Are we too focussed on monitoring?

[Kobus Jooste]

I have been spending a bit of time pondering monitoring in aquaponics again.  I think it has a lot to do with my aquatic ecology background – seasonal shifts in nutrient supply, nutrient sources and sinks, seasonal growth and die-backs.  Before embarking on the aquaponic journey I was well on my way to do a PhD on ecosystem energetics but the funding never panned out and I abandoned that idea. 

 

In aquaponics, things work a bit differently but it is not always all that easy to get your mind of details that you are used to expecting.  DO, ORP, BOD, COD, TAN, TOC…….and the list goes on when you try to balance the ecosystem energetics of an open aquatic system.  I see also on the automation group’s page that there are a lot of electronics junkies that seem to want to automate home systems to a great degree.  I think there is it also a case of curiosity meets ability.  As a further bit of detail, I will add that I have made this year my “Year of the plant”.  I have always had a better handle on the fish and water component, are having fun with designing components and all, but feel that I will only be truly happy with my abilities if I get sharper at knowing exactly what plants want.  This obviously lead to a bit of reading on the matter, and inevitably also looking at how these guys are treated in hydroponics.  I know enough about the differences between the two methods not to try and emulate hydroponic conditions, but still, having come from a fish side, I was considering all things.

 

Working with my research unit put me into finer water quality monitoring mode with ammonia, DO, nitrates, ORP, EC and TDS on the radar.  I am basically running fish and duckweed together with nothing else to deal with water quality except solids removal.  To have publishable results, and to track how the duckweed deals with trace elements added to the system, I need to have greater detail that for typical applications.

 

All of this got me wondering though.  Reading through terse exchanges on other forums reinforced the idea with me that there are two camps in monitoring theory:  The minimalists and the detail freaks.  I do not put people that are away from their systems often in any camp here – if you are not next to your system on a daily basis, or running a large commercial operation with a lot at stake, things are obviously different.  But for the rest of us, for the typical home system, what is eventually the difference between monitoring essentials and excessive sampling of factors that is of no consequence?  When a unit is hardly registering nutrients, what will parameters dealing with EC, TDS or ORP tell you?

 

If you keep the fish happy (give them food and have enough space and oxygen in the water), keep the bacteria happy (have a pH at which they can operate, enough oxygen and alkalinity to buffer the nitrification process) and keep the plants happy (nutrients, oxygen and a close pH match considering that bacteria comes first) will any of the parameters typical of finer aquaculture or hydroponic management matter? If you break this list of AP happy conditions down, you are referring to oxygen often, to pH often and then to basic nutrients linked to water quality.  The obvious choice is to look at Ammonia, because that is the basis for the nitrogen cycle and the thing that will kill fish the quickest if you do something wrong.

 

What is nice about the list above is that it is typically stable parameters in a well managed system.  pH is not 7.2 today and 6.1 tomorrow unless you have issues.  Without hardware failures, your DO will closely reflect the stocking densities and gravel surface area you chose to work with – design and management issues – again not concepts prone to sudden shifts.  Leaving dead stuff floating around or getting your feeding rates all wrong will show up in Ammonia readings, and it will show up quickly.  Then again, if you know your system well, you would probably “see” the issue from the appearance of the water, the behaviour of the fish or even the smell of the unit.  Are we insane to want to hook up a healthy ecosystem to life support?

 

Perhaps.  In defense of detail, it is possible that well-monitored systems may reveal some aspect of system performance that we could otherwise overlook or not take seriously at present.  Aquaponic practice is not totally developed and thus I think having a lot of detailed observations will be good.  That said, I do not think it should be considered vital at all.  As mentioned before, if you are away often or run a big commercial concern, the focus of monitoring is different, but for the rest of us, we perhaps need to accept the simplicity of what we are dealing with and settle into enjoying the stability we have created.

[Kobus Jooste]

Tanks Carey and Ellen for the feedback.  Perhaps if I look at the comments to this blog, I can re-focus the question as it is distilling: Are we too focussed on monitoring as an AP management tool in stead of good design principles?

 

I agree that running a research system, commercial system or for your own learning, having detailed monitoring can be important.  I also agree with the statement that WELL DESIGNED and PROPERLY MANAGED systems should be stable and give plenty warning of impending doom.  I see patterns in my research system that people with colour coded test kits may not even know about.  Every duckweed harvest or incremental food increase is visible in my readings, which is very satisfying (although not crucial for day to day management).

