Daily changes in TWS were estimated using the vertical and horizontal components of the GPS time series that recorded surface deformation due to changes in hydrological loading. We estimated the daily GPS positions using Jet Propulsion Laboratory (JPL) GIPSY-OASIS II in precise point positioning (PPP) mode (33), with resulting position solutions aligned to the ITRF08 reference frame. To correct for tropospheric delays, we used the Vienna Mapping Function 1 (VMF1) and nominals that are estimated every 5 min (16, 17), and we made corrections for ocean tidal loading (34). In postprocessing, we corrected for nontidal atmospheric and nontidal ocean loading using corrections from (10), which included hurricane-related pressure changes and the storm surge. The reduction in variance from the nontidal ocean and atmospheric corrections range up to 10 and 21%, respectively. Before these corrections, we applied a criterion to remove stations that exhibited large noise in their time series or were missing extensive data. We removed 19 stations that were missing more than 30% of their positions. We then removed an additional 15 stations that exhibited a variance that was three times that of the median variance estimated from the entire network. In the case that a particular GPS time series was missing a segment of data (those that passed the 30% criterion), we followed the approach of (13), where <3 consecutive days of missing data in the time series were replaced using a linear interpolation, and for data gaps of ≥3 days, we replaced missing values using the average position of the entire network. We found that the number of missing data positions on a given day for all stations across the entire network was <2% (equivalent to <4 of 219 stations; see fig. S16). This indicates that, when taking the network average position (in the occurrence a station was consecutively missing more than 3 days of data), it was not obtained from a small subset of the GPS network, and is therefore unlikely to be erroneously mapped into CME and removed by the ICA filter.

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.