Sumatran Coffee Processing: Why You Should Know Giling Basah

A guide to understanding the Giling Basah method, a Sumatran coffee-processing tradition.

Opung Michael, a Sumatran coffee farmer in the Lake Toba area of Northern Sumatra
Opung Michael, a Sumatran coffee farmer in the Lake Toba area of Northern Sumatra

In the Bahasa language of Indonesia, Giling Basah means “wet-hulled.” That’s not very exciting in itself … but understanding the processing method is key to understanding why you might like (or dislike) a certain coffee. Processing is a huge part of determining the cup character of a coffee. Yet we hear much less about it than we do other aspects of a coffee’s origin, such as altitudes, varieties of plants, etc.

Giling Basah refers to a coffee process that is specific to Indonesia and creates a signature flavor. Wet-hulled coffees can have more body and lower acidity, but they also fall short of the sweetness and aroma uncovered by other methods.

If you have any interest in coffees from Indonesia, and are looking for some reason why you like them (or perhaps why you don’t like them), it’s worth your time to learn about Giling Basah.

While it’s more attractive to discuss coffee in a vocabulary validated by the wine discourse (varietal, terroir), these matter little when viewed through the lens of processing. If a coffee is “wet hulled” versus “wet processed,” whether it is a Typica or Bourbon, cultivar will be secondary to the way the cup flavors were shaped by processing.

Identical Twins

“Wet-hulled” versus “wet-processed”: They sound similar, but judging by the results they could not be farther apart. Wet-processed coffee (also called washed coffee) is the most common method the world over for high quality specialty coffee. Most everything you taste from Latin America is wet-processed coffee. Kenyas, Rwandas, Burundi are also all wet-processed.

Indonesia has islands that have historically produced wet-process coffees: Java, Bali, Flores. But the biggest origin, Sumatra, as well as Sulawesi, have typically produced Giling Basah, wet-hulled coffee.

Coffee coming from the pulpers to the fermentation tank, at Finca Los Angeles, Huehuetenango, Guatemala
Coffee coming from the pulpers to the fermentation tank, at Finca Los Angeles, Huehuetenango, Guatemala

In the traditional wet-process method, the fresh coffee fruit has its skin removed (pulped) and it is left in concrete tanks to ferment overnight. By a process of acidification and the breakdown of pectins the persistent, sticky fruit layer can be “washed” off of the inner green bean (still coated by a parchment shell). Next, the coffee is dried by the sun for 20 days or so, until it reaches an 11% moisture level. It can then be bagged up, warehoused for a 30 day resting period, at which time the parchment layer is shelled off, the coffee is graded by density and size and hand-sorted for export.

Wet-hulling shares the same initial steps as wet-processing. In much of Indonesia it is small-holder farmers who carry out the first steps of the process. Farmers pick the coffee and pulp it, which means that they run it through a hand-crank drum with a surface like a cheese grater that peels off the skin of the fruit.

Then they ferment the coffee in any number of ways – in a polypropylene bag, a plastic tub, or a concrete tank, to get the fruit layer (mucilage) to break down. After overnight fermentation, the mucilage can be washed off and you then have wet parchment coffee. The green bean is inside the parchment layer, still swollen with water. It’s basically identical to wet-processing up to this point.

As mentioned, in wet-processing a farm would now slowly dry this coffee to 11% moisture. The green bean would become the small dried seed we know and the thin parchment shell is easily removed, as there is a substantial spacial gap between the parchment shell and the much smaller dried green bean.

A cross section cut of green coffee in the protective parchment layer. In a fresh coffee cherry, the seed is swollen and there is no gap between parchment layer and the bean. When the coffee is dried, it shrinks and the parchment truly forms a shell.
A cross section cut of green coffee in the protective parchment layer. In a fresh coffee cherry, the seed is swollen and there is no gap between parchment layer and the bean. When the coffee is dried, it shrinks and the parchment truly forms a shell.

But in Indonesia, the farmer doesn’t want to wait for all this to happen. Is it because they want to create a unique flavor profile, to shape the cup character, to get more body out of the coffee? No! It’s because they want to get paid! They want to do as little work as possible to process the coffee and get cash. And who can blame them? Coffee is a cash crop after all.

