What was the first number you looked at the last time you received a soil analysis? If the answer is pH, you are in the majority, and that’s where the problem begins. pH tells you how acidic the soil is, but it doesn’t tell you how much aluminum there is or how much space that aluminum occupies compared to calcium, magnesium, and potassium that the root actually wants to take up. Two plots on the same farm can have the same pH in the report and be in completely different situations, and if you lime them with the same dose because the pH said the same, one will be short and the other will be over-limed. The data that resolves this is already on the paper you paid for, in the row that says exchangeable aluminum.
The previous blog of this season showed that fertilizer can arrive in the wrong week and be wasted even if the amount is correct, and the first one showed that soil supply and effective absorption are different numbers. The third reason for this gap is the most structural of the three. There are plots where the problem is not the date or the dose, but the root cannot grow enough to reach anything because the aluminum in the soil solution burns the apex and forces it to stay in the first few centimeters. A crop with a stunted root has less soil volume explored, less water available in summer, and less capacity to respond to any fertilization, no matter how well planned it is.
None of what follows requires a special analysis or a different laboratory than the one you already use. The complete exercise consists of reading two more rows from the same report, doing a division, and taking a sample at an additional depth. It is probably the cheapest adjustment of this entire season, and also the one that takes the longest to show results because lime does not react in weeks.
Two plots with the same pH and a large difference in aluminum
The confusion has a legitimate technical root. pH and exchangeable aluminum measure things of different natures. pH is a measure of intensity, reflecting the activity of hydrogen ions in the soil solution at the time of measurement, and it changes quickly with moisture, temperature, and recent fertilization. Exchangeable aluminum is a measure of quantity, expressing how many centimoles of aluminum charge are retained in the soil's exchange sites, and it changes much more slowly. One data point tells you how acidic the solution is today; the other tells you how much aluminum the soil has stored to replenish that acidity tomorrow.
The practical consequence is better seen with two reports side by side. Imagine a first plot with a pH of 5.0, exchangeable aluminum of 0.5 centimoles of charge per kilogram, and bases totaling eight, including calcium, magnesium, potassium, and sodium. And a second plot, also with a pH of 5.0, aluminum of 1.5, and bases totaling only two and a half. In the first, aluminum occupies about six percent of the effective cation exchange capacity; in the second, it occupies more than thirty-seven. The same pH, more than six times the difference in what really matters, and two liming decisions that have nothing to do with each other. The second plot urgently needs lime, and the first probably doesn’t need it at all.
It is worth clarifying that this is not exotic information. In the work that evaluated lime requirement methods in twenty-five locations in the Colombian coffee zone, all selected for having a pH below 5.5, the routine analysis included pH, organic matter, calcium, magnesium, potassium, and sodium, exchangeable aluminum extracted with 1.0 molar potassium chloride, and cation exchange capacity. It is exactly the same package that any commercial soil laboratory provides. The data is there, the extractant is correct, and the producer does not use it.
What potassium chloride measures and what it misses
Here it is important to be precise because this is where it is decided whether the number is useful as is or needs interpretation. 1.0 molar potassium chloride is a non-buffered salt, and what it does is displace by exchange the aluminum retained electrostatically in the permanent charge sites of the soil. In that extract, the species Al³⁺ represents about ninety-three percent of the recovered aluminum, so the number that appears in your report is real aluminum and not an indirect indicator.
What the method does not measure is the activity of monomeric aluminum in the soil solution, which is the fraction that effectively comes into contact with the root apex. And that distinction has a field consequence that almost no one explains to you. Organic matter forms complexes with aluminum, and by doing so, it lowers its activity in solution; the literature on aluminum complexes with humic substances shows that the addition of organic matter reduces the activity of monomeric aluminum and that humic acids lower the aluminum saturation of the soil both by reducing that activity and by increasing the exchange capacity of the system. In farm terms, a soil with high organic matter may report a scary exchangeable aluminum and be much less limited than the number suggests because a good part of that aluminum is complexed and not free.
Keep the organic matter data from your analysis because in the limits section it changes the reading of everything else again.
This does not mean distrusting the analysis. It means that exchangeable aluminum is a good risk indicator and not a direct measure of toxicity, and that the decision to lime is made by crossing it with two other things: the base cations that accompany it and the organic matter that buffers it. Studies that compared methods found that the activity of Al³⁺ in solution, the aluminum saturation of the cation exchange capacity, and the aluminum extractable with calcium chloride all relate to plant growth, but saturation is the one you can calculate without requesting an additional analysis.
