top of page
  • Writer's pictureby Arielle Illia

Building A Root Cellar

Living in a small cabin in Alaska, with a solar system and a generator as our only means of providing energy, we both knew food preservation was always going to pose a challenge for us. What follows is our story on how we built an underground earthbag root cellar and what knowledge we acquired through the process.

We gardened extensively in Oregon and being in zone 8 we were able to garden into the fall season and store some of our root crops outside during winter under a thick layer of straw, but for the most part we'd harvest the majority of our crops late summer for preserving or for storing inside. At one point we did discuss building a root cellar outside but at the time we had a computer room that was infrequently used on the north facing side of our house, that remained shaded year round and stayed relatively cool. We decided it would be more practical to attempt to turn this into a storage room rather than construct one outside. This room worked great for our needs, we blocked out the window to exclude light and kept the door closed to minimize air exchange with the remainder of the house. We kept squash, apples, potatoes, canned goods, and our dry goods in there on shelves we built. At first I made the mistake of attempting to keep the room's humidity high since traditionally root cellars are in the 70 to 90 percent humidity range, but quickly found that mold thrives in these conditions. Needless to say we ended up purchasing a dehumidifier and running it every few days to keep the humidity at 50 percent or lower, this kept the mold at bay.

When we first moved to Alaska we brought quite a bit of canned goods and crops with us from Oregon, we shipped potatoes, onions, and garlic due to Canada's border regulations and we stored all of this in the cabin our first winter here. At the time our only other storage option was our conex but that wouldn't work since we don't heat it. During our time living here we have learned that our cabin temperature ranges from 40 to 90 degrees Fahrenheit depending on how regularly we tend to the fire in the spring, fall, and winter months. We have no dark spaces available in our cabin with the exception of a few cabinets and the humidity stays near 15 to 20 percent when our wood stove is running. In addition to these factors that make our cabin less than desirable for food storage, there is also its size (320 sqft on the first floor and 160 sqft in the loft) that makes it difficult to store such volume of food.

We considered many options when we talked about what we could do next winter, we knew we needed a place to store several hundred jars and our garden veggies. A root cellar came to mind of course, however building an above ground insulated building without heating it was out of the question since items inside would freeze in our harsh winters. We did debate on putting a few chest freezers in the ground for the crops and storing our canned goods in the cabin but we both felt a root cellar would be a feasible option and provide a means for greater food storage, so we jointly decided we'd be taking our construction below grade.

Ideally root cellars can be constructed into a hillside but we didn't have that option here so we knew integrity was going to be crucial to hold up to the forces of the earth surrounding this structure. We also had an allowance on this build that we were comfortable with, the higher integrity materials were generally costlier such as building with concrete, rebar, and cinder blocks. Other ideas we researched on the more costly end were septic tank containers and culverts. Shipping containers were not an option when we learned they are not designed to go underground and generally need to be reinforced if you would like them to hold up over time. We ruled many options out during this process and kept researching. We found a great deal of information on root cellar buildings but not as much when it came to Alaska specifically. We were able to track down two resources that we based a lot of our design from, one person's cellar was in a similar climate to ours and the other family's root cellar was in Fairbanks. Both designs used a rectangle wood based structure with only the ceiling insulated and a hatch style entrance with no ventilation. The family in Fairbanks had constructed several out of wood and finally replaced theirs with a cinder block structure.

Wood, pressure treated or not, was a commonly used material for our area with the understanding that these structures generally have to be replaced over time. We both gravitated towards cinder blocks due to longevity however the cost was tremendously greater with the materials for this structure, we also gave consideration to the skill level required and time needed to construct such a building, which ultimately are the reasons we turned away from this design.

From here we defaulted to a wood design, that is until we stumbled upon earthbags, also known as sandbags or construction bags. We found these structures were common in other parts of the world and frequently built underground. Our research often led us to dry, warmer climates where they were built. Initially we were confident in their strength but we questioned their ability to handle wet climates. We knew the material which earthbags are made from, polypropylene, holds up well to water and did confirm them to be appropriate for humid environments, and/or places that receive moderate rainfall. Their one known weakness is UV degradation which would not be an issue for us underground. Although we were not able to track down many sources that have documented building one in a climate comparable to ours, we both knew this option fit the bill in regards to cost, practicality, time allotment and in terms of a long term structure. We later confirmed any doubts we had about lateral pressure and snow load from two sources, a family in Canada and another in the Pacific Northwest that successfully built below grade with earthbags. At that point we shifted gears abruptly and started planning for an underground earthbag root cellar.

