The Coast and Mountain Walkers of NSW - Overnight walks to extraordinary places

On Water Treatment:
Finding It in Arid Areas and Devices for Collecting, Filtering and Purifying It

by Rob Jung

(Download this article as a pdf)

We usually plan trips around having a good supply of clean water on walks, although this is not always possible. Like many of you, I have encountered many situations in mountain, coastal and arid areas where obtaining good water was a difficult exercise. In the process I have learned some useful lessons and developed a few practical approaches. I will share some of these in this article. What I am about to say is far from comprehensive, and at least some of it is not new,but you are free to pick and choose what is useful to you. For the capable do-it-yourself person, I can supply patterns of my own constructions.


Finding water in arid areas

There are usually waterholes, somewhere. The best advice in locating them is to approach someone who knows the area well. This club has people familiar with different arid areas, so ask them. Topographic maps (e.g. in the Flinders Ranges) often mark waterholes and bores. Bear in mind that these may no longer be in good condition. This is particularly the case with bores, which fall into disrepair, and springs can dry up.

Certain geological structures can be suitable locations for collecting water or not. The rocky gaps in ranges in Central
Australia, which have significant catchments upstream, is one example of possible waterhole locations. In the Kimberley, creeks running through the ‘King Leopold Sandstone’ structure (a quartzite in the Kimberley Plateau), usually seem to express water and often abundantly. In the Flinders Ranges, in contrast to the Gammon Ranges where there are few waterholes, the granite streams dissecting the Mawson Plateau, NE of Arkaroola may have many waterholes. In some areas, such as in the Barrier Range (north of Broken Hill) and in the Fitzgerald River (SW WA), there is much salt in the landscape and when it rains the salts dissolve in the creeks and waterholes, making them unsuitable for drinking.


When planning a trip in an arid area, I use information from the Australian Government Bureau of Meteorology (BOM) site . I study historical rainfall records of a nearby location with rainfall records, to see what is normal. Then if I pick the right weather station, I can get daily rainfall records, to see recent patterns. I record this information on an Excel spreadsheet and graph the data to see what phase of the drought versus flood cycle the area is in, and how long ago the last significant rainfall was. An example of such a graph is shown in Figure 1., from my trips to Gundabooka.



Graph of rainfall patterns in Louth

Figure 1


. The rainfall data for Louth, was obtained from the BOM. Louth is 30kms west of Gundabooka, a National Park in arid north western NSW, so this graph indicates precipitation there.


My three trips to Gundabooka are denoted by the letters A, B (in 2005) and C (in 2010). The first (trip A) occurred at the end of a long dry period and we found little water and a parched landscape, that is until the last night, when there was a deluge. The next two trips followed abundant rainfall and the walking conditions then could not have been more different.


Refer to Rob Jung’s article 'Gundabooka' in The Bushwalker, Winter 2011, Vol 36 (3) pp12-14.



Checking the BOM rainfall data approach can also be very useful in evaluating other bushwalkers’ reports on the state of waterholes in past trips. They also need to tell you the date of their visit! In this case the historical rainfall data is viewed, and compared with the bushwalkers’ report on the condition of the waterhole at that time. There are complicating factors – rainfall is patchy. If there is a similar amount of rain for a number of stations in a region, you can be more confident that the rainfall was general.


Significant rainfall is when there is a fall of more than 25mm in a few days. It would be ideal if there were say at least 50mm for each of the past two months. Good recent rain has three benefits: supplies from ephemeral waterholes are more likely (“opportunistic” water), existing waterholes are more likely to be flushed, and water quality is likely to be better. To get favourable conditions you often have to be patient and wait.


An alternative is that you arrive with a mobile canteen (your car) to supply your water needs. For your walks, take water for two days - say 7 L in winter - and three days’ food. If water is found while out on the trip, stay out another day. If no water is found, return to the car on day two.


On the actual walk in an arid area, if by chance a good quality water source is located in an otherwise very dry area early in the day, it could be desirable to fill up for the night’s camp, even though that camp may be many hours away. The marked waterhole might not exist anymore (scenario in Figure 7.). The marked water site might be there, but there is no guarantee of its quality (Figure 2.). Opportunistically collecting water should be the overriding philosophy in these places, above the desire to travel light.


