The amount of groundwater that needs to be captured for a given plume, per day (Q) can be calculated as follows:
Q = T x i x L
where,
Q = gallons per day
T = transmissivity of the aquifer in feet2 per day, or the hydraulic conductivity in feet per day multiplied by the thickness of the aquifer in feet.
i = non-pumping gradient of the aquifer
L = the width of the plume perpendicular to the flow direction in the aquifer.
For instance, a typical sandy aquifer could have a transmissivity of 25 feet/day x 100 feet in thickness, a gradient of 1-foot per 100 feet (0.01), and a plume width of 100 feet wide. Doing the math, the amount of groundwater naturally moving through this area is about 2 gpm at steady state conditions. In other words, 2 gpm will capture the plume for this area (assuming no other sources of water are entering the capture area). However, the gradient to the well will not show a well-defined cone of depression and capture area for the pumping well. Keep in mind pumping water level is usually not representative of water level immediately outside of the well (due to well inefficiency) and usually is not plotted on the piezometric map.
Unfortunately, I don't' have an alternative method for showing capture area except if the values for groundwater parameters are correct, (T, i, and L), then a model can be generated to show the theoretical capture area at a very fine scale, which is actually occurring at the site.