One major factor Eric and I did overlook just a tad (okay not just a tad) was our water table. I will get to that soon but first let's talk about the shape of our construction, earthbags built in a circle are stronger than 90 degree corners, the same goes with arched roofs v flat roofs. I figured we'd touch on this subject since it was in our consideration early on, our only reasoning for straying away from these design elements was functionality and time, we also found many documented square or rectangle, flat roofed underground earthbag structures that have held up, which only reassured us it could be done. Rather than explaining why specifically we didn't do this or that, I will end at we considered it, found it not to work for us but if we were in the situation where it worked easily we would prefer it.

Once we did the math on how much space we would need in our root cellar, and allowing for growth, our design was a 8 foot x 8 foot structure with a 6 foot ceiling. Neither of us are that tall so we wanted to stick with the minimum depth we needed to dig and be underground. We roughly planned to use logs as the roof with a hatch opening. We went with this style because both the examples (previously mentioned) that we researched in Alaska had this type of entrance to minimize the amount of cold air coming into the cellar, however we were going to add ventilation that could be closed off in winter if need be.

What follows below is not a "How To", a tutorial or any form of instructions but rather simply a story of our experience that we wished to document and provide for folks that could possibly benefit from it, for either their own amusement or perhaps if they too are interested in constructing an underground cellar made from earthbags. Please proceed with caution if dark, damp, and amateur made underground structures scare you. However, if what you seek is honesty just keep scrolling.

I am semi kidding about that, Eric and I do feel incredibly safe in our root cellar but realize where it was not well thought out and will need to be improved.

The first step was digging the hole, by hand was not an option for us due to our time constraint and quite possibly our physical ability, although I can't be sure since we didn't ever have to test that. We spent a third of our budget on the excavator rental and previously purchased several hundred construction bags (14" x 26") at .40 cents per bag. We approximated around 500 to 600 bags for this project. We had the location picked out and were all set to go until our neighbor dropped off the excavator and clued us in on the water table. We quickly reworked the location for the root cellar and got started in hopes that we wouldn't hit water.

The first day of excavating took Eric a matter of a few hours before we had the hole almost completed, and by the next morning our hole was finished.

Upon first observation, we had great soil for this project, primarily compromised of silt and sand with a smaller amount of clay. When building with earthbags some people use a small portion of cement and then wet it to help stabilize the bags. We found this is generally not necessary if you have good soil.

Soil is a ratio of sand, silt, and clay, generally it is a combination of these three. Sand has the largest particle size, the best drainage and does not hold water or its shape very well when tamped. Clay, on the opposite spectrum, has poor drainage due to its small particle size and lack of space for air or water to circulate, therefore it often compacts. So you might ask what's the ideal ratio when filling earthbags with soil? Often earthbags are filled with the dirt removed from the excavation site, however it is crucial to not use only all sand or only all clay but rather a blend. You can face shifting with earthbags consisting largely of sand, and if they are composed of predominantly clay, they can expand with moisture and contract when dry, both scenarios can lead to unstable walls. In an ideal world the ratio would be 2 parts sand and silt to 1 part clay. Folks successfully use all different soil types for their projects and we were thankful our ground came close to the ideal make up.

Back to our excavation site, on the second day Eric was planning to excavate to about 10 feet or so, that way we had enough room for a 6 foot high ceiling with gravel laid, and with the addition of the logs as the roof and two feet of dirt on top. Just as I was about to tell Eric we were at that mark, there was a small wet spot forming at the bottom of the excavation site.