Photo of two waterholes in the Flinders Ranges

Figure 2


Flinders Ranges, May 2013.

We knew we would reach Yacki Waterhole (right) near our proposed camp and were certain of procuring water there (it is rated as a 95% waterhole). It was one hour into our day along Mainwater Creek and 5 hours to camp, when we opportunistically came across the large unmarked waterhole at left, GR 272385 [Photo: Sandra Mackiewicz]. We should have filled up here, as the water quality was excellent, unlike Yacki waterhole, which we later described as “Yukki”. Yacki was in a beatiful location, but the waterhole does not flush well.


The rainfall in the three previous months was good (for Arkaroola in February, March and April it was 35, 50 and 18mm respectively), which meant the likelihood of opportunistic water was good. On the previous day we passed up filling at another excellent opportunistic water source (GR 286 410), in preference to the inferior marked one at Mainwater Spring.

The etiquette around water when it is scarce

When water is very scarce as it often is in arid areas, it should be treated as precious and the last thing you should do is bathe in it! For this reason, I take a bucket and small sponge, which I can use to wash myself in a location well away from the water source. Using the sponge means that I can have an effective wash with the minimum amount of water. My bucket has a 5 L capacity, which means if there is suffcient water, I can immerse my head, wash my shirt in it etc. I also use my bucket as a supply reservoir in my water treatment processing. This is shown later in Figure 7.


Light collapsible 5-10 L buckets can be purchased commercially. However, typically they have a stability problem - when you are not looking they fall over, disgorging their contents. The old triangular prism shaped Paddy Pallin water bag works much better in this regard - it does not tip easily. My cylindrical bucket is more stable than the commercial ones. Its three C-fibre poles in the sides help to stabilize it.

Collecting and screening water

In many areas, water collects in rockholes on rocky ridges. This is usually ephemeral, lasting a few days to perhaps a week after rain. While a cup may be used to scoop out the water, these days I often use a large syringe (at least 60mL), with a wide nozzle, instead. These syringes may be obtained from medical suppliers. Although slower, syringing is much more efficient at cleaning out such pools and it is also much less likely to stir up debris in them. In the past I have used 8mm ID plastic PVC tubing for water collection, but now I have superseded that technique with the newer tools I carry.


I strain the water as I collect it, through a fine mesh nylon screen, to remove the floating material (Figures 3. and 4.). Other people use their cotton bandanas and handkerchiefs to do this. I have experimented with different hole sizes for the screen: 1mm (domestic sieve), 0.5mm (no-see-um mesh), 0.14mm and 0.07mm), and I have found the finest size to work best. It is often surprising what in seemingly clear water, is screened out (Figure 3.)


Photo of the contents of three mesh filters

Figure 3.


A. I always carry my fine 74µm mesh screen (15g), and it is surprising how often I use it. It removed this debris from 4 L of water from a clear looking creek on Bathurst Harbour, Tasmania.

B. It easily removed a thickish layer of algae from a small pool near Mt Zeil, NT.

C. Mark's silk screen easily filtered out this black residue which had settled in the bottom of this water tank connected to a hut, Northern Flinders Ranges, SA.

Why screen the water? Water purification techniques, whether they are the chemical treatment or UV treatment methods are compromised by particulate matter in the water, and are more effective if those solids are removed.

Photo of filtering water in the Yodellers Range

Figure 4.


In the Yodellers, water is usually very scarce. John is shown here filtering water from a large pool through an earlier version of my 74μm screen into a water container.

For a pool of this size, I would now use the cup and my screen-funnel, but I had not made that then.

With this wide and unconstrained screen, a syringe better directed the flow into the container.

In the latest version of my sieve (Figure 5), the mesh is incorporated into a very light nylon funnel, so that the water is directed into a wine cask container or plastic water bottle and there are no losses.

Photo of filter and funnel set-up in use

Figure 5.


A 74μm nylon screen disk is sewn into the top of a nylon funnel (12cm diameter). It effectively screens out debris, and connects to and directs flow into a winecask or other container, without losses. The yellow rim at the top of the funnel is made of whipper-snipper cord.

