Landscape factors influencing lake phosphorus concentrations across Minnesota

Cross, Timothy K.; Jacobson, Peter C.

doi:10.1080/10402381.2012.754808

Cited by 37 publications

(29 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our regression tree provides resource managers with a tool to quickly identify the eutrophic condition of reservoirs based on land cover and morphometry. For example, using the Ohio dataset, we found that very high eutrophication status reservoirs were only found in catchments composed of at least 49% row crop agriculture, which generally agrees with a recent effort in Minnesota examining landscape influences on lake TP concentrations (Cross and Jacobson 2013). The authors found a curvilinear relationship with catchment disturbance on TP; catchments with at least 40% disturbance had much higher lake TP concentrations than those with less disturbance.…”

Section: Discussionsupporting

confidence: 89%

“…Despite this complexity, agencies tasked with managing numerous reservoirs need to make decisions (e.g., where to stock fish, where to manage cyanotoxins) based on limited or no in-lake data. Thus, a recent emphasis is to create predictive models for lake management in which landscape-level attributes are explicitly incorporated and cost-effective approaches are considered (Catherine et al 2010, Soranno et al 2010, Cross and Jacobson 2013. These predictive models will be particularly powerful if relationships are robust across broad spatial scales.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting eutrophication status in reservoirs at large spatial scales using landscape and morphometric variables

Knoll¹,

Hagenbuch

Stevens

et al. 2015

View full text Add to dashboard Cite

Aquatic ecosystem management requires knowledge of the links among landscape-level anthropogenic disturbances and aquatic ecosystem properties. With large catchment area to surface area ratios (CA:SA), reservoirs often receive substantial terrestrial subsidies and can be particularly sensitive to eutrophication. Reservoir numbers and attendant management problems are increasing, and tools are needed to categorize their eutrophication status. We analyzed a dataset of 109 reservoirs in Ohio (USA) in an effort to classify eutrophication status using landscape-level features and reservoir morphometry. These predictor variables were selected because they are relatively stable and easily measured. We employed regression tree analysis and used a composite eutrophication variable as our response variable. Our regression tree analysis accurately divided 67% of Ohio reservoirs into 4 eutrophication status groups using 3 predictor variables: percentage of catchment area composed of agriculture versus forest; maximum reservoir depth; and CA:SA. We can infer that reservoirs with catchments containing >71% forest will likely be oligotrophic to mesotrophic. For reservoirs with <71% catchment forest, trophic status is determined by the relative extent of catchment row crops and either CA:SA or maximum depth. We applied our regression tree to a subset of reservoirs in the Environmental Protection Agency's National Lakes Assessment (NLA; n = 339 reservoirs). With a few exceptions, we categorized NLA reservoirs by eutrophication status despite their broad geographical range across the contiguous USA. Our results show that a few easily measured, stable parameters can classify reservoir eutrophication status. Models like ours may be useful for broad-scale management decisions.

show abstract

Section: Discussionsupporting

confidence: 89%

Section: Introductionmentioning

confidence: 99%

Predicting eutrophication status in reservoirs at large spatial scales using landscape and morphometric variables

Knoll¹,

Hagenbuch

Stevens

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Random forest models have been used in a variety of contexts in aquatic ecology, including predicting lake trophic status from landscape variables (Catherine et al 2010;Cross and Jacobson 2013), identifying habitat requirements of aquatic species (e.g., Hegeman et al 2014), and projecting the impacts of climate change on fish communities (e.g., Comte et al 2013). However, we are unaware of any other study using random forest to forecast recruitment of sport fishes.…”

Section: Predictive Model Of Walleye Recruitmentmentioning

confidence: 99%

Predicting walleye recruitment as a tool for prioritizing management actions

Hansen

Carpenter

Gaeta

et al. 2015

Can. J. Fish. Aquat. Sci.

