John Dinnall

Site Description

Catchment: Fitzhugh Run is a forested headwater stream in Baltimore County, Maryland (39°28’08”N, 76°34’19”W), draining 0.62 square miles of predominantly forested parkland within Gunpowder Falls State Park. The catchment drains as a tributary to Loch Raven Reservoir, providing a largely undisturbed forest buffer and representing one of the most protected land use settings among the monitored sites in this study.

Sensor Deployment: Three in-stream sensors were deployed at Downstream, Midstream, and Upstream positions, with a fourth Air sensor mounted as a dry atmospheric reference. The Downstream position showed active flow well submerged under water, at arm’s length from the stream floor. Midstream was fully submerged within a cavity-like area between an exposed tree root and a hill. The Upstream was then put in an area with wet soils.

Goal: To document flow patterns in this watershed and assess whether low-cost thermal sensing is a viable approach for detecting subsurface flow activity and flow permanence.

Data Note: Data is missing roughly from October to December because the iButton reached capacity. Lastly, we adjusted the sensors during initial deployment to ensure readings were on the correct network.

Study Area Map

Figure 1: Study area map of Fitzhugh Run, Baltimore County, Maryland. Sensor points indicate the locations of the Upstream, Midstream, and Downstream in-stream temperature loggers deployed along the longitudinal transect. The inset map shows Fitzhugh Run’s location relative to all study sites within Baltimore County.

Figure 2: Daily precipitation recorded at Fitzhugh Run from July 2025 through May 2026.

Figure 3: Stacked bar chart showing the proportion of days classified as wet (blue), dry (red), and uncertain (gray) for each sensor position at Fitzhugh Run across the full study period.

Figure 4: Daily wetness state classifications for all sensor positions at Fitzhugh Run from July 2025 through May 2026. Each row represents a sensor position (Air, Downstream, Midstream, Upstream) and each column represents a single day. Blue indicates wet, red indicates dry, and gray indicates uncertain classifications.

Sensor Images

07/10/25

Figure 5: Initial summer placement of all sensors, deployed along a longitudinal gradient. The Upstream sensor was placed in a small crevice with trickling water, the Midstream sensor in a section with gradual flow, and the Downstream sensor in a section with strong flow. One day after deployment, we moved the Midstream sensor farther upstream to capture stream dynamics better before the network merges into the Downstream reach.

01/21/26

Figure 6: Winter deployment in January. Leaf litter covers both the Upstream and Downstream sections, and ice has formed along the edges of the Downstream channel while flow remains active.

03/11/26

Figure 7: Early spring deployment in March. All in-stream sensors resided in wet conditions upon redeployment. Seasonal groundwater recharge sustained surface water in the upstream and midstream reaches while the downstream reach maintained continuous, robust flow.

05/29/26

Figure 8: Late spring to early summer transition period. All sensors were collected under wet conditions, while riparian vegetation began to establish around each sensor location.

Thermal Overview Plots

Figure 9: Hourly temperature recorded by all four sensors at Fitzhugh Run from July 2025 through May 2026. The Air sensor reflects ambient atmospheric conditions while the three in-stream sensors reflect channel thermal conditions at each longitudinal position. In-stream water temperatures begin in the low-to-mid 70s°F in summer before declining toward freezing by January, while the Air sensor exhibits considerably larger diurnal swings throughout the monitoring period.

Figure 10: Daily mean temperature for all four sensors at Fitzhugh Run from July 2025 through May 2026. Averaging hourly readings to a daily mean smooths short-term fluctuations and reveals seasonal temperature trends across the monitoring period. Convergence between Air and stream sensor temperatures may reflect overcast, cool, or rainy conditions suppressing diurnal contrast rather than reduced surface water presence.

Case studies

Purpose: To illustrate sensor response during different atmospheric and seasonal conditions. Each case study presents a four-panel diagnostic plot followed by a dryness score table for a single selected day.

4-Panel Plots: A four-panel diagnostic figure showing each sensor’s thermal behavior over a 24-hour period.

September 21st, 2025 (Late-summer dry day)

Figure 11: Four-panel diagnostic plot for September 21st, 2025. Temperatures ranged from 62–71°F under overcast skies, with wind gusts up to 12 mph. The Upstream and Midstream sensors show diurnal temperature swings similar to the Air sensor. The Downstream sensor stays thermally stable throughout the day, suggesting surface water enters the channel between the Midstream and Downstream positions.