The nylon screen is easily cleaned at the end of the filtering process by rinsing off the debris on top with a small amount of the cleaned water. Fine nylon or polyester screen fabric is available in many sizes from Sefar, Blacktown, NSW (see Table 1). I think that a finer size could be used (perhaps 50 – 60 µm), but since finer pore sizes block more easily and have less % holes / area, I would not use a smaller screen size than that. Finer sizes are also more expensive. It is however expensive.


Not much mesh fabric is required to make a filter. If a number of people are interested one could form a syndicate, buy some and cut the fabric and split the cost between the participants.


Table 1. Nylon sieve fabric available from Sefar (Oct 2014)


 Size, µm 

 Catalogue No. 

 Roll width, 

Cost /m
 (GST extra) 


















Cost is / m, which is the minimum length cut or shipped. Delivery is $40. The first number in the catalogue number refers to the fabric (nylon), the second to the hole size and last number refers to the % area of holes in the fabric.For Sefar, it is best to phone (02 8822 1700), as their website has no detailed information about their fabric screens.


A cheaper alternative screen fabric is to use a fine fabric such as silk, from old inner sheet etc.

Collecting water from deep wells

Occasionally, being able to collect water from deep wells is very very useful. There are a number of these wells in the Flinders Ranges. A rope connected to a billy with a good stable handle is required for this task. If using a billy, one with a long narrow profile is the most stable for hauling up the load. A billy however is not ideal for this task as it is “tippy”, or conversely can be hard to fill. If constrained within a narrow bore casing (e.g. Geraldton Historical Society Bore, Gunbarrel Hwy), its rigid shape means that the billy cannot fill – it floats very well on top. I have made my own very light water collectors from siliconised nylon, which are specially designed for these situations. They have a weight bag on one side, which forces the collector to tip and thus fill when it reaches the water level. At the water line only the top part tips on the weighted side and not the whole collector, so that it can fill within a confined space. Figure 6 shows it being used to collect water from two wells in the Flinders Ranges.


Photos of author obtaining water from deep wells

Figure 6.

The author is shown having just collected 2.5 L of water from 4-5 m down two different wells in the northern Flinders Ranges. That was the only water available in these areas.

The water collector is very efficient. In B we had scooped out and distributed 40 L of water for the five of us within about 20 mins.



Filtering water using pump filters


I made many outback trips and obtained water from many places before I purchased a pump-type water filter. Once I did, I soon found that these devices are especially useful in areas of dubious water quality. Because using them is time consuming, I still only deploy them when necessary. My first preference is for sieving, then applying some kind of treatment. The treatment I use is boiling and / or adding chlorine-type tablets (Katadyn Micropur).


The pump filtration device I purchased was an MSR Sweetwater. In practice I have found it to work satisfactorily, although there are some extra accessories required and techniques to be understood to make it work well. My comments which follow, relate to this MSR unit.


The main filter cartridge does clog eventually, so before atrip where you may well use it, make sure you have a spare cartridge. I store the spare in the sealed bag until quired and after the trip after I have washed it.


The filter comes with two long lengths of tubing, a float, strainer and a bottle-type scrubbing brush. Don’t do what I first did and save weight by cutting the tubing. There will be awkward situations where having the supplied length is very useful


Also don’t leave the pump’s scrubbing brush behind (again to save weight). That is very false economy! The inner surface is where the filter residue accumulates and I always scrub this surface down soon after use, to remove the caked on solid deposit. If you don’t, the expensive filter will block up much quicker than it might, and consequently have a much shorter life. The scrubbing brush is also useful for cleaning the small wire strainer (intake screen) supplied with the kit. That frequently blocks if you don’t pre-strain the water.


If you don’t buy the next accessory(?) – the prefilter – you will find you will be using the scrubbing brush a lot when you filter murky water. In that case you may need to clean the filter every litre if you are lucky, or in more extreme cases every few hundred mL. That is very frustrating and slow. The Sweetwater pump expresses water by pumping in both directions (up and down). When the main filter is becoming blocked, the pressure required builds up and the restricted output, flows from the filter in only one pumping direction. The prefilter cartridge is a thick “depth” filter, which means it has a capacity for removing a high solid load. It is a coarser filter (with larger hole size) than the big filter, but it removes most of the solids within its bulk. The prefilter is called the “Sweetwater siltstopper” and the Australian MSR agents import them.