Another cross section cut of fresh parchment coffee, the stage at which an Indo would be wet-hulled. Note the swollen green bean and lack of gap between parchment and seed.
Another cross section cut of fresh parchment coffee, the stage at which an Indo would be wet-hulled. Note the swollen green bean and lack of gap between parchment and seed.

So they take their clean wet parchment coffee, dry it a few hours until it has 50% moisture content, and sell it to a collector middleman at a local coffee market. They get paid faster and do less work this way. Each town in the coffee producing areas has a section where the collectors compete to buy the wet parchment coffee. The coffee market coincides with the town’s weekly market day.

Coffee collectors at the Siborongborong market sit atop their wet parchment coffee. Note the "litera" cans.
Coffee collectors at the Siborongborong market sit atop their wet parchment coffee. Note the “litera” cans.

The collector accumulates the humid parchment, sometimes mixing it, and sometimes keeping it in separate bags. It matters little; they might pay differently based on the wetness of the coffee, or penalize for a lack of cleanliness (usually in the form of incomplete fermentation, a reddish partially dried mucilage still clinging to the parchment layer). But in any case there is no discrimination for quality and it is all eventually mixed.  Collectors buy based on the “litera”, a standard 1 or 2 liter volume, typically using a can found in the area. They often buy the coffee from farmers and sell it to a larger collector or mill at the same price, but they buy using a heaped can-full (called a Bocho, a hill) and sell by the level can-full. The difference is their profit.

The wet-huller used to strip the coffee beans of fruit and parchment layers, which is about 3x the size of a dry huller since it has to generate much more friction.
The wet-huller used to strip the coffee beans of fruit and parchment layers, which is about 3x the size of a dry huller since it has to generate much more friction.

The mill might dry the coffee a little bit more for a day or two, but in general they send it to a special machine (the wet-huller) when the coffee still has 25-35% moisture content. This machine uses a lot of friction to take the tightly attached parchment layer and tear it from the water swollen green bean, which at this stage is often white and looks nothing like the green bean we finally see.

The wet-huller is massive, generates a lot of friction and transfers heat to the coffee. It is powered by a car or truck engine, often times! The heat it imparts to the coffee is not good for resulting cup quality.  Coffee is soft when it is wet, not the hard dried seed we know well, and is prone to being crushed. It is rubbery like a pencil eraser, but sadly it is not resillient like one. The friction of the wet huller can mangle the coffee and crush the ends in particular.

An array of defect green coffee beans pulled from coffee on a Sumatra drying patio. Crushed/split coffee on the left.
An array of defect green coffee beans pulled from coffee on a Sumatra drying patio. Crushed/split coffee on the left.

After hulling the coffee is laid out to dry, totally unprotected by any outer layer, on a patio, on a tarp, on the road, or sometimes on the dirt! Drying without the shell is rapid, so the mill is able to sell the coffee and get paid quickly. In no other coffee origin would they consider laying the green bean out to dry without the parchment shell protecting it.

But this is Indonesia, the great exception.    In the best mills, the coffee is treated with care in this state, dried on very clean patios. In the worst case it is laid out on the road. In any case, direct contamination isn’t the only concern for quality. Overly rapid drying is bad for coffee, and the parchment shell creates a neat little environment for each green bean to dry. It stabilizes moisture, moderates diurnal temperature swings, and lessens the impact of direct sun. Take that away and you have the possibility of harsh and uneven drying.

Wet-hulled coffee beans in Sumatra, white and swollen from water, called Gabah or, after hulling, Labu.
Wet-hulled coffee beans in Sumatra, white and swollen from water, called Gabah or, after hulling, Labu.

Coffees that are dried well, with an even and slow loss of moisture, will last longer when they arrive at the importing country; good tastes won’t fade quickly into papery or burlap bag flavors. The soft coffee is also damaged on the patio by being crushed underfoot, or in the process of unbagging and rebagging it daily.

So why is the coffee stripped of it’s parchment shell when it is wet? Why is the green bean laid out to dry without this protective layer? Again, time is money. The longer you have coffee in your possession, the more money you lose. The quicker you can sell it, the quicker you make back your investment and keep your cash flowing, and the quicker you can take your profit.

