5 | DISCUSSION
By comparing historical and contemporary food webs in the San Juan River, we gained insight into trophic relationships among species, which supports a prior hypothesis that resources could be limiting the reestablishment of Colorado Pikeminnow in this modified system (Franssenet al ., 2007; Franssen et al ., 2019). Our results indicate the fish community reduced its overall trophic resource use, potentially increasing competition for resources, and Colorado Pikeminnow experienced a reduction in resource use likely due to a shift to feeding lower in the contemporary food web. Together, these results suggest trophic resource availability may be reducing prey fish populations, and in conjunction with the absolute loss of Roundtail Chub, could be hindering Colorado Pikeminnow’s ability to completely transition to piscivory and limit their ability to persist in the San Juan River.
Our results are consistent with other studies that demonstrated decreased resource use by fishes following river regulation (DeLonget al ., 2011; Turner et al ., 2015; Thoms and DeLong, 2018). Here we were also able to identify species turnover per se (loss of Roundtail Chub and addition of Red Shiner) as having a limited effect on this trophic contraction. The reductions in trophic resource diversity could stem from the loss of upstream inputs, limited floodplain exchange, and overall habitat simplification which can reduce primary basal and secondary consumer diversity and abundance (Vannoteet al ., 1980; Junk et al ., 1989; Turner et al ., 2015; Ward et al ., 2002; Winemiller et al ., 2010; Zeni and Casatti, 2014). Additionally, deviations of the natural flow regime can shift macroinvertebrate communities (Bunn and Arthington, 2002), potentially altering sources of carbon incorporated into the food web (e.g., relative abundance of allochthonous shredders vs. autochthonous scrapers). The changes we observed in the San Juan River fish community’s trophic structure indicated resource sharing increased which has the potential to affect population dynamics of native fishes.
Results assessing changes to trophic niche overlap indicated competitive interactions among all species potentially intensified between time periods. Invertebrate size, abundance, and diversity can substantially decrease after dam construction and all fishes included in our analysis likely forage on invertebrates at some period of their lifetime (Bunn and Arthington, 2002; Zeni and Casatti, 2014; Caldwell et al ., 2018). While changes to the San Juan River invertebrate community are unknown, if invertebrate numbers are insufficient, competition for this prey could reduce overall fish abundances (Van Poorten et al ., 2018). Indeed, the densities of invertivorous Flannelmouth Sucker, Bluehead Sucker, Speckled Dace, and Red Shiner have declined in the San Juan River over the last two decades following ongoing deviations from the natural flow regime (Pennock et al ., 2022). Given the surprising and significant increase in niche overlap by Colorado Pikeminnow into all other species, this may also be indicative of a paucity of fish prey given their presumed piscivorous nature (Vanicek and Kramer, 1969; Franssen et al ., 2007; Franssen et al ., 2019). Bioenergetics modeling assessing production and current fish prey abundances could be a useful path forward in quantifying the ability of the system to support Colorado Pikeminnow, however, to be information such modeling may need to include ontological prey preferences and Colorado Pikeminnow predation rates.
Competition for resources can cause intraspecific individual specialization (Araújo et al. 2011; Bolnick et al. 2010), which may explain some trophic dispersion changes of fishes between time periods. Two trophic generalists, Flannelmouth Sucker and Fathead Minnow, able to consume algae and detritus in addition to invertebrates increased their relative niche breadth (i.e., standard ellipse area) between time periods. Although this diet flexibility could be advantageous in situations where resources are limited, if specialization is towards resources that are less energetically efficient, population abundances could decline (Floeter et al ., 2004). Colorado Pikeminnow and Speckled Dace deceased their diversity of resource use, which could be in response to environmental change because they are habitat (Speckled Dace; Gido et al ., 1997) or trophic (Colorado Pikeminnow; Vanicek and Kramer, 1969) specialists. Thus, these species may be unable to adapt when trophic resources when they become limited (Bolnick et al ., 2010; Araújo et al ., 2011; Sanchez-Hernandez et al ., 2021).
Results from the stable isotope mixing models suggested Colorado Pikeminnow shifted towards increased insectivory in the current food web. Although we introduced uncertainty into our finding by estimating isotopic variation of historical invertebrates, we think our results are plausible because complete piscivory has been documented for wild-spawned Colorado Pikeminnow >150 mm TL elsewhere in the Colorado River Basin (Vanicek and Kramer, 1969) and the δ15N signature of the two largest historical Colorado Pikeminnow sampled here also indicated high piscivory rates. Additionally, contemporary consumption of invertebrates was described in a previous investigation where the diet of Colorado Pikeminnow 100 – 350 mm TL in the San Juan River was estimated to be ~25% invertebrates (Franssen et al ., 2019). However, due to similar isotopic signatures among prey fishes, modeled results had relatively high levels of uncertainty. Further insights into prey consumption through stomach analysis may be required to identify potential prey preferences of Colorado Pikeminnow and confirm the relatively high predicted consumption of invertebrates in the current food web.
Combined, cur results suggest Colorado Pikeminnow fed on relatively few fish prey contemporarily compared to historic conditions possibly due to reduced prey fish abundances. Although historical fish prey abundances that supported Colorado Pikeminnow in the San Juan River are unknown, a shift to greater insectivory could have implications for growth and survival of stocked individuals that infrequently recruit to the adult life stage (Clark et al ., 2018; Franssen et al ., 2019). For piscivorous fishes like Colorado Pikeminnow, the ability to move through ontogeny from an invertebrate diet to more energetically efficient fish prey is necessary for rapid growth (Mittelbach and Persson, 1998) and subsequent survival (Nilsson and Brönmark, 2000; Persson et al ., 1996).
For piscivores like Colorado Pikeminnow, the establishment of some nonnative species could be useful in subsidizing a reduced prey base (Boltovskoy et al ., 2018; Maerz et al ., 2005). For example, our mixing models suggested Colorado Pikeminnow consumed a significant quantity of nonnative Fathead Minnow in both periods and nonnative Red Shiner in the contemporary period. Increased densities of these small-bodied nonnative fishes have been correlated with higher body condition in Colorado Pikeminnow suggesting they could be an important food source (Osmundson in review ). Thus, managers may need to consider how reductions in these prey species could impact the availability of trophic resources for Colorado Pikeminnow.