Smart Irrigation Connectivity: LoRaWAN vs Cellular vs Satellite vs WiFi, and What Each Costs Per Acre - DripMaster Agri

Smart Irrigation Connectivity: LoRaWAN vs Cellular vs Satellite vs WiFi, and What Each Costs Per Acre

The sensor is the easy part. You can buy a soil moisture probe that runs five years on two AA batteries for under a hundred dollars, shove it in the ground, and it will happily report water content every fifteen minutes. The hard part of smart irrigation connectivity is moving that reading from the probe to your phone when the field sits two kilometers from the farmhouse, the nearest cell tower is a rumor, and there is no WiFi for miles. That network layer is where most farm automation projects quietly die, and nobody warns you about it in the brochure.

The Smart Irrigation Connectivity Problem Most Guides Skip

Before you spend another dollar, understand what you are actually transmitting. A soil moisture reading is a few bytes. A valve command is a few bytes more. Even a weather station reporting wind, temperature, and humidity every ten minutes pushes less than five kilobytes a day. You do not need bandwidth. You need range, battery life, and gear that survives rain, dust, and a tractor driving past. That fact rules out most consumer networking gear.

The Four Ways to Get a Field Online

There are really only four serious options, and they sit on a spectrum from cheap hardware with no range to expensive gear that works anywhere.

WiFi is what you already have in the house. It is fast, the hardware is nearly free, and it dies at about 100 meters from the router, or 200 with a good outdoor access point. It also wants mains power and internet at every hop. WiFi works for a greenhouse or a high tunnel next to the barn. For a 40-acre field it is a non-starter.

LoRaWAN is a low-power, long-range radio protocol that runs on license-free bands: 868 MHz in Europe, 915 MHz in North America. It tops out around 27 kilobits per second, which sounds pathetic until you remember you need only a few bytes. In exchange you get roughly ten kilometers of range line of sight, and node battery life measured in years. You buy a gateway, put your own sensors on it, and you own the whole network. No monthly bill.

Cellular NB-IoT and LTE-M ride on the phone network. Every sensor has its own SIM and talks straight to the tower, no gateway needed, which is the draw. You pay the carrier a dollar or three per device per month, and you are limited to places that actually have coverage. On a rural farm, coverage is the word you should trust least.

Satellite IoT, mostly from Swarm (now under SpaceX) and Iridium, is the last resort and a genuinely good one. It works anywhere with a view of the sky, no tower, no gateway, no nothing. You pay for it: modules run over a hundred dollars and service is five to fifteen dollars a month per device, with latency measured in minutes rather than seconds.

LoRaWAN: The Workhorse When There’s No Infrastructure

If your fields are spread out and your cell signal is patchy, LoRaWAN is almost always the right answer, and it is cheaper than people expect. A single outdoor gateway costs between $150 and $500 and covers a radius that will swallow most farms. Soil moisture nodes with a LoRaWAN radio run $40 to $150. After that the network is free. You are not renting anything.

The tradeoff is that you become your own network administrator. You mount the gateway somewhere high with a clear view, keep it powered, and when a node stops reporting you troubleshoot it yourself instead of calling a carrier. For a lot of farmers that is a feature, not a bug. The system keeps working when the cell tower goes down, and your data never lives on someone else’s server unless you put it there. I have watched growers fall in love with LoRaWAN for exactly this reason, then tear their hair out when a node on the far fence line goes quiet and they have to drive out to find why. Owning the network means owning the problems too.

One caveat: that ten-kilometer figure assumes line of sight. Hills, dense tree lines, and metal sheds eat signal fast. In rolling country you plan on a gateway every few kilometers, or you raise the antenna and see what happens. It is radio, not magic.

Cellular: Pay Someone Else to Run It

NB-IoT and LTE-M are the sensible choice when you have one or two sensors near a town, or when you have zero appetite for running your own gateway. The hardware is cheap because the module is a five-dollar chip soldered into the sensor, and there is nothing to mount or maintain. You screw the probe in, it finds the tower, and it starts sending.

The monthly fee looks small, and per device it is. But it never stops. Ten sensors at $2.50 a month is $300 a year, every year, forever, and the number only grows as you add nodes. Over five years that subscription quietly passes the cost of the LoRaWAN gateway it was meant to avoid. Add the coverage risk, and cellular is a coin flip you should take only after confirming the signal with a cheap test device in the actual field, not the driveway.

Satellite: When There Is Literally Nothing Else

If you farm somewhere the cell map is blank and the nearest neighbor is a mountain range, satellite IoT is the answer, and five years ago it was not affordable. Swarm’s pitch is a tiny module for a little over a hundred dollars and service around $5 a month, which is remarkable for something that talks to space. The catch is data volume and latency: you get enough for sensor readings and the occasional command, not enough to stream anything, and a message can take minutes.

For remote livestock water points, a rain gauge at the back of a thousand-hectare property, or a pump that has to report it is alive once a day, satellite is the only option. Budget for it like a utility, not a one-time purchase.

What It Actually Costs Per Acre

Numbers beat adjectives, so here is a breakdown for a 40-acre operation with ten soil moisture nodes and four valve controllers.

  • LoRaWAN: one $300 gateway plus 14 nodes at $60 each is about $1,140 one-time, $0 a month. Call it $28 an acre and you own it.
  • NB-IoT or LTE-M: $0 upfront if the radios come built in, but 14 devices at $2.50 a month is $35 a month, $420 a year, $2,100 over five years.
  • Satellite: 14 devices at $12 a month is $168 a month. Only worth it where nothing else reaches.
  • WiFi: one outdoor access point covers the farmyard and maybe the near field. Stringing power and backhaul across 40 acres costs more than a LoRaWAN gateway and still underperforms it.

The more nodes you add, the more LoRaWAN wins, because its cost is in the gateway rather than the node, and it carries no per-device subscription. Cellular and satellite scale linearly and never stop charging.

How to Decide Without Overthinking It

Walk through three questions and you will land in the right place.

If you have reliable power and internet where the sensor lives, and it is within a couple hundred meters of the router, use WiFi. This is the greenhouse and high tunnel case, and the only place WiFi belongs in a field.

If you have confirmed cellular coverage and fewer than a handful of nodes, and you would rather pay a small bill than run hardware, use NB-IoT or LTE-M.

If you have many nodes, spread-out fields, or no cell coverage and no taste for a subscription, build a LoRaWAN network. Buy one gateway, test the range, and add more only where the signal actually dies.

If you are in the middle of nowhere with no tower and no neighbor, satellite is the tool. It is expensive, it is slow, and it works.

The Part the Brochures Skip

Two warnings before you buy anything. First, connectivity is a plumbing problem, not a farming decision. A perfectly networked sensor that reports soil moisture every ten minutes is useless if you do not know what that number means for your crop, your soil, and how your drip irrigation system should respond. Get the agronomy right first, then bolt on the radios. Second, every battery-life claim assumes a small payload sent infrequently. If you set a node to stream readings every thirty seconds because you can, the batteries die in weeks and you will blame the technology. Send what you need, when you need it, and a LoRaWAN node really will run for years on a couple of AA cells.

The connectivity question has a right answer for your farm, and it is usually cheaper than you think. Figure out your range and your node count, and the choice between a $300 gateway and a $2.50-a-month SIM mostly makes itself.