Freshly wet-hulled green coffee, left, and partially dried green coffee, right. Both are laid direct on the coffee drying patio, and the lack of the protective parchment layer permits faster dry times.
Freshly wet-hulled green coffee, left, and partially dried green coffee, right. Both are laid direct on the coffee drying patio, and the lack of the protective parchment layer permits faster dry times.

Drying coffee without the parchment is fast! Consider that in Central America the time from harvest of coffee cherry to export might be 2 to 3 months, including 20-25 days to try on the patio, then 30 days of “rest” (called repose) in the warehouse for the coffee moisture to stabilize. In Indonesia coffee is often exported in under 1 month from picking the cherry to loading the vessel at port!

The Crazy Indonesian Climate

To attribute Giling Basah to greed is false. Yes, people want to get paid quickly, but that is a factor in any coffee system. Indonesia faces a challenge in coffee production that few countries encounter to such a dramatic extent: weather. To be more specific, in Indonesia the farmer is trying to grow coffee in a humid climate. This means that the cherry is on the tree for many more months than in climates with clear weather patterns, with one or two blooms of the coffee flower per year. There are places in Central America where a coffee picker might visit a tree 3 times in the harvest to select the ripe fruit, and one last time to “clean” the tree of everything remaining. In Indonesia the farmer is harvesting the coffee weekly for up to 8 months of the year, revisiting a tree countless times to remove the ripe coffee.

More importantly, drying coffee in Indonesia is a nightmare. A “good” day for drying might have 4 hours of partial sun in the morning, and then rain in the afternoon. The coffee has to be collected from the patios before rains come, and the labor to get the coffee even slightly drier each day is taxing on the business of the coffee mill. Many of the mills we work with are building covered drying patios to increase drying times, since even in a rain shower there is heat and air movement to dry the coffee.

Small scale, raised-bed, covered coffee drying at a site we buy from in Indonesia. This is ideal for higher cup quality.
Small scale, raised-bed, covered coffee drying at a site we buy from in Indonesia. This is ideal for higher cup quality.

Keeping the parchment shell on each green bean would slow this process radically, and when coffee in parchment has been drying more than 30 days it shows problems in the cup. So in this light, stripping off the parchment shell when wet guarantees a better outcome than leaving it on in most coffee regions of Indonesia. Dry-processing coffee, where the entire fruit layer is left on the bean, is even more impractical in Indonesia as it would take even longer to dry the green bean. Mechanical drying or covered drying in hotter climates might make it feasible.

The Indonesian port cities that coffee is exported from, such as Medan in Sumatra, are exceptionally hot and humid. Oftentimes, coffee dried to 11% moisture at altitude in a place like Lintong gain water content in Medan as they are being prepared for export. There are logistic and legal reasons that specialty grade coffees cannot be exported with high moisture, especially when they are to be bagged per 60kg unit. The importer would be cheated of a large volume of coffee if it was shipped wet.

So in Medan they often “flash dry” the coffee for several hours in the intense sun, and quickly bag it up for export in jute. You can bring the moisture level back down from 14% or 15% to 11 or 12% quickly, but it is terrible for the cup quality of the coffee and something they would do nowhere else. They wouldn’t have to anywhere els either, because other coffee origins don’t suffer such humid climates.

The “Indo Coffee” Taste

Let’s be honest: On a cupping table of well-processed Central American coffees, an Indonesia wet-hulled coffee would likely be thrown out as defective. The earthy and foresty flavors – herbal, sometimes mossy or even mushroomy – would be attributed to processing errors, and the coffee labeled as substandard.

Ibu Manurund depulps coffee cherry from the day's harvest at her home in North Toba region, Sumatra.
Ibu Manurund depulps coffee cherry from the day’s harvest at her home in North Toba region, Sumatra.

So why this schism in the way the coffee trade treats wet-hull Indonesia coffees?   It comes down to taste: If a Sumatra supplier can consistently provide the same coffee processed the same way, be it fruity or earthy, there are buyers who see this as a uniquely different flavor profile, and a welcome break from Central America, Colombia or Kenya coffees. And of course the bottom line is that their customers like it. Those who like minimally-processed wines or wines with complex flavors of leather, peat moss, fir, cedar, humus, tannins, will see something in the Sumatra flavor profile.

