I ran a routine LC-MS sweep after last week’s storm and sucralose clocked 130 ng/L at a roadside culvert even as E… coli counts were already falling. As a wastewater fingerprint, why does this artificial sweetener persist in baseflow, and have you seen it outlast dye or caffeine in your catchments?
Sucralose’s the Teflon of tracers — hardly biodegrades, barely sorbs, shrugs off sunlight — so it rides baseflow after the bugs and caffeine drop; I’ve seen it outlast rhodamine WT by a storm or two in flashy urban creeks. Try a dawn baseflow grab and compute a caffeine:sucralose ratio (“<0.1 = wastewater”) alongside EC to fingerprint the source — if there’s septic influence, the tail can smear; does that match your culvert?
I pair sucralose with acesulfame-K and track the ‘ace-K:sucralose’ ratio through the baseflow recession; if the ratio stays low while microbes fall, it usually flags a steady groundwater-fed wastewater leak rather than a short storm pulse at the culvert. Small caveat: in shallow, sunlit reaches rhodamine and caffeine drop from photolysis/adsorption regardless, so the sweeteners look ‘persistent’ by comparison — do you have ace-K or specific conductance on those samples?
I’ve had this at a roadside culvert too (about 120–140 ng/L) and a quick check that helped was pre‑dawn grabs through baseflow and a sucralose vs chloride plot — when Cl is “back to background” but sucralose holds, it usually flags septic‑fed shallow groundwater rather than transient road inputs. Small caveat: did your LC‑MS run include an isotopically labeled internal standard — culvert DOC can suppress ionization and fake a “persistent” tail.
Sounds like classic groundwater carryover: when ‘E. coli falling’ but sucralose steady, I’ve traced it to shallow plume inputs that lag stormflow. If you can, run a quick NaBr slug and look for long tailing; if sucralose persists longer than the conservative tail, it points to a persistent subsurface source rather than in‑stream storage — it’s the glitter of tracers. Have you tried pairing it with carbamazepine on the same grabs to confirm a wastewater signal?
A quick add-on: run fDOM/tryptophan fluorescence through the baseflow recession; if fDOM returns to background while sucralose hangs about 130 ng/L, that points to a dilute, low-reactivity groundwater plume rather than lingering stormwater, @sking-style. Small caveat: make sure it’s not LC‑MS carryover — drop a field blank/auto‑rinse between grabs; sucralose can be the last guest to leave. Did you see any diurnal wiggle in fDOM?
Pair it with acesulfame‑K and watch their ratio during the recession; if the ratio stays flat while ‘E. coli’ falls, that ‘130 ng/L’ plateau is the glitter‑in‑the‑carpet kind of wastewater moving on slow paths, not in‑stream hangup. Caveat: ace‑K can photodegrade a bit in clear, sunlit reaches, so do pre‑dawn or shaded grabs. Do you have ace‑K on your LC‑MS?
I’d peg it to hyporheic/near-bank storage feeding a lagged wastewater signal; sucralose is the sticker-on-a-bumper of tracers — low sorption, slow biodegradation — so it can outlast caffeine and even Rhodamine WT in shaded culverts. Quick step: push a mini‑piezometer 10–30 cm into the bed and sample porewater for sucralose and chloride; if porewater exceeds the channel and the sucralose–Cl line stays tight through recession, you’ve got a groundwater bleed, @kpatel98. Only caveat is high DOC and warmer temps can nibble it down a bit, but not enough to erase that persistence.