Clean up water purification

[kevingranade]

For something that's pretty central to survival, water purification is quite haphazard.

Some cleanups we can do:

1. Differentiate between clear (but possibly contaminated), cloudy (obviously dirty), and tainted (from a very dangerous source) water.
2. Differentiate between biological and chemical contamination. (internally, this will not be player-visible without a test)
3. Carry contamination across recipes as called for.
4. Differentiate between sterilized (boiled/chemically treated, not that tasty) and filtered/distilled water (very clean).

To be more concrete, I'm proposing:
New Items:
cloudy water (obviously dirty, probably unsafe, from rivers)
toxic water (obviously dirty water from an obviously dangerous source)
sterilized water (flat or chemical-tasting water, biologically safe, output of boiling or chemical purification, lower morale boost than clean water)
water test kit (item used to detect chemical contamination in water)
New item attributes:
bio_contamination
chem_contamination
Code changes:
Enhancing recipe code to copy contamination from inputs to output
Enhance recipes to indicate whether they transfer contamination
Item changes:
Let iodine be used to sterilize (percent chance?) water.

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

[KA101]

If water from the same source (river/lake/etc) retains the same cleanliness, etc when I get more of it later, then yeah, this sounds interesting. Nontrivial chance for tedium though, I'm thinking?

[illi-kun]

This issue was closed as it appears inactive.

Reducing open issues to those which are (or will) be actively worked upon helps us focus our efforts. This issue has not been deleted - it still appears in searches and if it contains relevant information you are encouraged to continue to link to it.

If this issue was a bug

It should be reopened if it can be reproduced in the current build. You can obtain the most recent copy here. Please check there is not a more recent report of this bug before doing so. If no more recent report exists you should continue the discussion in this issue.

If this was a feature request

If the consensus was that the idea was good you could consider submitting an implementation via a PR. If you want to comment further please do so here as opposed to opening a new issue. Before posting check nobody has already made the same point and consider whether your comments are likely to lead to an implementation. If you have doubts about either consider instead voting for the issue

If you want to work on this issue

Then either assign it to yourself or if you are unable to do so claim it via adding a comment. Please don't assign others or make a general request for action.

[MrHrulgin]

Here's my take on water and its sources:

1. Terrain: Fresh shallow and deep water in flowing and still versions, shallow and deep salt water as well as pool water and sewage.
   a. Still water would be clear by default. Unless there's a current to keep it moving solid particles would drop to the bottom and even sewage would settle out within a few days of the sewer system ceasing to function. This would have no effect on its potability.
   b. Flowing water would be cloudy, but not inherently any more unsafe than still water in the region. How to model that variation in potability effectively is an open question, discussed below.
2. Water from pre-cataclysm municipal sources (currently hot water heaters and toilets). Clean and clear, though potentially dropping in safety over time if water contamination is modeled.
3. Water from wells, shallow and deep. Presumably water heaters and toilets in more remote houses and locations would also be from local wells. Of variable safety depending on local groundwater.
4. Water that is visibly off or definitely tainted. Pools of water in gutters with a sheen of oil, toilet tanks with brightly colored toilet bowl cleaner pucks, water from swimming pools etc. Possibly water flowing from pipes like you can see at homeless encampments. This water would not be made potable by boiling, as these contaminants wouldn't be neutralized by heat. Possibly a more advanced water filtration system would be able to deal with this stuff.

Questions:

1. How to test water safety? Many minerals or pollutants will change the water quality in ways that are easily noticed ( taste, color, odor and staining of fixtures). This can affect all of the types of terrain as well as wells, and should be very visible to the player once they're looking at it. Some of this water, while having alarming characteristics, is still safe to drink. (Or, in extremis, having deleterious health effects that are much longer-term than dehydration.) I haven't found anything online that shows how to make your own kits that test for potability, so they're single use and pretty irreplaceable?
2. How to model water safety? Water flows the entire length of the Mississippi in 90 days, so within a few months pollution from industrial sources would have gone into the ocean or terminal lake, but in that same timeframe you'd see new issues from lack of maintenance on those same industrial systems. Is it possible to affect groundwater quality based on distance from special locations? The groundwater will be worse next to a dairy farm than it will far away from human construction. Presumably there would be a groundwater safety quality and a flowing water safety quality, and the sum of the two is how dangerous the rivers are to drink? There's some work being done on lakes and rivers; possibly model groundwater boundaries based on that system? Depending on the complexity of the groundwater modeling, there could be significant amounts of useful trace minerals in the water.
3. Longer term water storage: should water in unsealed containers become contaminated? Water in toilet tanks and water heaters that might be good in the first few weeks may not be good indefinitely, and water that's boiled and poured into an open tin can on the counter will not stay good for long. This opens up possibilities for preserving water sources much like other foods are preserved. You shouldn't expect to boil 60L of water, leave it sitting by your stove and then come back in a month and find it fine.

