Year 2019, 2020, and 2021 continuous measurements, every 15 minutes, were made of depth and stream temperature in a small headwater stream, Saw Mill Brook, Burlington, MA, draining a highly suburban catchment (72% residential) in the Ipswich River watershed. Discharge is determined from stage using discharge vs stage regressions.
This file consists of depth data collected using a HOBO datalogger deployed in a stilling well in the stream channel.
HOBO Datalogger: Depth and water temperature are recorded by the HOBO data logger. Depths are corrected for barometric pressure changes using readings in an additional HOBO datalogger deployed at Marshview Farm. The HOBO is deployed in a PVC tube, which was attached to ree-bar driven into the stream bed. The PVC tube has holes drilled to allow adequate water circulation.
ADDITIONAL RESEARCH LOCATION COORDINATES: In addition to the decimal degree coordinates given above, locations are also given here in Massachusetts State Plane coordinates. The stream is called SawMill Br. and is located in Burlington MA, in the Ipswich River watershed. Small amounts of forest remain (~14% in 1999), much of which is along the stream channel near the gaging site.
STATION STATE X STATE Y
YSI-SB 225840 919310
NOTES AND COMMENTS:
A fallen tree upstream of the monitoring station in 2016 affected the geomorphology and sediment deposition within in the stream. As a result, the monitoring station began accumulating sediment in the Fall of 2016,and by Spring 2017 was noitceably affecting stage readings. The monitoring station was thus moved to the opposite side of the stream on May 20, 2017 at approximately 9:30 AM EST, but reamined within the same hydrological control. Staff gage readings and discharge measurements taken since the move confirm a stage offset of 0.10m. That is, 0.10 m is subtracted from all depth readings from May 20, 2017 at 9:30 AM EST and later. With this correction, the current rating curve remains in place (see below) with an r2 value of 0.98.
Discharge rating curve :
Discharge (l/s) = 1380.4*depth^4.4564
where depth is Adj_HOBO_Depth
Since our Hobo dataloggers are placed in stream reaches with minimal artificial engineering (i.e. v-notch weirs, etc.), there are occasional environmental factors than influence the Hobo depth readings that must be corrected to accurately reflect the true conditions of the stream reach. These include silt/sand build-up in and around the stilling well or downstream debris dams that artificially elevate the water level of the pool where the hobo is located. We attempt to correct these changes to provide best estimates of stream depth and corresponding discharge. Especially with debris dams, we assume a slow build-up of debris and apply an accumulating off-set based on the amount of time betwen site visits. We note the reasons for all corrections using the "Flag" column in the spreadsheet and the explanations are below.
During 2019 and 2020, the depth readings from the HOBO logger became noticeably unstable, particularly during logger downloads. That is, when the logger was removed from the housing to be processed and then placed back in the housing, the stage reading often shifted by one to two centimeters. Because the HOBO dataloggers are placed in stream reaches with minimal artificial engineering (i.e. v-notch weirs, etc.), there are occasional environmental factors that influence the HOBO depth readings that must be corrected to accurately reflect the true conditions of the stream reach. These include silt/sand build-up in and around the stilling well or downstream debris dams that artificially elevate the water level of the pool where the HOBO is located. We attempt to correct these changes to provide best estimates of stream depth and corresponding discharge. The greater than normal instability beginning in April 2019 is likely a result of two possible issues. First, kinking in the chain which holds in the HOBO. Secondly, and more likely, the benthic environment under the HOBO housing appears to be rough and the HOBO appears to rest on the bottom of the stream. As such, if the HOBO rests on slightly different parts of the stream bed, the depth could change by millimeters to centimeters.
To account for this inconsistency, the HOBO stage record was compared to staff gage readings during 2018, 2019 and 2020. Staff gage measurements are made during every site visit, which occur every one to three weeks. In comparing these records, no noticeable shifts were observed until April 2019. These shifts in the HOBO stage record were most dramatic during data downloads, although some occurred during intervening periods. The shifts were accounted for in the Correction Factor and implemented throughout the record. When occurring at a data download, the shift is implemented immediately, e.g., the Correction Factor changes from 0.1 to 0.16 in adjacent measurements. When a shift was needed during an intervening period, the Correction Factor was adjusted during the rising or falling limb of a storm in accordance with USGS discharge processing protocols (i.e. Sauer, V.B., 2002, Standards for the Analysis and Processing of Surface-Water Data and Information
Using Electronic Methods: U.S. Geological Survey Water-Resources Investigations Report 01–4044, 91 p.). These changes are further confirmed by comparison of flow calculation with flow measurements over the period of record. These continue to align well and are reflected in the rating curve.
NA - not available
1 - datalogger malfunction
Data are missing from 2021-07-12 8:30 to 2021-07-22 10:15 due to the datalogger station being knocked over during a storm during which the logger drifted about 5 meters downstream and was foundn laying on the stream bed. On 2021-07022 the logger station was reinforced, the logger was reinstalled, and a discharge measurement was made from which a new correction factor was determined.
Version 01: March 17 2022. Data sources for 2019, 2020, 2021 were combined under one ID to upload as a single dataset to Drupal and EDI; data and metadata uploaded to comply with importation to DEIMS7 and LTER Data Portal. using checks with MarcrosExportEML_HTML (working)pie_excel2007_Jun2019.xlsm 6/7/19 12:58 PM for QA/QC to EML 2.1.0