Early in their education, meteorology students learn about the atmospheric scales of motion. Atmospheric scales of motion refer to the relation between the life span of a weather system to its horizontal extent.
At the smallest or “microscale,” turbulent eddies extend over meters to tens of meters in distance and exist for seconds to a few minutes in time. Microscale motion can be seen in a swirl of leaves in autumn. At the largest or “global” scale, polar front and subtropical jet streams encircle the Earth with its meandering long wave patterns that last for weeks or more. Following the seasonal increase and decrease in solar heating, the west to east flow of jet streams migrate north during spring and south in fall. In the mid latitudes, the changing positions of jet streams is evidenced by a change in paths and frequencies of cyclones.
The understanding of atmospheric scales is critical for predicting weather phenomenon. If one wishes to predict microscale eddies, then sensors would have to be arrayed on the order of a meter or less and measurements recorded over seconds. On the other extreme, weather balloons carrying radiosondes are launched twice daily from weather offices many kilometers apart. These instrumented balloons take measurements as they ascend from the surface to the upper atmosphere. Collectively, they can adequately discern the spatial and temporal dimensions of jet streams’ long wave patterns. The distance between instruments and frequency of measurement dictate what weather phenomena can be resolved in an observational network.
Today, the National Weather Service (NWS) networks of weather stations and radar can just barely resolve “mesoscale” weather phenomena, such as thunderstorms and tornadoes which extend over a few kilometers (little more than a mile) and last hours. It is because these networks provide only borderline spatial and temporal resolutions for severe weather, like tornadoes, that the NWS issues “boxes” for areas at risk and watches and warnings for specific time intervals. The public understands that not everyone will be hit by severe weather in a box. But that the risk is high for someone at some point in the box to be hit during the time interval of warning. By using boxes, the NWS is acknowledging the limitations in the spatial and temporal resolutions of its data-collecting systems.
The Importance Of Scale
Like weather, scale is important for precision agriculture. The average crop farm size in the U.S. is about 1,100 acres, or 4.45 square kilometers. If we assume that the average farm is a square, then spatial resolution of its one side would be 2.11 kilometers (1.3 miles). This spatial resolution is at the lower boundary of the atmospheric mesoscale and would be within a NWS box in the case of severe weather. It is also at the very edge of the finest spatial resolution of numerical weather models. The Real-Time Mesoscale Analysis (RTMA) model, developed by the National Centers for Environmental Prediction (NCEP), has a spatial resolution of 2.5 kilometers (1.55 miles) and a temporal resolution of one hour. It can provide real-time analyses of weather variables important to agriculture. This analysis model coupled with forecast models at cruder spatial resolutions can provide a weather data feed for programs supporting precision agriculture at the farm scale.
While the best weather analysis models are at the edge of the farm scale, precision agriculture sits at the “field” scale within the microscale of atmospheric motion. Since most decisions in precision agriculture are made at the field scale, biases and inaccuracies will be inherent in the cruder-scale weather analyses and forecasts. The existence of such biases and inaccuracies can be easily appreciated with a simple example. At its farm-scale resolution, the best weather analysis model can only provide a single hourly temperature for an entire crop field. Clearly, a single hourly value cannot accurately represent at the same time the temperature in the shade of a canopy or at the moist surface of a soil. There will be significant differences in temperatures across a field due to the aspect of the land, presence of a canopy, soil texture and moisture content and other physical properties in and around a field. The differences — while hard to measure — will define the microenvironment of plants in a canopy and habitats for pests. These differences are real and represent inaccuracies in weather model output valid at a farm scale.
The Role Of UAVs
Biases and inaccuracies inherent in farm-scale weather analyses and predictions can be corrected by either onsite measurements or other sources of field-level information. Enter the unmanned aerial vehicle (UAV). The UAV or “drone” is a relatively new technology being pursued in precision agriculture. Equipped with the appropriate camera and transmission platform, a UAV can remotely survey a field at a sub-decimeter (5 centimeters) resolution. A UAV can be deployed in short notice and supplement other technologies for collecting field-level data, such as satellite and aerial imagery, in-ground sensors or app-assisted observations. Its field-scale resolution complements the farm-scale weather analysis and forecast models.
The UAV is another tool in the precision agriculture toolbox. Along with weather and agricultural models, remotely-sensed imagery, in situ sensors and manually-collected measurements, it is another source of data for making field-level decisions. Like the sciences and technologies supporting it, precision agriculture tips the scales towards better crop management and more profitable production.