4.2 Time-since-fire influences nitrate and DOC effect sizes
Nitrate increased in wildfire-impacted watersheds, and remained elevated for up to 10-years post-burn, consistent with previous research (Rhoades et al., 2019; Rust et al., 2018), but with an observable lag effect (Fig. 2). Nitrate effect sizes weren’t significantly higher compared to the unburned watershed until one year following a wildfire, indicating a delay in nutrient response (Fig. 2). This may be due, in part, to a lag in increased soil nitrate pools that have been observed in the first year post-fire (Wan et al., 2001), which have been tied to shifts in key controllers of soil nitrification post-burn, such as pH and ammonium availability (Gustine et al., 2022), limited vegetative uptake of nutrients (Adkins et al., 2019; Gustine et al., 2022; Smith et al., 2011), and/or climate related drivers decoupling nitrate mobilization from its post-fire production in soils (Hanan, D’Antonio, et al., 2016). We observed that variation in nitrate response was higher immediately following wildfire, and became more constrained three years after wildfire, which may indicate that spatial and temporal variability in soil nitrate pools are linked to hydrologic export across the studied systems, as nitrate mobilization is often tightly coupled to high hydrologic connectivity and flushing events post-wildfire (Bladon et al., 2008; Gustine et al., 2022).
While prior studies have generally reported increased nitrate export to streams following a wildfire (Coombs & Melack, 2013; Mast & Clow, 2008; Rhea et al., 2021), consensus regarding post-fire DOC dynamics remains elusive, suggesting that mechanisms driving DOC dynamics may vary more dramatically through time. We found that DOC decreased significantly in burned watersheds compared to unburned counterparts, and remained lower for at least five years following wildfire (Fig. 2). Lower DOC following wildfire observed in our meta-analysis may be related to alterations to DOC source pools post-wildfire. For example, organic carbon (OC) source pools may become depleted following a wildfire due to combustion of topsoil, thus limiting the amount of DOC available for export (Wagner et al., 2015; Wang et al., 2012). Moreover, wildfire can transform biomass to recalcitrant forms of pyrogenic carbon, which can be resistant to microbial degradation and less soluble in water, resulting in lower amounts of DOC released in burned watersheds compared to unburned watersheds (Barton et al., 2024).
The persistence of fire effects on in-stream concentrations indirectly reflect terrestrial ecosystem recovery, thus we hypothesized DOC would return to background conditions over time (Clay et al., 2012; Rhoades et al., 2019; Rodríguez-Cardona et al., 2020). Our meta-analysis confirmed this, showing a non-significant difference in DOC effect sizes greater than five years post-fire. Due to wildfire effects on hydrologic and biotic processes, we expected greater variability immediately following a wildfire and then becoming more constrained through time. However, we did not observe a significant difference between the variability of DOC response through time. This may be due, in part, to hydrologic controls on DOC post-fire. Run-off increased significantly in regions with a Mediterranean climate where the shrub vegetation recovered slowly and transportation decreased for extended periods of time (Kinoshita & Hogue, 2011), while in more mesic climates, streamflow is linked to interannual variability in precipitation (Holden et al., 2012). Variability in streamflow across climates is highest in the first two years post-fire, and gradually decreases through time (Saxe et al., 2018).