Peets and, in earlier days, Starbucks, made a signature flavor out of dark-roasted Indonesia coffee. The black oily cup was perfect for overloading with cream and customers saw this as “strong” coffee, something they identified as special because it wasn’t like Grandma’s can of Folgers. Light roast coffee was seen as “weak.” The thick body and low acidity of Indonesia further advanced the idea of a deep, brooding character (referring to both the cup, and the customer … or at least how they wished to be seen).

Cross section of a dark roasted coffee shows gas pockets formed by carbon dioxide formation.
Cross section of a dark roasted coffee shows gas pockets formed by carbon dioxide formation.

The fact is, dark roast levels obscure a multitude of defects in the cup. When Sweet Maria’s started back in ’97, right in the middle of this “specialty coffee as carbon-water” era, the samples of Sumatra coffees I looked at were hideous. And for perhaps a decade after that, many Indonesian coffees were exported with impunity, as “Grade 1 Mandheling” seemed to mean nothing, and 100+ physical defects per 350 gram sample of green beans was common. Indonesian exporters, perhaps as the most dubious in the coffee world, were not only given a free pass on the cupping table, but in terms of cleanliness of coffee preparation as well.

Recovering Giling Basah from the Dustbin

This might make it sound like all wet-hulled coffee is bad, since this method isn’t really rooted in creating good tasting coffee, but rather speeding up the process and mitigating weather problems.

But there is good versus bad wet-hulled coffee. There are mills drying coffee on patios so clean you could eat off them, covered in a green house-like structure to protect from the unpredictable Sumatra rains, treating the coffee with great respect, and consistently producing great lots. It takes a lot of cupping and identifying a different set of reference points to determine what a really good wet-hulled Sumatra should be.  We look for sweetness in the cup, an expanded definition of sweetness than one might use when cupping other origins. This could be raw sugar, like muscavado, or molasses. It could be unique syrups like brown rice syrup, or sorghum syrup. In any case, a coffee with no sweetness is rarely, if ever a good coffee.

Better tasting and more stable coffees will result from better drying methods - here a coffee collector has built covered drying while still using exposed patios when the weather is good.
Better tasting and more stable coffees will result from better drying methods – here a coffee collector has built covered drying while still using exposed patios when the weather is good.

We look at the rustic elements to distinguish gross and off-putting flavors like dirt from positive “clean-earth,” like humus or other relatively clean, natural scents and tastes. While slight green herb and mossy is good, vegetal notes that are too bitter hint at poor processing or under-ripe fruits.

In our lab we also check the defect count, ultra-violet appearance of the coffee, water activity, humidity, and density of the bean. These tell the story of the coffee, but ultimately we find that cupping reveals the truth just as well. Samples are cupped at lighter roast levels so nothing is obscured by the roast taste, as well as more developed levels to show what flavors are possible with a darker treatment.

New Directions in Old Coffee Tradtions

Sumatra and Sulawesi have traditionally been Indonesia origins that were wet-hulling all their coffee. There have been instances of fully-washed coffee mills operating in Sumatra, like those of the dubious Holland Coffee operation in Aceh. In Sulawesi there was the plantation and wet mill of Sulatco and the forced demucilage (i.e. machine washed) coffee of Toarco’s Pedemeran plantation.  Other islands were producing fully wet-processed coffee like the PTP Government Estate coffees of East Java: Jampit, Blawan, Kayumas. Bali coffee was often wet-processed, and newer coffee projects in places like Flores, and East Timor are washed coffee.

The market likes wet-hulled flavor from Indonesia so some retrograde buyers are requesting Giling Basah process from places like Bali and Flores that did not historically do this method. This is part of a shift from an orientation to the Japanese market, which wanted cleaner cup character, to a part of the American market that operates with the ’80s mentality serving up French roast Indo coffee.  We have a new designation for our coffee projects in West Java, Sulawesi and Sumatra. We call it dry-hulled coffee. We could call it wet-processed, but we mean to distinguish it from wet-hulling. Dry-hulled coffees cupped next to their wet-hulled counterparts show more brightness and a lighter body.