Aluminum saturation: the number that sets the dose
The calculation is a division, and you can do it with the report in hand. You add the exchangeable aluminum plus the exchangeable bases, calcium, magnesium, potassium, and sodium, and that gives the effective cation exchange capacity. Then you divide the aluminum by that sum and multiply by one hundred. The result is the percentage of aluminum saturation, and it is the criterion that agronomy of acidic soils uses to decide whether to lime and how much.
Reference ranges help to locate yourself. Above pH 4.8, exchangeable aluminum tends to fall below 0.2 to 0.3 centimoles of charge per kilogram, and aluminum saturation drops to single-digit values, considered non-toxic for most crops. Below pH 5.5, aluminum solubilizes with increasing strength, and it is precisely in that range where much of the coffee soils are found.
Coffee, moreover, has a peculiarity that changes the objective. Research on acidity correction in Colombian coffee soils associated the greatest plant growth with a pH range between 4.9 and 5.7, with calcium between 5.1 and 11.2 centimoles of charge per kilogram. That range is lower than what generic fertility tables suggest, which usually point to a pH close to 6.5 because they are built with transient crops in mind. The practical reading is that liming coffee with corn criteria is over-liming, and that a reasonable goal in many plots is to bring the pH to 5.5 and lower aluminum saturation, not to raise the pH as much as possible.
As we already addressed in our blog Your soil can retain more phosphorus than your crop can use, which you can read here, there are soils where phosphorus gets trapped in forms that the root cannot reach. Acidity and aluminum are precisely the chemical mechanism behind that fixation, so correcting aluminum saturation not only frees the root but also changes the efficiency of each bag of phosphorus you apply afterward.
Knowing which plot is losing roots
Do you have today, per plot, the pH, exchangeable aluminum, bases, and calculated aluminum saturation, along with the date of the analysis and the date of the last lime application? In most farms, soil analyses exist, they are in a folder or an email, and they are not compared between plots or over time, so no one knows which plot is improving and which is acidifying year after year. Harvis helps you keep the analysis by plot with its date, calculate and see aluminum saturation without doing the math by hand, record lime applications with doses and materials, and compare the series over several years to know if the amendment you applied is working or if the plot has returned to square one. Write to us, and we will set it up based on the analyses you already have.
The first twenty centimeters are not where the problem is
Almost all soil sampling in the country is done at a depth of twenty centimeters, and for good reason, because that’s where the highest concentration of absorbing roots is and where fertilization reaches. The lime requirement study in the coffee zone took its samples exactly at that depth. The problem is that the root does not stop there, and neither does aluminum.
Research in highly weathered tropical soils found that aluminum saturation in the forty to sixty centimeter layer was most strongly associated with yield, indicating that subsurface acidity is a limiting factor that standard sampling simply does not see. A plot may have the first twenty centimeters corrected and a chemical barrier waiting just below, and the plant responds well in winter and collapses in the first strong summer because its root never went down to seek the water that was deeper.
Lime does move down, but slowly and little. In the trial that followed the effect of lime over eighteen months at two depths, the zero to twenty centimeter layer rose from pH 5.08 to 6.47 with three tons per hectare, while the twenty to forty centimeter layer barely rose from 5.15 to 5.85, and the maximum of the subsoil appeared later than that of the surface. In other words, the deep effect exists, it is of lesser magnitude, and it arrives with a delay.
For the subsoil, there is a different tool, and it is important not to confuse it with lime. Agricultural gypsum moves with percolation water along the profile, provides calcium at depth, and reduces aluminum toxicity in the subsoil, and long-term literature in tropical soils describes it as a complement to liming to improve the deep root environment. What gypsum does not do is raise pH because it is not a neutralizing material. Applying gypsum expecting to correct surface acidity is wasting money; applying lime expecting to correct the subsoil is waiting in vain.
From target saturation to bags of lime
With the criteria clear, the dose can be calculated in two ways, and it is important to know which one the person recommending it is using. One way aims to raise pH to a target value; the other aims to neutralize a fraction of exchangeable aluminum to bring saturation below a threshold. For coffee, the second path usually gives lower doses that are more consistent with the pH range that the crop prefers.
In the laboratory, there are methods of different demands, and that explains why two recommendations may not coincide. The incubation method is considered the standard, and in the Colombian work, it required fifty days between sample preparation, incubation, and pH measurement. The titration method with calcium hydroxide provided an estimate in five days. When you compare a recommendation made with a quick method against another made with the standard, the difference is not anyone's fault; it is the method. Studies that evaluated lime requirement methods on reference soils found that the recommended dose varies appreciably depending on the chosen procedure.
