affords the plants full nutrient uptake opportunity as photosynthesis gets underway. Thus, at 07:30, just after completion of the dosing session, the system is highest in all interference factors. COD increases suddenly and ORP plummets. The aquarium does not fully recover its normal redox condition for about five hours. This is NOT the time to take the sample. However, much to my chagrin, it is the time I have been taking the daily sample.
2) This aquarium is crystal clear in appearance. There is always a 25 micron filter in at least one of the processing loops. This does NOT guaranteeadequate clarity for photometric testing of a randomsample from the water column. I have been frustrated by frequent "check sample" errors from the NO3 tester. Since I began passing the sample through filter paper I have gotten no further errors; the instrument produces a number.
3) The API nitrate test requires vigorous shaking of reagent bottle #2. The instructions specifically instruct 30 seconds before dispensing to the test tube. I had a discussion with an API tech about inconsistent results with this popular test. He emphasized the importance of this shaking. But I don't think you can shake enough. Before even starting the test, shake #2 for 30 seconds, then shake #1 for at least 5 seconds (not specifically directed). Then begin the test, shake the test tube after adding #1. Shake #2 rapidly with real vigor for a minimum of 30 seconds before dispensing. Shake the test tube as directed for the full 60 seconds. I found that the extra shaking before starting the test was needed for consistency of results. This is especially important on first use of a new kit, or if the kit has been idle on the shelf. I think I am being a little overoptimistic about my ability to parse results under 10 ppm . Also, I have a hard time telling any result over 10 ppm until the sample approaches 40 ppm; I can't tell the difference between the colors in this range on the color comparison card! It's a hobby grade kit and I don't expect more from it, which is why I need to find confidence with the photometric test instrument. My use for this kit is to verify the ballpark where the photometer is reading, to validate trust in its accuracy.
4) Scrupulous sample purity and cleanliness are important. All test tubes, test vials, and cuvettes are rinsed thoroughly with tap water after use, including caps. Immediately before use they are rinsed with aquarium water by fully submersing them. Photometer cuvettes must be dry and cleaned to optical instrument standard. I use Hanna's alcohol-based cleaner and a microfiber cloth. It is folly to attempt shortcuts. Sample contamination is a real problem, especially with phosphate testing. Water spots and body oil film on photometer cuvettes will cause errors, even if not obvious.
This aquarium's chemistry is maximally stable (flat redox) between 22:00 and 07:00. By 07:00 plant uptake for a full photoperiod, and two series of streaming water changes, will have cleared interference factors to whatever minimums are achievable before the next day's supplement dosing. The sample is taken at this time, filtered, and let stand before conducting the test. I believe the NO3 testing debacle is resolved due to these measures.
I did a 10% water change on Thursday, 15 April; then another 10% on Friday, 16 April. Nitrate tested at lower concentrations accordingly; the numbers are moving in a logical direction. The water changes of course impacted hardness parameters which are now a little skimpy as a result. Still out of bounds in terms of the target trophic status, the NO3 must be reduced further. The NeoMag reactor will catch up.
The water changes put a good sized dent in the K2CO3, which I boosted yesterday, and again today (after testing). I was interested to see the pH data-log curves "go bouncy" since Friday's water change. The running average pH is holding to specification, but an increase in outlier data points suggests a slight instability, probably due to decline in buffer. I expect the curves to normalize when we get back 2 degrees more of KH.
With time the GH will rise; ++Ca and ++Mg are okay for now and will only go up unless there is another big gulp water change soon, which is not contemplated. Before, as noted, the NeoMag reactor ran away with the GH (and added considerable KH as well) after a couple of weeks in the filter loop. I used only half as much this time in hopes of a gentler reaction.
This 7-day plot of ORP and dissolved oxygen (DO), illustrates the normal diurnal cycling as it occurs in this aquarium. The marker at Apr 13, 23:45 is shown for reference. Note the severe notch in ORP corresponding to a small similar notch in DO beginning at 07:00 and bottoming at 07:30. This is the time when all dosing events occur. A larger DO dropout is averted by aeration that fixes a 5 ppm O2 floor. See 16 March entry for details on this. Note that ORP does not fall below 470 mV, and DO does not fall below 5 ppm. The notch marks the moment when all the supplements enter the system.