Ateng coffee variety with bronze tips. Ateng is, like catimor, a hybrid disease-resistent strain that has Robusta genetic material.
Ateng coffee variety with bronze tips. Ateng is, like catimor, a hybrid disease-resistent strain that has Robusta genetic material.

But dry-hulling also fails to hide bad cultivars, and a vast percentage of Indonesia is planted in these ignoble plant types. The most popular is Ateng, a type of coffee with robusta inputs to give it disease resistance, and even worse is TimTim, which is pure Timor hybrid, a spontaneous cross of arabica and robusta. In wet-hulling, the processing flavors mask these varietals and their woody or metallic taste in the cup. But if you improve your wet-process method and dry coffee to 11% before hulling, you start to taste the bad plant material.

In our projects we have had farmers sort out Ateng and TimTim, to plant better types like Jember or S-linea, or old local Typicas that are still present from the Dutch era.  The ascendency of Indonesian coffee will necessitate buyers to make these demands, to stop accepting defect-riddled coffees, to roast in a way that reveals the character of the green coffee, and to teach coffee customers what a clean, well-prepared Indonesia coffee can truly be.

–Thompson

A few more Sumatra resources in our coffee library:

Here’s a video Tom put together discussing coffee trading and cultivation traditions in Aceh

Check out Tom’s photo travelogue from a 2019 visit to Aceh, Sumatra

7 Responses

  1. What a fascinating article. I am learning so much, and a lot of it points to me drinking mediocre coffee (excusable, IMO), AND being a snob about it (less excusable, IMO). I always preferred dark roasts, and Sumatran coffees were some of my favorites. A “dark, brooding” Peets guy in a nutshell (I still like Peets). I certainly never paid much attention to processing methods. The current trends towards lighter roasts and the like are all really new to me. I am trying to keep an open mind and have been enjoying my explorations outside of my comfort zone. I included a bit of Sumatra with my first bean/popcorn popper order, so I am sure I will continue to get more of an education as I get hands-on, including trying some lighter and more medium roasted versions of the Sumatra.

    These articles are very informative, and I also enjoy learning about the coffee economies and business side of things from the perspectives of the farmers and others at the origin countries as well!

    1. Thanks for the comment! Its nice to hear that your interest in starting to roast is paired with an interest to learn about where the coffee comes from!

  2. Thompson,
    Thank you for the thought and effort that you put into this article. I found it to be educational, informative, and a genuinely interesting “Here’s what you need to know” primer. It was serendipity that brought me to the “Sweet Maria’s” website and your article has whetted my appetite to continue on to read the site’s other pages about coffee and its production.

    Thank you (again) for putting this article together …. It is appreciated.

    Regards,
    Paul Cabral

  3. Great article – thank you.

    We settled on 2/3 wet-hulled and 1/3 espresso blend for our coffee – we think it balances things very nicely and we get nothing but compliments about our coffee. We buy all our coffees from Sweet Maria’s, and enjoy the variations in taste that come from using different SM’s espresso blends and the different Sumatra and Sulawesi wet-hulled coffee as they are available from SM’s.

    Most of our coffee consumption at home is via a Gaggia Syncrony, an older, super-auto espresso machine that will put three different amounts of water through the puck. But we also use this same coffee for drip and French Press, with good results.

    We are home roasters but we keep a few family members and friends in coffee, and make a lot for my students and parents of students – I give music lessons at home for most of my living and a complimentary coffee or coffee beverage is always offered, no charge.

    I encourage everyone to try mixing wet-hulled and more traditional, higher-acid coffees together – it really makes for a great cup. What we do is about as far from making single-origin coffee as one could get, but we make what we like and we like what we make.

    1. Yes – I agree that blending wet-hulled and wet process coffees is a great idea! It allows you to get control of the amount of wet-hulled rustic flavors, since they do vary lot to lot, while also adding additional top end flavors to the cup. Great point Steve…

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Reddit
Facebook
Twitter
LinkedIn
Email