I agree with the types of water laid out by Kevin. There's been recent upgrades to recipes to track calories from the base ingredients and will carry the (rotten) effect across; will that system work for water contamination? (See here) Will the types of contamination have different effects? Requiring different purification types? I don't know if there were as many options for water purification when Kevin proposed this. As it stands, boiling, the water purification tablets, charcoal water filter and the electronic water filter all would deal with biological contaminants, but only a distillation system like a still or a solar distiller would consistently separate the chemical contaminants from polluted water. Some chemicals have a lower boiling point than water, so even those wouldn't be removed IRL, but we might handwave that issue.

[kevingranade]

This issue has been mentioned on Cataclysm: Dark Days Ahead. There might be relevant details there:

https://discourse.cataclysmdda.org/t/why-are-swamps-in-this-game-filled-with-salt-water/25349/6

[I-am-Erk]

There's really no reason not to implement this now, it should be pretty trivial (to start on the framework, at least)

[I-am-Erk]

I was requested to add some implementation details here, so here are some general thoughts. Let me know here or on discord if you're interested and want more implementation talk.

• water could use the item fault system to track purity.
• there are three major contamination types for water that we could track. Two are fundamental, one is hard. The fundamental ones are "particulate" and "biological", the tricky one is "soluble".
• particulate contaminants make it hard to clear biological contaminants. Particulate can be removed through several forms of filtration. Filtered water is safer but still not safe.
• once particulates are filtered the biological contaminants can be removed in many ways such as boiling, UV exposure, chemical cleaners.
• soluble contaminants are generally rare and represent things like chemicals in the water, or salt, or other stuff that isn't removed by filtration and boiling. We have two options here, we could just call this water unsalvageable or we could make it possible to distill it to get pure water. I think allowing distillation to purify it ultimately is a pretty decent option but worth noting it requires a bit more knowledge than one might expect... You do need to make sure that you know what to keep and what to discard.

[nornagon]

Thinking through some more technical/implementation details.

• Currently the way to repair an item fault is to select (m)end on the item, which seems like a sort of strange verb for purifying water. The crafting system seems like a more natural fit here, perhaps?
• Breaking contamination down into particulate, biological and chemical makes sense. That's roughly how the EPA breaks it down, too (they add "radiological" as a fourth category). In theory, water with only particulate contamination would be safe (but unpleasant) to consume, but it's unlikely that a survivor will encounter particulate-contaminated water that isn't also biologically contaminated, so probably not worth modeling.
• Modeling increasing biological contamination over time for clean or sterilized water in unsealed containers seems reasonable to me. This is essentially similar to food going rotten.

flowchart LR
cloudy_water -->|filter| water -->|sterilize| sterilized_water -->|distill| clean_water
sterilized_water -->|time unsealed| water
clean_water -->|time unsealed| water
water -->|distill| clean_water

Loading

• Water in lakes, rivers and in puddles would become cloudy_water. Water in toilet tanks would become clean_water (degrading to water over the course of a few weeks). Water in water heaters and such is already clean water; additionally, those containers are SEALED so the water in them would stay clean indefinitely.
• I don't think it's reasonable for a survivor to be able to remove arbitrary soluble contaminants from water. Further, I think this case is rare enough to not worry about terribly much—there are many better sources of water, a survivor doesn't need to resort to complicated sewer-water distilling mechanisms. We have mutations if you want to roleplay a sewer rat.

I think it's reasonable to assume a survivor would have a pretty good idea of what water is how safe to drink. Drinking from a river, or puddle in the dirt? Probably not great. Drinking from a toilet tank? ehhh if it's been a few months it'll probably look pretty yucky. So I don't think it's worth trying to hide the particulate/biological contaminant status of water from the player behind a test kit—especially as such test kits are not reasonably craftable by a survivor, so they'd be a consumable resource. I could see it being interesting/fun to have certain chemical contaminants be undetectable without a test kit, e.g. radiological or other contaminants leached from a toxic waste dump, perhaps with the possibility of causing mutations if you drink too much?

[Edited to add a line from water --> clean_water via distillation without sterilization, as distillation also removes contaminants.]

[nornagon]

I've been looking through the current system for water contamination, wanted to note down what's currently there to capture existing thought. Whether or not we want to preserve any of this behavior, we should at least know what we're replacing :)

How does water become contaminated?

The definition for water doesn't include anything pertaining to contamination, so how does it become contaminated?

• water sometimes has poison in it, depending on where it spawns. This is set on the item at spawn time in map code.
• SPAWN_WITH_LIQUID + FRESH_WATER on a terrain spawns it with some water. This is only used for t_puddle and t_puddle_underground currently, which both also have the MURKY flag, which causes the water to have poison = rng(1,6), as well as parasites = 5 and contamination = { { disease_bad_food, 5 } }.

