Tue, Aug 16, 2022: 1:45 PM-2:00 PM
518B
Background/Question/MethodsFreshwater lakes and reservoirs provide valuable ecosystem services that are threatened by anthropogenic change, including land-cover change. To date, most evaluations of the effects of land modification on water quality have relied on remotely sensed data, which often miss important features such as houses located underneath primarily forested cover. We hypothesized that incorporating housing density into land-cover classifications would improve our ability to predict Secchi depth, a key metric of water quality. However, when we tested this hypothesis for 51 lakes across Maine and New Hampshire, including housing density in our models provided only marginal improvements in fit. We are currently repeating these analyses for 28 lakes in forested watersheds across Michigan and Wisconsin as part of a regional comparison using Secchi depth data from the US Water Quality Portal (WQP, https://www.waterqualitydata.us/). As before, we will use general linear models and partial least squares to compare the explanatory power of two suites of land cover classifications – metrics derived directly from the 2011 National Land Cover Database (NLCD) and “demographically enhanced” metrics that incorporate housing density based on the 2010 US Census – within a 100 m buffer around each lake.
Results/ConclusionsConsistent with ecological expectations, clearer water (i.e., deeper Secchi depth) was associated with higher percent cover in forest and lower percent cover in wetlands and agricultural land use in both regions. In the Northeast, there was limited evidence to support the hypothesis that demographically enhanced land cover categorizations improve our ability to predict average summer Secchi depth. The best-fitting model with NLCD data included the percent cover in forest and wetland (F2,48=26.5, P< 0.0001, R2=0.50); inclusion of high-density housing ( >80 houses/km2) under the forested land cover improved R2 to 0.55 with little impact on AICc. However, the association with high-density housing was positive, not negative, which we interpret as indicating both human congregation around high-quality lakes and the initial resilience of lakes to development. The Midwest analysis is still underway, but preliminary findings suggest some intriguing differences from the northeastern lakes. If these results continue to suggest regional differences, an important next step will be to determine the geographic conditions under which it will be worthwhile to obtain the demographically enhanced land cover categorizations, which is time-consuming and can only be done on a decadal time scale following the US Census.
Results/ConclusionsConsistent with ecological expectations, clearer water (i.e., deeper Secchi depth) was associated with higher percent cover in forest and lower percent cover in wetlands and agricultural land use in both regions. In the Northeast, there was limited evidence to support the hypothesis that demographically enhanced land cover categorizations improve our ability to predict average summer Secchi depth. The best-fitting model with NLCD data included the percent cover in forest and wetland (F2,48=26.5, P< 0.0001, R2=0.50); inclusion of high-density housing ( >80 houses/km2) under the forested land cover improved R2 to 0.55 with little impact on AICc. However, the association with high-density housing was positive, not negative, which we interpret as indicating both human congregation around high-quality lakes and the initial resilience of lakes to development. The Midwest analysis is still underway, but preliminary findings suggest some intriguing differences from the northeastern lakes. If these results continue to suggest regional differences, an important next step will be to determine the geographic conditions under which it will be worthwhile to obtain the demographically enhanced land cover categorizations, which is time-consuming and can only be done on a decadal time scale following the US Census.