Then comes the quality of the material, which is where the most money is lost without realizing it. The relative power of total neutralization combines two things: the chemical purity of the material and its reactivity, which depends on particle size. In the trial that compared fine lime of about one-tenth of a millimeter against coarse lime of one to two millimeters, applied at three and six tons per hectare, the fine fraction produced faster and more uniform pH increases due to its greater surface area and higher solubility. Buying cheap coarse lime is buying a fraction of the effect you are paying for. And the choice between calcite and dolomite is not about price but about analysis because dolomite provides magnesium in addition to calcium and makes sense when your report shows low magnesium, while using it where magnesium is already high disrupts the cation balance.
When liming is not the answer
The first limit is over-liming, and it is more common than it seems because no one measures it afterward. In the same lime trial, three tons per hectare raised the surface pH to 6.47, and six tons raised it to 6.60, both values well above the range of 4.9 to 5.7 associated with the best coffee growth. Raising pH above that range reduces the availability of iron, manganese, zinc, copper, and boron, so you correct one problem and create another that you will later try to solve by buying micronutrients. The way to avoid it is simple: set a dose against a saturation target and re-analyze before repeating.
The second limit is the one we saved above. If your analysis shows high organic matter along with high exchangeable aluminum, the real toxicity is probably lower than the number suggests because part of that aluminum is complexed by organic matter and is not active in solution. In those soils, it is advisable to be conservative with the dose and give weight to the plant's behavior, not just the table. And it works in the other direction as well; a soil with low organic matter and moderate aluminum may be more limited than the number indicates.
The third limit is time, and it is what makes many producers conclude that lime does not work. In the eighteen-month follow-up, the maximum pH values appeared between twelve and eighteen months after application, depending on the layer and the particle size of the material. The lime you applied three months ago has not finished reacting, and measuring pH at sixty days to evaluate whether it worked is measuring too soon. This forces planning liming as a decision for the previous campaign, not as a remedy for the current cycle.
The fourth point is about transferring the figures. The doses of three and six tons, the resulting pH, and the reaction times of the cited trials correspond to specific soils, climates, and materials. What transfers is the criterion: calculate aluminum saturation, set a target consistent with the crop, and verify with a subsequent analysis. The table does not transfer.
Four zones, two depths, and a scale of lime
The exercise fits in a plot and in a campaign, and does not require buying equipment. Walk through the plot and divide it visually into four zones based on what you already know about it: the high and low parts, the one that always yields less, the one that gets waterlogged. You don’t need a map; you need four zones that make sense to you and that you can find again next year.
In each zone, take a composite sample of fifteen to twenty subsamples, and take it at two depths, from zero to twenty and from twenty to forty centimeters. That’s eight samples in total for the plot, and that is the only real cost of the exercise. Label each bag with the zone and depth because an unlabeled sample is wasted money.
When the results arrive, calculate the aluminum saturation of each one with the division above and write it next to the pH on a single sheet. You will see two things at the same time: how much it varies between zones of the same plot and how much it worsens as you go deeper. That sheet is what decides whether you need a single dose, different doses by zone, or if the problem is below and surface lime will not touch it.
Apply lime only in the zones that need it, with the dose calculated against a saturation target, and note the date, the material, its neutralization power, and the kilos per hectare. If two zones ended up with similar saturations, combine them and treat them the same, as the idea is not to complicate the operation but to stop applying uniformly where there is nothing uniform.
Repeat the analysis in twelve months, in the same zones and at the same two depths. That second analysis is what turns the exercise into your information because it tells you how much pH has risen, how much saturation has decreased, and whether the material you bought performed as the label said.
The root is hindered by a number that is already in your analysis
There is something uncomfortable about this topic, and it is worth saying. There is nothing to buy to solve it; the data has already been paid for, the calculation is a division, and the most effective material is ground rock that costs a fraction of what a bag of fertilizer costs. That’s why no one explains it to you insistently, and that’s why it remains one of the most widespread limitations of coffee soils in the tropics. A plot with high aluminum saturation will respond little to anything you apply because the problem is not the nutrient supply but the size of the root system that can go seek them.