Eric and I proceeded to assess the situation and realized what this meant. We needed a change of plans, and ultimately decided to add back 3 feet of dirt and gravel in hopes that the water table wouldn't come up much higher. I did some research during our dilemma to see if we would be able to foresee how high the water table would rise and came across some interesting information. It turns out that you can get a good idea of the water table fluctuations just by looking a bit closer at your soil. Through my brief research I learned about the color of soils and iron deposits being factors that helped determine water table levels. If there was a specialist making the conclusion I am sure they would have learned a great deal about our soil but since it was just me I wasn't able to determine too much beyond that the most superficial soil had a gray hue and was sand like and the deeper soil was brown and clay like, there also was notable mottling from iron. With time pressing us due to other summer plans, family visiting soon and the excavator rental primarily we opted to just go for it. We came back up with two feet of excavated dirt and about 12 inches of gravel for the base of the cellar floor. Honestly we both had high hopes and figured we would take the gamble, worst case scenario we could pump out the water if that should occur. The decision to not come back up even higher relates back to our concern with the insulation of the structure and if too much of it was above grade we may possibly face freezing in the winter.

At this point you may be wondering about drainage, we did consider a drainage solution but did not feel it was necessary since we were adding a gravel floor and lining the structure. The thought of lining the floor was also a possibility but it wasn't in the plan nor is it traditional for root cellars to have a non earth floor. High humidity was what we were after at the time. Looking back I can't say I would have changed this, however I feel the shallow water table was, and is, the biggest issue with our root cellar build in our opinion.

Now with the site prepared the real work began. After shaping the hole with hand shovels and leveling it, we filled the hole with approximately one foot of gravel from large to small diameter and we marked our perimeter (also accounting for the bag size once tamped). Keeping straight lines is important for the integrity of the earthbag walls and it can be easy to get off track. After filling several dozen earthbags we laid our first course of sandbags. For the next week this was our routine; fill, tie, and tamp. To say the process is exhausting is an understatement, on top of this project we spent the remaining hours of these days fishing, gardening, and preserving, only to sleep for a matter of a few hours before starting all over again. Like I mentioned before, there was a push to get this done quickly so we kept at it but of course made sure to document it.

An important note when filling the bags is that you want to avoid large rocks that could contribute to air space and possibly shifting later on down the road, some folks filter their dirt, we manually picked out large clumps as we put it down the chute. The rocks sorted themselves our fairly easy so this task was not too hard with our soil type.

In between every layer of earthbags we added two layers of barb wire to prevent slipping, this is a crucial step I personally would not skip, there are others we saw that included rebar in their walls for reinforcement and buttressing the corners, and honestly from all the designs we came across not one of them was identical to another, so we knew people had constructed them in all different ways but with that being said, it does make sense to take the extra precautionary steps. For us, after tamping these bags we saw firsthand just how incredibly strong they are. The largest earthquake our root cellar has withstood was a 5.2 magnitude and I must say earthquakes this size are what I would call a "baby" earthquakes, we feel them all too frequently since moving to Alaska so our perception is skewed. And although fairly confident the earthbag shelter would hold up in the 7.1 we had last November, I can't say the same for the jars, but I genuinely do not think we will have one of those in the near future and if I could predict the future I most likely would not be here writing about our root cellar build. Onward!

Once we built up to around three feet we added the housewrap and lined the outside edge of the sandbags with a few inches of gravel before back filling with dirt and tamping it into place. We back filled at several stages throughout the project to prevent air pockets. Several folks used polyethylene plastic rather than housewrap for lining their structure which also seems to be a good option, keeping in mind most materials will break down given time. We repeated this process until we reached six feet from the base of the gravel floor to the top, approximately 21 courses. We did ensure to stagger the bags as we laid them, although not entirely apparent since the bags are not like bricks. When we reached a section where a whole bag didn't fit we adjusted accordingly, we tied a smaller earthbag (with the tie strings) and tucked them down and tamped them into the next bag. By the time we built this thing up, it felt very much like a brick wall in strength and I might add, stayed incredibly cool the entire duration of the process even in 80 degree Fahrenheit temperatures. That's the power of the earth I suppose.

Now we put in our shelves, wood was going to be our primary choice but without protection we both knew mold would form, plastic and metal were other options. Metal which most likely will rust ended up being the least costly option and we felt good about proceeding with that decision.

