What causes seasonal signal fade on a line-of-sight microwave link

If a hop's RSL dips every spring, holds steady through summer, and climbs back in late fall, that pattern has a name in the trade: seasonal fade, and on a microwave backhaul link it's almost always leaf-out.

It's not rain fade and it's not multipath

Rain fade shows up as a sharp, short-duration hit tied to a storm cell crossing the path. Multipath fade shows up at night or in still, humid conditions, often on long over-water or valley paths, and it comes and goes within hours. Seasonal fade is slower and more stubborn than either. It tracks the calendar, not the weather radar. A hop that's clean in January starts losing a dB or two in April, keeps losing margin through May and June as the canopy fills in, and doesn't fully recover until the leaves drop in October or November.

What's actually happening is that deciduous trees along the path, the ones you cleared or that were short enough to ignore when the link was engineered, put on a full season's growth of leaves and new branch extension. A stand of oak or maple that barely touched your 0.6 Fresnel clearance in the dormant months can intrude into the zone once it leafs out. The intrusion causes partial attenuation and diffraction loss from vegetation sitting inside the first Fresnel ellipse, degrading the signal without killing the path. A hard outage would mean the trees are fully blocking line of sight; this is a softer effect, and it's the reason the drop reads as a seasonal trend instead of a fault.

Why growth that was fine last year isn't fine this year

The part that catches planners off guard is that nothing about the hop changed. No new construction, no antenna misalignment, no equipment drift. The trees did the changing. A sapling that cleared Fresnel by six feet five years ago has put on five years of vertical growth and is now sitting inside the envelope during full leaf. Species matters here too: fast-growing softwoods and certain hardwoods can gain several feet of height a year under the right soil and light conditions, which is a lot faster than most RF teams revisit a path profile.

This is also why the same hop can show worse seasonal fade year over year even if last year's trim held. Canopy height resets the clearance math every growing season, and a path survey from three or four years ago is describing a canopy that no longer exists.

How to confirm it's vegetation before you dispatch

A few things separate leaf-out fade from other seasonal culprits like ice on the antenna or seasonal ducting:

• The fade correlates with local leaf-out and leaf-drop dates for the species along the path, not with a specific weather event.
• RSL degradation builds gradually over weeks, not hours.
• The affected spans are the ones with tree lines near midpath, where the Fresnel zone is widest.
• Trim history (or the lack of it) lines up: hops that were cleared two seasons ago are clean, hops overdue for trimming are the ones fading.

If those four line up, you're looking at vegetation encroachment, and the fix is a scheduled trim along the specific span.

Catching it before the fade ticket

The problem with leaf-out fade is that by the time it shows up as a fault ticket or a customer complaint about degraded throughput, the canopy has already grown into the clearance envelope for a full season. Walking every hop to eyeball tree height against a Fresnel chart doesn't scale once you're managing more than a handful of paths, and a crew dispatch per path adds up fast across a backhaul network.

Link Vegetation Monitor runs an annual multispectral pass with a 3D canopy-height estimate along your link paths and flags, per link, where vegetation height is closing in on your Fresnel clearance envelope, so trimming gets scheduled against canopy growth instead of a fade complaint. If seasonal fade keeps showing up on the same hops year after year, it may be worth seeing where your canopy sits against clearance before the next leaf-out season starts.