Actually local well data is absolute groundwater level which measured quarterly in a year and having units as meters. GRACE GWS results are changes of groundwater storage in cm. How to compare these two data sets?
1. You have to generate the TWS by deducting depth to groundwater level values from 30 m obeservation well. (since the monitoring wells in the region are shallow wells with 30 m depth ), then you can use this formula for calculating the GWS
ΔTWS = ΔRESS + ΔSMS + ΔGWS
ΔGWS = ΔTWS- ΔRESS -ΔSMS
Where total water storage (TWS). Changes in TWS include changes (Δ) in surface water reservoir storage (RESS), soil moisture storage (SMS), and groundwater storage (GWS):
2. For each pixel of GRACE data you have to multiple each each pixel scaling factor for TWS anomaly
and following paper you can use
Article Dwindling groundwater resources in northern India, from sate...
Article Estimating groundwater storage changes in the Mississippi Ri...
Article A comparison of terrestrial water storage variations from GR...
I would recommend you to compare the anomalies instead of absolute values. you can estimate anomalies from well observations by subtracting the observed levels from the mean. Compare observed GW anomaly with the anomalies of TWS and GWS (derived from TWS as explained by # Leelambar Singh ), which gives an idea of how effectively the GW storage changes are being captured by the GRACE in your area.
However, be aware that well observations of a single point may not represent the GW storage changes on the scale of GRACE pixel. In addition, there may be questions related to storage changes in the deeper aquifer, if exists, as the well is penetrating just 30 m.