What you can do starting Monday fits into five decisions. Pull out the soil analyses you already have and calculate the aluminum saturation of each plot by adding aluminum plus bases and dividing because that number organizes priorities in ten minutes and at no cost. Stop comparing plots by pH and compare them by saturation since two plots with the same pH can be six times apart. Add a second depth, from twenty to forty centimeters, in the next sampling since that’s where acidity limits yield without the surface analysis showing it. Calculate the dose against a saturation target and against the pH range that coffee prefers, not against the 6.5 of the generic tables, and ask about the neutralization power and the particle size of the material before buying it. And schedule the verification analysis for twelve months because lime reaches its maximum effect between twelve and eighteen months, and evaluating it before is concluding that it didn’t work. The producer who makes that division today knows which of his plots have a chemical ceiling and can start to lower it; the one who continues to read only pH will keep fertilizing plots where the root has nowhere to grow and will keep attributing the result to the weather.
Do you want to have the analysis of each plot stored with its date, the calculated aluminum saturation, and the history of amendments to know if lime is working? Write to us via WhatsApp or schedule a demo, and we will show you how Harvis turns your analyses into a comparable series by plot.
Consulted References
- Mitigating Soil Acidity, Impact of Aglime (CaCO₃) Particle Size and Application Rate on Exchangeable Aluminium and Base Cations Dynamics. (2025). Sustainability, MDPI, 17(18), 8135. https://www.mdpi.com/2071-1050/17/18/8135
- Evaluation of a rapid method for determining lime requirements in soils of the Colombian coffee zone. (2021). Revista de la Academia Colombiana de Ciencias Exactas, Físicas y Naturales, 45(177). https://doi.org/10.18257/raccefyn.1405
- Correction of soil acidity, effects on initial coffee growth. Revista Cenicafé. https://publicaciones.cenicafe.org/index.php/cenicafe/article/view/13
- Correction of soil acidity, chemical alterations of the soil. Revista Cenicafé. https://publicaciones.cenicafe.org/index.php/cenicafe/article/view/1
- Assessment of soil pH and lime requirement methods and recommended lime rates for six reference soils across US land grant institutions. (2025). Soil Science Society of America Journal. https://acsess.onlinelibrary.wiley.com/doi/10.1002/saj2.70116
- Effects of liming on soil physical and chemical properties in Europe and North America, a review. (2025). Agrosystems, Geosciences & Environment. https://acsess.onlinelibrary.wiley.com/doi/full/10.1002/agg2.70175
- The meta-analysis study on the effects of the quality of lime materials on the soil physicochemical properties and crop yields in acid soils. (2026). Frontiers in Soil Science. https://www.frontiersin.org/journals/soil-science/articles/10.3389/fsoil.2026.1725559/full
- Long-term soil acidity dynamics and crop yield response to phosphogypsum and limestone with different reactivities in a no-till Oxisol. (2025). European Journal of Agronomy. https://www.sciencedirect.com/science/article/abs/pii/S1161030125002990
- Liming Method and Rate Effects on Soil Acidity and Arabica Coffee Nutrition, Growth, and Yield. Journal of Soil Science and Plant Nutrition. https://link.springer.com/article/10.1007/s42729-021-00550-9
- Lime requirement determination for acidic soils based on the measurement of soil mobilized aluminum with a portable colorimeter under field conditions. (2025). Journal of Environmental Management. https://www.sciencedirect.com/science/article/abs/pii/S0301479725041386
- Comparison of exchangeable Al, extractable Al, and Al in soil fractions. Canadian Journal of Soil Science. https://cdnsciencepub.com/doi/10.4141/cjss90-028
- Aluminum status of synthetic Al–humic substance complexes and their influence on plant root growth. Soil Science and Plant Nutrition. https://www.tandfonline.com/doi/full/10.1111/j.1747-0765.2007.00114.x
- Extractable Al and Soil Solution Ionic Concentrations in Strongly Leached Soils, Effects of Liming. International Scholarly Research Notices. https://www.hindawi.com/journals/isrn/2012/105127/
- Calibrated SoilOptix estimates of soil pH and exchangeable cations in three agricultural fields, implications for managing spatially variable soil acidity. (2024). Heliyon. https://pmc.ncbi.nlm.nih.gov/articles/PMC11402233/
- Improvement in Chemical Attributes of Oxisol Subsoil and Crop Yields under No-Till. Agronomy Journal. https://acsess.onlinelibrary.wiley.com/doi/abs/10.2134/agronj2013.0031
- Calcium sulphate, phosphogypsum and calcium carbonate in the amelioration of acid subsoils for root growth. Plant and Soil. https://link.springer.com/article/10.1023/A:1004285113189
- Food and Agriculture Organization of the United Nations. Arabica Coffee Manual, chapter on soils and acidity. https://www.fao.org/4/ae939e/ae939e06.htm