Tree proximity dominates meter-scale microbial activity beneath mature Faidherbia albida trees on a sandy tropical soil in Zambia: evidence from spatial regression and blocked validation

Imagine the soil beneath your feet not as a flat, uniform blanket of dirt, but as a bustling, invisible city where trillions of tiny creatures—bacteria, fungi, and microbes—live, eat, and breathe. These microscopic residents are the engine of the earth; they break down dead leaves and turn them into food for plants, a process scientists call "microbial activity." But this city isn't spread out evenly. Just like a human city has busy downtowns and quiet suburbs, the soil has "hotspots" where life is frantic and "coldspots" where it's sluggish. One of the biggest factors creating these hotspots is the presence of trees. Trees drop leaves, drop roots, and change the soil chemistry, creating what scientists call "fertility islands"—pockets of super-fertile ground right under the canopy.

However, figuring out exactly how these islands work is tricky. Is the soil better just because it's close to the tree? Or is it better because the soil there happens to hold more water, or has more carbon, or is a different texture? And if we know the soil is better near a tree, can we predict exactly how good it will be at a specific spot, or does the pattern change too much to guess? This is the puzzle researchers in Zambia set out to solve. They wanted to map the invisible life of the soil around a specific type of African tree to see if distance from the tree was the main boss, or if other factors like soil texture or carbon content were calling the shots.

---

The Tree's Invisible Halo

In a sandy field near Lusaka, Zambia, a team of researchers decided to play detective with the soil. They focused on five mature Faidherbia albida trees. These aren't your average trees; they are "reverse phenology" stars, meaning they lose their leaves during the rainy season (so they don't block sunlight from crops) and grow their leaves during the dry season (providing shade and food). The scientists wanted to know: does the "fertility island" around these trees look like a perfect, smooth circle of super-soil, or is it a messy, patchy mess?

To find out, they didn't just take one sample. They treated the field like a giant grid. Imagine a checkerboard where every square is 2 meters by 2 meters. They dug up soil from 169 different spots on this grid, covering a 24 by 24-meter area. At each spot, they measured how much the microbes were "breathing" (a process called microbial respiration), which tells you how active they are. They also checked the soil's personality: how much carbon it held, how acidic it was, and whether it was made of sand, silt, or clay.

The Main Discovery: Distance is King

The results were clear, and they pointed to one main rule: How close you are to the tree matters the most.

The scientists found that the soil right next to the tree (within 0 to 2 meters) was a microbial party. The activity level there was an average of 4.19. But as you walked away, the party died down. By the time you were more than 6 meters away from any tree, the activity had dropped to an average of 2.50. That's a 68% drop in microbial energy just by stepping away from the tree's shadow.

Think of it like a campfire. Right next to the fire, it's blazing hot. A few steps away, it's warm. Ten steps away, you can barely feel the heat. The study showed that the "heat" of the soil follows this same pattern. The closer you are to the Faidherbia albida, the more active the microbes are. This relationship held true even after the researchers used complex math to account for other things.

What Didn't Cause the Party?

You might think, "Well, maybe the soil near the tree has more food (carbon) or a better texture, and that's why the microbes are happy." The researchers tested this idea, and here is where it gets interesting.

They looked at Soil Organic Carbon (SOC), which is basically the amount of dead plant stuff in the soil. They found that SOC was indeed higher near the trees, and it did seem to help the microbes a little bit. It acted like a partial pathway: the tree drops leaves $\rightarrow$ carbon goes up $\rightarrow$ microbes get happy. However, when they tried to use carbon levels to predict the activity at a new spot, it didn't work very well. Carbon was part of the story, but it wasn't the whole story.

Then there was soil texture (the mix of sand, silt, and clay). The field was mostly sand (about 80.94%), which is like a very loose, gritty soil. The researchers wondered if the tree was somehow changing the sand into silt or clay to make the soil better. They used a special math trick (called isometric log-ratio coordinates) to make sure they didn't get confused by the fact that sand, silt, and clay always add up to 100%. The result? Texture didn't matter. The mix of sand, silt, and clay didn't explain why the microbes were more active near the tree. The soil texture was just as sandy far away as it was close up.

Similarly, the pH (how acidic or basic the soil is) didn't show a clear pattern that explained the microbial activity.

The "Map" vs. The "Prediction"

Here is the twist in the story. The researchers built a model to predict where the microbes would be active. They tried to use the distance to the tree, the carbon, the texture, and the pH to guess the activity at a spot they hadn't measured yet.

The result? The model was okay at finding the average trend (it knew that, on average, being close to a tree is good), but it was terrible at predicting the exact activity at a specific spot. When they tested their model on new blocks of land, it only got about 4.8% of the variation right.

Imagine trying to predict the weather. You know that, on average, it's hotter in July than in January. That's a good rule. But if you try to predict exactly what the temperature will be at 2:00 PM on a specific Tuesday in July, you might be way off because of a random cloud or a breeze. That's what happened here. The "tree effect" is a real, strong average trend, but the soil is messy. There are little patches of high activity and low activity that the simple rules of "distance" and "carbon" couldn't explain.

Why This Matters

This study teaches us two big lessons for farming and nature.

First, don't treat the soil like a flat sheet. If you are a farmer in Zambia (or anywhere with these trees), you can't just take one scoop of dirt from the middle of the field and assume it represents the whole farm. The soil right under the tree is a different world than the soil in the open field. You need to sample them separately to understand what's really happening.

Second, proximity is a powerful clue, but not a crystal ball. We know that being near the tree makes the soil more alive, but we can't yet say exactly why at every single spot. It's not just the carbon or the sand; it's a complex mix of roots, moisture, and tiny animal habitats that we are still learning to map.

The researchers concluded that while the tree creates a "fertility island," that island isn't a perfect circle. It's a bumpy, patchy landscape where the tree is the main boss, but the local neighborhood still has its own surprises. So, while we know the tree is the star of the show, the rest of the cast is still improvising.