Search for: All records

Wildfires significantly alter hydrological and biogeochemical processes, impacting downstream water quality and posing risks to ecosystems and human communities. Following the 2022 Hermit’s Peak-Calf Canyon (HPCC) wildfire in New Mexico, the largest wildfire recorded in the state of New Mexico, we deployed high-resolution in-situ sensors at three locations along a > 160 km fluvial network to investigate event-scale solute transport dynamics and their environmental drivers. Our objective was to evaluate how post-fire runoff events influenced water quality behavior across spatial (headwaters to mid- and high-order streams) and temporal (event to seasonal) gradients. We found that acute water quality impacts were most severe near the burn area, where turbidity reached ~8,500 FNU and dissolved oxygen fell below regulatory thresholds. These extremes, largely missed by traditional discrete sampling, were strongly driven by storm event size and seasonal variability. In contrast, farther downstream, solute export behavior was better predicted by longer-term indicators such as time since the fire and vegetation recovery metrics. Our analysis reveals distinct spatial shifts in concentration-discharge behavior that depend on the water quality parameter type, event features, and site position in the watershed. These findings highlight the need for longitudinal, high-frequency monitoring to detect and anticipate wildfire-induced water quality risks and inform more adaptive, spatially targeted watershed management strategies. 

We definewebsto be the collections of producers and consumers (e.g., functions and calls) in a program that are constrained: in higher-order languages, multiple functions can flow to the same call, all of which must agree on an interface (e.g., calling convention). We argue that webs are fundamentally theunit of transformation: a change to one member requires changes across the entire web. We introduce a web-centric intermediate language that exposes webs as annotations, and describe web-based (that is, flow-directed) transformations guided by these annotations. As they affect all members of a web, these transformations are interprocedural, operating over entire modules. Through the lens of webs we reframe and generalize a collection of transformations from the literature, including dead-parameter elimination, uncurrying, and defunctionalization, as well as describe novel transformations. We contrast this approach with rewriting strategies that rely on inlining and cascading rewrites. Webs are an over-approximation of the semantic function-call relationship produced by control-flow analyses (CFA). This information is inherently independent from the transformations; more precise analyses permit more transformations. A limitation of precise analyses is that the transformations may not maintain well-typedness, as the type system is a less-precise static analysis. Our solution is a simple and lightweight typed-based analysis that causes the flow-directed transformations to preserve well-typedness, making flow-directed, type-preserving transformations easily accessible in many compilers. This analysis builds on unification, distinguishing types thatlookthe same from types that have tobethe same. Our experiments show that while our analysis is theoretically less precise, in practice its precision is similar to CFAs. 

The increasing severity and frequency of wildfires in forested watersheds pose significant challenges to water quality management. This study examines the impacts of the 2022 Hermit's Peak-Calf Canyon gigafire, the largest wildfire in New Mexico's history. The wildfire burned over 1,382 km², affecting a key watershed that supplies drinking water to Las Vegas, NM. We conducted a longitudinal assessment of post-fire water quality dynamics across a 170 km fluvial network, analyzing flow, water quality parameters, nutrient and metal concentrations, and mobilization patterns. We found that post-fire nutrient concentrations exceeded pre-fire medians by up to two orders of magnitude. Our analyses revealed solute-specific transport patterns that are difficult to predict with static watershed- or fire-specific characteristics (e.g., burned area and percent severities). ${\text{NH}}_4^{+}$, ${\text{PO}}_3^{-}$, and ${\text{NO}}_2^{-}$were closely and positively associated with discharge and turbidity near the burn perimeter, while ${\text{NO}}_3^{-}$and TON exhibited strong mobilization trends ~170 km downstream. In contrast to nutrients, calcium, magnesium, and manganese levels showed no significant pre- vs. post-fire shifts, while concentrations of trace metals like Cr³⁺, Pb²⁺, Zn²⁺, and Sr²⁺surpassed background levels and public health thresholds. Our findings emphasize the significant propagation of wildfire disturbances over hundreds of kilometers and suggest the need for integrated watershed management strategies, including the management of large-scale flood control mechanisms to mitigate the far-reaching impacts of water quality disturbances post-fire. 

Gradually typed programming languages permit the incremental addition of static types to untyped programs. To remain sound, languages insert run-time checks at the boundaries between typed and untyped code. Unfortunately, performance studies have shown that the overhead of these checks can be disastrously high, calling into question the viability of sound gradual typing. In this paper, we show that by building on existing work on soft contract verification, we can reduce or eliminate this overhead. Our key insight is that while untyped code cannot be trusted by a gradual type system, there is no need to consider only the worst case when optimizing a gradually typed program. Instead, we statically analyze the untyped portions of a gradually typed program to prove that almost all of the dynamic checks implied by gradual type boundaries cannot fail, and can be eliminated at compile time. Our analysis is modular, and can be applied to any portion of a program. We evaluate this approach on a dozen existing gradually typed programs previously shown to have prohibitive performance overhead—with a median overhead of 2.5× and up to 80.6× in the worst case—and eliminate all overhead in most cases, suffering only 1.5× overhead in the worst case.