Dryness Scores — September 21st, 2025
Sensor warming_score cooling_score daily_score conflict_flag wetness_state
Air 0.4997 0.9989 0.7493 FALSE dry
Downstream 0.0000 0.0000 0.0000 FALSE wet
Midstream 0.4997 0.9989 0.7493 FALSE dry
Upstream 1.0000 1.0000 1.0000 FALSE dry

Table 1: A summary of September 21st’s sensor thermal classification. The Midstream and Air sensors recorded the same daily score (0.7493), and both were classified as dry. The Upstream sensor was the driest position, with a score of 1.0. The Downstream sensor scored 0 and was classified as wet, and no conflict flags were raised.

January 9th, 2026 (Mild winter day)

Figure 12: Four-panel diagnostic plot for January 9th, 2026. Temperatures ranged from 44–51°F, with wind gusts up to 9 mph and light showers beginning around 6 PM. The Upstream sensor started below the Midstream sensor but warmed past its peak by early afternoon, consistent with its exposed position. After about 3 PM, temperatures at all sensors leveled off for the rest of the day.

Dryness Scores — January 9th, 2026 (Longitudinal Gradient)
Sensor warming_score cooling_score daily_score conflict_flag wetness_state
Air 1.0000 1 1.0000 FALSE dry
Downstream 0.0006 0 0.0003 FALSE wet
Midstream 0.0000 0 0.0000 FALSE wet
Upstream 0.5017 0 0.2508 TRUE wet

Table 2: A summary of January 9th’s sensor thermal classification. All three instream sensors were classified as wet, with no conflict flags. The Upstream sensor had the highest warming score of the instream sensors, but its cooling score of 0 kept its daily score just under the 0.3 wet threshold. The Air sensor scored 1.0 and was classified as dry.

January 25th, 2026 (Cold winter day)

Figure 13: Four-panel diagnostic plot for January 25th, 2026. Temperatures ranged from 14-25°F under cloudy skies, heavy sleet, and wind gusts up to 16 mph. The Air sensor was the only sensor with large diurnal swings above its morning minimum, as shown in the normalized panel. All instream sensors showed little fluctuation. Air temperature was noticeably below freezing, while the stream sensors stayed around the water-freezing temperature of 32°F.

Dryness Scores — January 25, 2026 (Longitudinal Gradient)
Sensor warming_score cooling_score daily_score conflict_flag wetness_state
Air 1.0000 1 1.0000 FALSE dry
Downstream 0.0000 0 0.0000 FALSE wet
Midstream 0.5003 0 0.2501 TRUE wet
Upstream 0.5003 0 0.2501 TRUE wet

Table 3: A summary of January 25th’s sensor thermal classification. All three instream sensors were classified as wet. Midstream and Upstream each recorded a daily score of 0.2501 and triggered conflict flags because their warming and cooling scores disagreed. Downstream scored 0, and Air scored 1.0 and was classified as dry.

May 20th, 2026 (Spring Precipitation day)

Figure 14: Four-panel diagnostic plot for May 20th, 2026. The rainy day had a temperture high of 63°F and had a low of 53°F. The air rises extremely faster than all of the other sensors esepcially almost ot the point of 15 degrees higher than its morning minimum. Behind that sensor is the downstream sensor oddly.

Dryness Scores — May 20th, 2026
Sensor warming_score cooling_score daily_score conflict_flag wetness_state
Air 0.9994 1.0000 0.9997 FALSE dry
Downstream 0.5006 0.1670 0.3338 FALSE uncertain
Midstream 0.0000 0.0000 0.0000 FALSE wet
Upstream 1.0000 0.1668 0.5834 TRUE uncertain

Table 4: Summary of May 20th’s sensor thermal classification. Midstream scored 0 and was classified as wet. Downstream scored 0.3338 and Upstream scored 0.5834, and both were classified as uncertain. Upstream also triggered a conflict flag because its warming score of 1.0 disagreed with its cooling score of 0.1668. Air scored 0.9997 and was classified as dry.

Analysis

Fitzhugh Run maintained consistent surface water at its Midstream and Downstream positions, which were classified wet on 87.5% and 80.7% of days respectively, with dry classifications on only 3% and 7.2% of days. The Upstream was classified dry on 46.2% of days and wet on only 39%, with the heaviest drying in summer 2025 and partial recovery during the winter months, as the daily dryness heat map shows. Summer also produced a reference inversion, in which the Upstream sensor warmed and cooled more sharply than the Air sensor, so Air was classified wet on 25.1% of days. This pattern reflects an exposed channel head with little water buffering rather than a failure of the Air sensor. These results suggest perennial flow at the Midstream and Downstream positions and seasonal intermittency at the Upstream position, though the classification cannot be extended to the broader stream reach without additional spatial sampling.