Note the nonconformity of the DO curve beginning Friday Apr 16 at 11:00. This is when a full dose (200 ml) of Dr Tim's WasteAway chemo-heterotroph culture was added to the system. Organic matter began to be digested immediately with increased oxygen requirement. This is a graphical representation of biological oxygen demand (BOD). DO began recovery at 12:00. Beginning at about 02:00 Saturday, aeration was triggered to compensate the accumulated deficit. This nonconformity in the DO curve is typical each time this bacterial culture is added, and it is evidence that there is some labile organic matter always present in the system. The ORP exhibits a slight uptick after the bacterial treatment which reflects the reduction in dissolved organic matter (DOM).
Recent back-to-back big gulp water changes diluted parameters significantly.
Potassium and buffer still lean, but rising. I added 50 ml of K2CO3. The solution is 1 g/10 ml. These additions are mild in effect. I want the +K to be within 30 and 40 ppm and the KH between 7 and 8 degrees. With the current rate of daily water change 6.7 ml three times daily of K2CO3 solution has been doing that more or less uniformly for a long time now, although there was an occasional need to use a little SeaChem Alkaline Buffer to make a tweak. The objective is to maintain 30 ppm CO2 at pH 6.84, and +K at luxury uptake level, the exact concentration not being critical.
No change in GH yet. These figures for ++Ca and ++Mg as secondary nutrients are probably just fine for the plants, yet the habitat is a true softwater environment with a GH under 3 degrees. I would like the NeoMag reactor to maintain the GH at between 3 and 4 degrees in balance with the current rate of daily water changes. A slow build is preferable to runaway hardness, so I remain patient. I am still adding 14 ml of MgSO4 (1 g/ 10 ml solution) daily to support the desired Ca:Mg ratio. Since +K has been going in as K2CO3, no K2SO4 has been dosed for a very long time. Although --SO4 is a prominent anion in SeaChem Flourish Comprehensive and Trace, the greatest part of sulfate is going in as MgSO4.
I mentioned earlier the minor instability in the pH arising from the drop in KH occasioned by large water changes. Just as a point of interest, the following pH traces illustrate this effect. Both pH-L and pH-R exhibit increased outlier data points beginning after the water change on Friday, 16 April. The system running average held at circa 6.84 but there was clearly greater variability in the data. The second graph shows the return of the normal pattern in the traces as pH re-stabilized. The pH 7.33 spike at 10:30, 19 April, is pH-L detecting the --CO3 "rush" just as I was adding an extra dose of K2CO3. Similar smaller spikes at 07:00 each day in the pH-L trace show moments of buffer addition at the scheduled time. The doser outlet is at the left end of the tank, so these spikes will appear only in the pH-L trace.
This is merely a technical footnote. Were it not for data-logging, this phenomenon would have gone completely unnoticed. There was never danger in this case of parameters deteriorating into disaster, or anything of that sort. However, in terms of how managed aquatic systems behave it is a feature of interest to me. It illustrates what parameters we know to be specifically related by laws of chemistry look like as they evolve in real time in a real system.
The streaming water change regimen and the NeoMag reactor appear to be balancing on a stable GH. As the sole source of ++Ca, the NeoMag is working. At present the small daily supplementation of MgSO4 seems needed. Beyond that, I will not obsess about increasing the GH even if it remains at 2.5 degrees (44 ppm Ca/Mg). I will revisit this when these numbers change.
SeaChem Nitrite/Nitrate kit is showing approx 2 ppm NO3, and the API also confirms at << 5. Having corrected my lab technique, particularly the sampling schedule, my confidence in the Hanna photometric test has returned. So I will resume reporting NO3 from that test. I will continue to confirm NO3 readings with other tests.
Nitrate has continually decreased since supplementary KNO3 dosing was stopped over a month ago. All NO3 at this time is autochthonous (from environmental metabolism only). I have been feeding the fish generously with all types of food, including sinking. They do a great job of clearing it up, but it is a lot of nutrient. Between the streaming water changes and the jungle's nutritional requirements, the system is maintaining oligotrophy. I am encouraged once more that NO3 can be sustained without resort to supplementation.
The decline of +K and KH over the last few weeks is a signal that the K2CO3 dosing is not keeping up. Manually dosed boosts get spent fairly quickly. The auto-dose rate of K2CO3 is being increased from 60 sec (10 ml) to 90 sec (15 ml), three times daily.