155

View full text Add to dashboard Cite

Abstract:We classified walleye (Sander vitreus) recruitment with 81% accuracy (recruitment success and failure predicted correctly in 84% and 78% of lake-years, respectively) using a random forest model. Models were constructed using 2779 surveys collected from 541 Wisconsin lakes between 1989 and 2013 and predictor variables related to lake morphometry, thermal habitat, land use, and fishing pressure. We selected predictors to minimize collinearity while maximizing classification accuracy and data availability. The final model classified recruitment success based on lake surface area, water temperature degree-days, shoreline development factor, and conductivity. On average, recruitment was most likely in lakes larger than 225 ha. Low degree-days also increased the probability of successful recruitment, but primarily in lakes smaller than 150 ha. We forecasted the probability of walleye recruitment in 343 lakes considered for walleye stocking; lakes with high probability of natural reproduction but recent history of recruitment failure were prioritized for restoration stocking. Our results highlight the utility of models designed to predict recruitment for guiding management decisions, provided models are validated appropriately.Résumé : Un modèle de forêt aléatoire pour la catégorisation du recrutement de dorés jaunes (Sander vitreus) s'est avéré exact dans 81 % des cas (prédiction correcte du succès ou de l'échec du recrutement pour 84 % et 78 % des années-lac, respectivement). Des modèles ont été élaborés à partir de 2779 évaluations obtenues pour 541 lacs du Wisconsin de 1989 à 2013, et de variables prédictives reliées à la morphométrie des lacs, à l'habitat thermique, à l'utilisation du sol et à la pression de pêche. Nous avons sélectionné les variables prédictives de manière à minimiser la colinéarité tout en maximisant l'exactitude de la catégorisation et la disponibilité des données. Le modèle final catégorisait le succès de recrutement en fonction de la superficie du lac, des degrés-jours de température du lac, d'un facteur d'aménagement des berges et de la conductivité. En moyenne, le recrutement était plus probable dans les lacs de plus de 225 ha. De faibles degrés-jours se traduisaient également par une probabilité accrue de succès du recrutement, mais principalement dans les lacs de moins de 150 ha. Nous avons prédit la probabilité de recrutement de dorés jaunes dans 343 lacs considérés comme candidats pour l'ensemencement de dorés; la priorité en ce qui concerne l'ensemencement aux fins de rétablissement a été donnée aux lacs présentant une forte probabilité de reproduction naturelle, mais un historique récent d'échec du recrutement. Nos résultats soulignent l'utilité de modèles conçus pour prédire le recrutement pour ce qui est d'orienter les décisions de gestion, pourvu que ces modèles soient validés adéquatement. [Traduit par la Rédaction]

show abstract

“…Phosphorus concentrations directly affect hypolimnetic oxygen concentrations (Molot et al 1992;Jacobson et al 2010); phytoplankton productivity, including increased blooms of bluegreen algae, which can result in summer oxygen depletion in shallow lakes (Papst et al 1980); and filamentous algal density (Maberly et al 2002). Watershed land use is a primary driver of nonpoint nutrient loading in lakes with significantly higher concentrations of nutrients in runoff from agricultural, urban, and mining land uses than forests, grasslands, and wetlands (Heiskary et al 1987;Wang et al 2010;Cross and Jacobson 2013).…”

Section: Water Quality Habitatmentioning

confidence: 99%

“…A simple, yet direct watershed disturbance variable (percentage of urban, agriculture, and mining land uses in a catchment) was developed by Cross and Jacobson (2013) using National Land Cover Database 2001 land use GIS data. The percentage land use disturbance variable was significant in models predicting total phosphorus concentrations in Minnesota lakes (Cross and Jacobson 2013). Catchments with undisturbed land uses lie primarily in the Northern Forests ecoregion (CEC 1997; Level 1) and generally provide good water quality to lakes and streams in that region ( Figure 5).…”

Section: Water Quality Habitatmentioning

confidence: 99%

Full Issue PDF, Volume 41, Issue 6

2016

Fisheries

View full text Add to dashboard Cite

Landscape factors influencing lake phosphorus concentrations across Minnesota

Cited by 37 publications

References 39 publications

Predicting eutrophication status in reservoirs at large spatial scales using landscape and morphometric variables

Predicting eutrophication status in reservoirs at large spatial scales using landscape and morphometric variables

Predicting walleye recruitment as a tool for prioritizing management actions

Full Issue PDF, Volume 41, Issue 6

Contact Info

Product

Resources

About