Let's take a quick moment to talk about why light, temperature, and humidity is so important when storing foods. Light degrades canned foods and will cause potatoes to sprout prematurely. Dark is key for long term storage. Temperature is also important since most vegetables and home canned foods cannot freeze, if crops freeze and unthaw they will quickly spoil and if mason jars freeze you risk breakage and seal failure. High temperature is also not ideal for long term storage and can lead to the quicker breakdown of canned goods and decrease the time vegetables will store. Lastly there is humidity, this is a tricky one since certain crops do better at different ranges. A traditional root cellar has a high humidity, in the 80 to 90 percent range, which is ideal for most crops but not all of them. Garlic, onions, and squash are an exception to this and prefer closer to 60 percent humidity, if stored at a higher level they can rot, on the contrary cabbage, carrots, beets, and potatoes prefer 80 to 90 percent humidity for optimal long term storage. That's quite a difference if you are attempting to store all your goods in one area like we did. High humidity will also cause rust on metal lids over time, this may not be a big deal if you consume them annually but too much rust can possibly contribute to a seal failure.

Finally it was time for the roof, in retrospect I could argue this was by far our biggest mistake in terms of longevity, primarily because we opted to not peel the logs we used. So I guess as one would anticipate, when the logs on the ceiling were exposed to the high humidity and condensation formed, the bark was quick to mold. They are in tact for now and we feel safe with them for this year and quite possibly longer before they will need to be replaced.

When constructing the top we notched the logs just slightly to help with lateral forces and laid them on top of the earthbags, allowing space for the hatch. Pictures work better here than my words.

With the hatch framed, we laid more earthbags on top to cover gaps and sharp edges of the logs so our plastic cover would not be punctured. We used 6 mil poly to cover the roof, we cut out an opening for the hatch and laid it over before we covered the roof with dirt. The hatch is constructed of 2 x 4 lumber with foam board inside and another layer of 2 x 4 wood that we added at a later date and treated with linseed oil. We opted to install one 2" pipe vent near the ceiling and later install another at ground level to help with air exchange. Ventilation for root cellars is important to prevent ethylene gases from building up. The lid is scrap metal roofing we had with leftover insulation made to fit snugly.

After we completed covering the whole structure with approximately 2 feet of dirt we knew it was time to fill it up. Over the course of three to four months we canned hundreds of jars and brought them down to the root cellar. The hatch routinely gets brought up as an issue by others however it is not an issue for Eric and I, we are both comfortable with it but cannot argue that walk in stairs would be more convenient.

In the month of August the root cellar started to cool down to 57 degrees Fahrenheit where it stayed for some time, the moisture was very high at the beginning then slightly decreased. We found as we were putting the jars down there just some of them within a two week period or so would get a small amount of mold on them, we fixed this problem by wiping all the jars with vinegar prior to storage and haven't had an issue since. I suspect traditional root cellars had varying mold levels as well. Through the late summer months and early fall our goal was to decrease the humidity in the cellar. We were able to get our humidity down to 70 percent with a fan drawing air from outside into the hatch opening, we knew that we wouldn't be able to continue drying it out once we dug up the potatoes and the winter set in.

During July and August we had several downpours but September and October was when the real rain came, and naturally the water table rose. With the rain came some erosion as well of the top soil, and we needed to intervene when the ground started freezing and unthawing. We lined the edges above ground with rock and that seemed to do the trick for the time being, ultimately winter came for good and the ground is frozen for the season. We had planned to do something more permanent but seeing that we were already having issues with the water table and moldy roof we figured we would be removing the dirt and redoing a part of this structure sooner than later.

Late September was the first time we saw wet gravel on the base of the cellar. We had a pump to deal with the situation but were a little distraught to see how quickly it rose afterwards, primarily because we had to run our generator during the pumping operation.

The weather is doing some freaky things so predicting it is not my business we just have to work with it. At first we spent a good amount of time running the generator to pump the root cellar often, later we learned that we could just let it fill with water and move items up to the higher shelves. The water never came higher than 20 inches or so, although it more commonly sat around 8 inches and would drain slowly. At the time of this post there is no water on the floor and we suspect we won't have an issue until spring thaw. In regards to condensation in the structure, it has improved then gotten worse when the water table rose, it does drip on our jar lids and creates the ideal environment for mold on the logs although the growth itself has slowed with the decrease in temperature for now. Rust has really not been an issue, even on the metal shelves, we both imagine given more time it could be.