When filtering very dirty water, like the muddy puddles shown in Figure 7 you can see a dark band of discoloured particles gradually working into the prefilter. The manufacturers claim these can be restored by pumping through clean water in the reverse direction, however this is not a situation readily available on a trip, so it is best to regard the prefilter as disposable. On a trip, I take a new prefilter and a kit of three spares. The spare filters are called the “siltstopper replacement kit.” Filtering outback waters is often a slow process, and for a group using these tools regularly, there should be one filter unit for each 3-4 people.


Photo of using water filter system and of using alum

Figure 7.


On a Flinders Ranges trip (May 2014), there was no waterhole in the indicated location at Hannigan’s Gap (GR 746 721), but fortunately nearby there was an emergency source - this muddy puddle on a road (A). The top layer water was clearer than the water lower down, so the water was carefully decanted from the surface of the puddle into the 5 L bucket.Water was filtered from the bucket flowing first through the prefilter and then through the main filter. The filtered water was clear.3½ L of clear water was filtered, after which I discarded that prefilter.


In 2016 we returned, to find the same water situation (B). This time we used alum coagulation (C) to clear the water, which was much easier.


While filtration is effective, it greatly assists to carefully scoop the water to be filtered. Never stir up the water: what can happen is discussed in Figure 8. That is why a syringe may be the best means of transferring it. Stirring the water risks dispersing the solids, which can easily convert a pleasant task taking perhaps fifteen minutes, into a tedious and difficult one, taking more than an hour.


The MSR filter pump comes with a float, and for a single person working a waterhole this may be effective by allowing the intake to be well below the surface. However for larger group working around a small waterhole (e.g. Figure 8), working space is often restricted and carefully decanting the water into a bucket and filtering from that, away from the waterhole, is likely to be more practical.


Photo of obtaining water at Yacki Waterhole

Figure 8.


Obtaining water from Yacki waterhole, Gammon Ranges. This waterhole is poorly flushed and much used by wildlife, evident from the abundance of droppings surrounding it. Hence it was not surprising it was covered in a thick green layer of algae.


My fine screen is valuable in removing algal scum, but I didn’t have my 74µm screen then.


We cupped water into my bucket to filter it. From that, it took much more than an hour to filter 3 L, and getting that required frequent cleaning of the main filter (I had no prefilter). Night was approaching and I was getting concerned, so I decided to throw out the remaining feed water and try again. The others had departed and the waterhole had not been disturbed for a while. This time, very carefully, I filled the bucket by putting my cup about 15 cm below the surface and slowly drew it up. I was careful not to stir the water. The result was that we were able to filter another 3L in about fifteen minutes.



How much do the filtration devices weigh?


Table 2. Weight of the filtration devices discussed


MSR Sweetwater pump and filter

196 g

Connecting tubing

61 g

Cleaning brush

10 g

Metal screen intake filter

12 g

Siltstopper prefilter

28 g

Kit of 3 spare prefilters

27 g

All of the above MSR kit in a bag

366 g

Nylon 5 L bucket

53 g

60 mL syringe

28 g

74 µm screen-funnel

15 g

2.5 L Well water collector

15 g

Other ways of clearing water


A better way to clear muddy water is to use a coagulant such as Alum (aluminium sulfate). This procedure works well, is simple and is discussed in two other articles in this section.Epsom salts (magnesium sulfate) and gypsum (calcium sulfate) are not effective coagulants, despite some claims that they are.


An alternative method of clearing muddy water, though less effective and slower is by boiling it. If the water is boiled for 5 minutes and allowed to stand, the solids can settle to the bottom of the billy. The clear water can be decanted into another suitable container and the muddy residue discarded. The syringe and screen could be handy during that process. Sufficient fuel, time and storage containers resistant to hot water (not PET bottles), are all necessary for this method.


There are other ways of obtaining good water near the sea, such as distilling seawater or purifying seawater with a hand pumped reverse osmosis systems (RO). Hand operated RO systems are not very good and hard to find. For vehicle based travellers and boat owners, with 12v power supplies, good battery operated RO units are readily available.


(Download this article as a pdf)

Logo of The Coast and Mountain Walkers of NSW

Site Search

Copyright © 2018 The Coast and Mountain Walkers of NSW Inc

© 2015 CMW