 

I think that I will follow up with some thoughts on design vs monitoring as well as treating monitoring results as indicators of trends or absolute gospel. 

 

Carey, on Automation I'm a follower, not a detractor.  As long as you set the system up correctly, having automation should not be a problem.  I see many benifits and are currently trying hard to secure funds for a complete sensing and control system for the research unit.  Let us take temperature and pH as examples.  having a control system capable of keeping these in the "ideal" band not only will give better yields, but in terms of making adjustments, having small changes regularly is better than trying to make larger adjustments every now and then.  Again, on the design and management side, there will be suggestions put forward for why automation may not be needed, but for me, it is not something that I strive to do without.  Here is an example: In my research unit, there is no nitrification.  I therefore worry about total ammonia levels related to temperature, pH and the level of uniodized ammonia.  If I had a control system, I could program it to automatically start lowering the pH (I try to keep it mid 7's) to a level that supresses uniodised ammonia levels (say 6.8) until I have been able to stabilise the unit. I can also have it send me a warning.  I'd rather work in the knowledge that I have this back-up than telling myself that I should not let ammonia rise above 2 mg/L.  A feeder can malfunction or a valve could fail cutting off the Lemna bed and then I have hours to fix it before I start loosing fish if it is summer.

[Carey Ma]

Personally I do believe in testing and monitoring especially for beginners and for research.

One thing I would like to add is weather or temperature changes may affect a system. It would be nice if we could separately log and combine our results to write an algorithm for automation.

 

Do you think automation detracts our products from being holistic? 

[Ellen Roelofs]

I think it just depends on why you're testing in the first place.  If you are testing to detect problems, once you learn to visually identify those problems there is no need for further testing. If you are testing to create a record whereby to evaluate long term changes, then you keep testing at regular intervals, but maybe consider reducing sample frequency as conditions stabilize and you get sick of recording the same results over and over again. :)

[Gina Cavaliero]

It is somewhat refreshing and relieving for me to hear Nate and Chris's opinions, as they are very similiar to mine and especialy after the other monitoring discussion we were having, Kobus!   I do understand and appreciate the need for monitoring on a very detailed level, but realistically I think that may be better reserved for the R & D folks, not the commercial folks where those couple hours of testing and recording can be better utilized elswhere.   Those of us that are intimately involved with our systems on a daily level are capable of quite advanced visual monitoring.  I don't know if I would call it lazy Chris, or just a level of confidence via competence.  Again, these systems are so stable, a significant outside introduction or event would have to occur for a 'crash' to happen. I also think the larger size of the system is a critical factor in maintaining a healthy, stable balance.  A small system is going to reflect issues quicker if there is something off.     

[Kobus Jooste]

I like the concensus on stability from all of the old hands.  It brings me to some other conclusions though in terms of the pressure I was under to try and please some people over here that wants to see aquaponics behave like "meals on wheels" - just spitting out produce at will.

 

1. AP is likely going to be stable in a low fish density scenario, thus people wanting a lot of fish yield will have to look at other methods, some of which may not be so stable, but which may well end up becoming rather expansive

2. AP is not so "wild and woolly" that it has to be described as the marraige of two technical disciplines (aquaculture and hydroponics) where a person need to become familiar with the technical competancies of both.  It becomes far easier. 

[Nate Storey]

The longer I do this, the less I care about monitoring.  For the most part, my fish and plants tell me more about the system than a test kit could anyway.  Plus, it's so stable/predictable that I really don't need to worry about crashes or anything. . .

[Kobus Jooste]

I must admit messing with rainwater systems, overstocked (on purpose) systems and set-ups with no biological filtration has made me a little jumpy.  I'm reverting my home system to LD now just to enjoy it again.

[RupertofOZ]

Likewise Chris... I only test when/if  I observe any change in my water quality... and in my case, with an unbuffered system.. it's almost always a pH issue... readily, and quickly adjusted...

[Chris Smith]

I must say that I have gotten a little lazy with my testing. I know my systems so well that I do not do much testing at all.  I look for signs of problems and will test as needed. My plants and fish usually will tell me when there are problems brewing. I occasionally test for pH but that rarely changes due to calcium carb buffering.These systems are so stable once they are up and running that I feel there is little need to test a lot. When I used to do hydro I was testing and balancing EVERY day. That got to be a big hassle. I now have 8 independent systems operating on my farm. It would get expensive testing all of then regularly.

[Terri Mikkola]