The temperature in the root cellar has dropped extremely gradually over the course of a few months and we have not noted large fluctuations, if it's 28 degrees Fahrenheit or -10 it may change by 1 to 2 degrees. So far we are still hovering at the 33 degree mark, I know you're thinking that's awfully close to freezing but from all that I researched it is common for root cellars to stay literally just above freezing in climates such as ours. That's all you need them to do and I might add it makes for wonderful crop storage for the potatoes and carrots, they come out of there harder then when we pulled them out of the ground. On the downside, I overestimated the ability to be able to store garlic, squash, and onions in there, with the high humidity we ran into mold issues very early on when attempting to store squash. So for now we keep our onions and squash in the house, it's not ideal for long term storage but they have lasted longer in here than what they would have down there.

Although there are some major changes we need to consider we are still incredibly happy it worked as intended, we will have potato seed for this spring and I won't have to buy potatoes for planting. In the past most of our potato varieties made it to February but we typically had only a few left that there were never any remaining for planting by the time March rolled around. This root cellar has been perfect for the potatoes and my hope is they will go right into March and April for planting season.

There is something beautiful about this project that we built to sit there and think it really works, like actually really works amidst all of our other failures in regards to it. We joke about turning it into a swimming pool or filling it back up with dirt but we both know it has the potential to be something great. We are constantly thinking much like Winnie the Pooh's classic "think, think, think" sessions to come up with the perfect resolution, and although we have yet to find it, we have an idea where to start.

Food storage does continue to provide a challenge for us but as a result we have turned to learning traditional ways of storing and preserving meat and crops by fermenting, pickling, curing, and drying which in turn helps us conserve energy and minimize our plastic use. We both love our own little personal grocery store and are delighted to take a trip to the cellar to stock up for the week.

At the end of the day we are both proud of our decision to build a root cellar, perhaps it meant trying something that didn't work out but we cannot deny that there was a tremendous learning experience to be gained and ultimately we would have rather tried and failed than to have never tried at all. We firmly believe in trying new things and learning as we go. The root cellar was no exception to that and at the end of the day life is about the chances you take, there is unrenowned beauty in that I promise you.


Recent Posts

See All


Jan 13, 2022

OMG, LOVE YOU BOTH TO PIECES!!!! That's a good-lookin root cellar! 4-sure! Where exactly did you place the barbed wire? Never saw it in your Youtube videos either... super curious about the mysteriously hidden barbed wire. .:) Also, a side note, for the high and frozen snow up there... why did you make the entrance flat? NOT TO CRITICIZE AT ALL, but we live in the high mountains of Colorado, and we try to build structures with a slant roof for easy snow-removal / access. Still on your 1st year of your "Videos" in order... don't know if you slanted the root cellar entrance or not. ... ?


Aug 08, 2021

Not sure what you mean, “didn’t work out “ as it worked for your jars and potatoes, right?

Aug 12, 2021
Replying to

They mentioned it didn't work because they were experiencing too much water in the cellar. Then the walls started collapsing from the amount of water in the ground and it was too much weight against the sandbags.


Jay Held
Jay Held
Oct 07, 2020

Thanks Arielle! This was very helpful. I just built a mini root cellar under the cabin. Right now (October 7) inside the root cellar it's 57 degrees with 82% humidity. I have 2 inch vent pipe at the bottom and a 2 inch vent pipe out the top but I think I may need to put in 4 inch vent pipes to keep better airflow and maybe that will help reduce humidity. Just jars and potatoes going in our cellar.


Darrell Batton
Darrell Batton
Feb 16, 2020

Excellent write up. And almost a one-of-a-kind piece of reference work in the earthbag arena...many thanks for that! I am particularly interested to see where this portion of your adventure unfolds. Keep up the good work!


Feb 10, 2020

I'd love to see more of your preserving recipes on the web site, and wonder if you considered pickling eggs or canning chicken meat last summer.

bottom of page