  Cataclysm-DDA/src/map.cpp

  Lines 2247 to 2257 in 9eaa5d4

  	if( new_t.has_flag( "SPAWN_WITH_LIQUID" ) ) {
  	if( new_t.has_flag( "FRESH_WATER" ) ) {
  	item water( "water", calendar::start_of_cataclysm );
  	// TODO: Move all numeric values to json
  	water.charges = rng( 40, 240 );
  	if( new_t.has_flag( ter_furn_flag::TFLAG_MURKY ) ) {
  	water.poison = rng( 1, 6 );
  	water.get_comestible()->parasites = 5;
  	water.get_comestible()->contamination = { { disease_bad_food, 5 } };
  	}
  	add_item( p, water );

• Water from water-source terrain ("examine_action": "water_source") with the MURKY flag gets the same treatment—poison = 1..6, parasites = 5, and 5% contamination with food poisoning.

  Cataclysm-DDA/src/map.cpp

  Lines 5308 to 5316 in 9eaa5d4

  	if( has_flag( ter_furn_flag::TFLAG_MURKY, p ) ) {
  	for( item &ret : get_map().i_at( p ) ) {
  	if( ret.made_of( phase_id::LIQUID ) ) {
  	ret.set_item_temperature( std::max( weather.get_temperature( p ),
  	temperatures::cold ) );
  	ret.poison = rng( 1, 6 );
  	ret.get_comestible()->parasites = 5;
  	ret.get_comestible()->contamination = { { disease_bad_food, 5 } };
  	return ret;

• Water from water-source terrain with the TOILET_WATER flag is clean 1/3 of the time, otherwise has poison = rng(1,3).

  Cataclysm-DDA/src/map.cpp

  Lines 5321 to 5327 in 9eaa5d4

  	if( has_flag( ter_furn_flag::TFLAG_TOILET_WATER, p ) ) {
  	for( item &ret : get_map().i_at( p ) ) {
  	if( ret.made_of( phase_id::LIQUID ) ) {
  	ret.set_item_temperature( std::max( weather.get_temperature( p ),
  	temperatures::cold ) );
  	ret.poison = one_in( 3 ) ? 0 : rng( 1, 3 );
  	return ret;

• Water from other water-source terrain will have a varying chance of poison from 1/3 to 1/20 depending on the DEEP_WATER and CURRENT flags.

  Cataclysm-DDA/src/map.cpp

  Lines 5345 to 5363 in 9eaa5d4

  	if( terrain_id->has_examine( iexamine::water_source ) ) {
  	int poison_chance = 0;
  	if( terrain_id.obj().has_flag( ter_furn_flag::TFLAG_DEEP_WATER ) ) {
  	if( terrain_id.obj().has_flag( ter_furn_flag::TFLAG_CURRENT ) ) {
  	poison_chance = 20;
  	} else {
  	poison_chance = 4;
  	}
  	} else {
  	if( terrain_id.obj().has_flag( ter_furn_flag::TFLAG_CURRENT ) ) {
  	poison_chance = 10;
  	} else {
  	poison_chance = 3;
  	}
  	}
  	if( one_in( poison_chance ) ) {
  	ret.poison = rng( 1, 4 );
  	}
  	return ret;

• Water from rain has a 1/10 chance of being poisoned per, uh, attempt? I think this chance might currently differ depending on whether the funnel is in the reality bubble or not.

  Cataclysm-DDA/src/weather.cpp

  Lines 246 to 250 in 9eaa5d4

  	if( contents.can_contain( ret ).success() ) {
  	// This is easy. Just add 1 charge of the rain liquid to the container.
  	// Funnels aren't always clean enough for water. // TODO: disinfectant squeegie->funnel
  	ret.poison = one_in( 10 ) ? 1 : 0;
  	put_in( ret, pocket_type::CONTAINER );

What are the effects of poison?

• Consuming water with poison > 0 will give you food poisoning for (poison * 30) mins, and if the poison is above rng(2,4), you'll be extra poisoned for (poison * 10) mins. Unless you're an Eater Of The Dead.

  Cataclysm-DDA/src/consumption.cpp

  Lines 1099 to 1107 in 9eaa5d4

  	// If it's poisonous... poison us.
  	// TODO: Move this to a flag
  	if( food.poison > 0 &&
  	!you.has_trait( trait_EATDEAD ) ) {
  	if( food.poison >= rng( 2, 4 ) ) {
  	you.add_effect( effect_poison, food.poison * 10_minutes );
  	}
  	you.add_effect( effect_foodpoison, food.poison * 30_minutes );

• Crafting anything with poisoned water removes the poison. This is a bug; only recipes which heat the water should remove poison.
• Combining water with differing poison levels (e.g. by Unloading) will result in the poison level of one being overwritten by the other, depending on which one is being poured in. This is a bug.
• Finally, NPCs won't accept poisoned food from the player, so this can be used as a water test kit. Also seems like a bug :)

  Cataclysm-DDA/src/npc.cpp

  Line 3233 in 9eaa5d4

  if( it.get_comestible_fun() < 0 || it.poison > 0 ) {