ࡱ> RTr6EFGHIJKLMNOPQWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~,_--a..c//e00g11i22k33m44o55` =bjbjss 4o tv8 N-hV:jllllll$ht9 ;;; ([[[;  j[;j[[O J jwZL\lE<-BpqZqpq "[# / {[R-;;;;NNNaNNN    Chapter 2 Classification and Condition Assessment For New Hampshires Lakes Arlene Olivero Aquatic Ecologist/GIS Manager The Nature Conservancy, Eastern Regional Science Office Doug Bechtel Director of ConservationScience The Nature Conservancy, New Hampshire Chapter Table of Contents LAKE CLASSIFICATION Purpose 5 Lake classification background 5 Lake classification methods 10 Primary classification parameters 12 Secondary classification parameters 16 Classification results and discussion 18 Part 1. Primary classification parameters across all lakes 18 Secondary classification parameters across all lakes 24 Part 2. Classification of lake types 33 Biological system types 35 Biological system type descriptions 37 Very acidic ponds 37 Acidic ponds 39 Neutral shallow ponds 41 Acidic shallow lakes 43 Neutral shallow lakes 45 Acidic deep lakes 47 Neutral deep lakes 50 Lake condition background 53 Watershed and lakeshore development 53 Nonindigenous species 56 Hydrological change 59 Lake condition methods 62 Land Use 62 Isolation from roads 63 Dams 64 Conservation lands 64 Census and housing density projections 65 Lake Condition Results 67 Part 1. Results across all lakes 67 Isolation from roads 67 Land use 70 Dams 73 Conservation lands 76 Census and housing density projections 78 Part 2. Most and least natural lakes 81 Most natural lakes 81 Least natural lakes 84 Top 10 lakes 87 Condition discussion 91 Literature cited 93 Appendix 1. List and data of most and least natural lakes. (Chapter 2_AppendixI.xls) List of Figures Figure 1. Lakes by size class (for water bodies >= 2 acres). Figure 2. Deep Lakes. Figure 3: Acidification sensitivity class as modeled by CART analysis. Figure 4. High elevation lakes. Figure 5: Potential seepage lakes, or lakes with no mapped connection to a river system. Figure 6. Lakes with 200m shoreline buffer containing >25% wetlands. Figure 7. Lakes with 200m shoreline buffer containing >25% coarse sediments. Figure 8. Example of sinuosity classes. Figure 9. Biological system types. Figure 10. Isolated lakes. Figure 11. Lakes with development in 200m shoreline buffer and HUC12 watershed. Figure 12: Dammed lakes by dam type. Figure 13: Lakes with >=80% conservation land within 200m shoreline buffer. Figure 14. Lakes expected to experience housing density change from 1990-2020. Codes reflect change over time. Figure 15: Lakes embedded in a census block by rate of housing unit increase 1990-2020. Figure 16: Most Natural lakes by absence of dams, natural land cover, and proximity to roads. Figure 17: Least Natural lakes by presence of dams, natural land cover, and proximity to roads. Figure 18: Top 10 lake occurrences for each lake type. List of Tables Table 1: CART model output for lake ANC classes. Table 2. Elevation range of lakes. Table 3: Frequency and acreage of New Hampshire lakes by size class. Table 4: Depth Classes by Lake Size Classes Table 5. Acid neutralizing capacity (ANC) from CART model by lake sizes. Table 6: Lakes by elevation class and size class. Table 7. Lakes by connectivity and size class. Table 8: Wetlands in 200m shoreling buffer by size class. Table 9: Coarse sediment in 200m shoreline buffer by size class. Table 10. Sinuosity by size class. Table 11: Nineteen unique combinations of lake type based on size, depth, and modeled ANC variables. Table 12: Seven biologically relevant system types, based on likely patterns of species assemblages and relevant ecological processes. Table 13: Biological System Types by elevation zone. Table 14: Biological System Types by connectivity class. Table 15: Biological System Types by 200m shoreline buffer area dominant geology class (local geologic setting). Table 16: Land use model attributes for 200m shoreline buffer around lakes. Table 17: Lake types by isolation from road (500m, .5mi, 1mi). Table 18: Lake Types by percent developed in 200m shoreline buffer (developed = residential, commercial, industrial, transportation). Table 19: Lake types by percent developed in 200m shoreline buffer and HUC12 Watershed. Table 20: Lake types by presence or absence of dams. Table 21: Lake types by dam type. Table 22: Lake types by percent conservation land in 200m shoreline buffer. Table 23: Lake types by census blocks housing density class in 1990 and in 2020. Codes reflect change over time. Table 24. Most natural lakes by lake type. Table 25: Least natural lakes by lake type. Table 26. Statistics of top 10 lakes by lake type and conservation status within 200m shoreline buffer. Table 27. Statistics of Top 10 lakes by lake type and housing density projected changed 1990-2020 within census blocks. Purpose The objective is to develop a lake classification for the New Hampshire Statewide Wildlife Grant Project that 1) could be comprehensively applied to all lakes in the state and 2) would provide a useful context for evaluating patterns in biological, water quality, and socioeconomic variables, and 3) would provide quantitative data for future analysis. The lake type classification will use a physical environmental classification framework where local scale lake morphology characteristics define lake types within a larger environmental setting of elevation, geology, and landform patterns. An assessment of lake current condition and threats will also be performed using GIS measurable variables related to land use, roads, dams, and predicted housing density growth. Lake classification background Lakes are inland depressions containing standing water. Lake biotic communities vary widely within and between lakes due to a host of zoogeographic, morphological, physiographic, and hydrologic factors that operate at different spatial and temporal scales. At a very coarse scale, the current distribution of lakes is the result of landform patterns shaped by long-term glacial and geoclimatic processes (Silk and Ciruna, 2004). These processes created and molded physical lake habitats and connected stream networks over thousands of years and have led to the current associated large-scale patterns in aquatic biota distributions (Abell et al. 2000). Hierarchical environmental aquatic classifications, such as the USFS Hierarchical Framework of Aquatic Ecological Units (Maxwell et al. 1995), suggest that any small scale lake type classification be nested within a larger hierarchical classification framework of freshwater zoogeographic regions, sub-regions, and watersheds to address the influence of environmental and evolutionary processes that operate over larger spatial scales and longer time frames (Maxwell et al. 1995). Within this hierarchical stratification, individual lake types can be further defined by lake morphometry, local geology, and hydrologic connectivity (Maxwell et al. 1995). These local variables largely determine species diversity and are main factors determining the biological interactions and resultant biological communities found in lakes at smaller scales. (Eadie and Keast 1984, Marshall and Ryan 1987, Schupp 1992). Lake morphometry variables that drive ecological processes within lakes include lake surface area, depth, shoreline complexity, volume, and maximum length or fetch. Lake area has been used consistently in the literature as the best predictor of lake species richness (Minns 1989, Tonn & Magnuson 1982, Matuszek & Beggs 1988), however other morphological variables also exert a strong influence over key ecosystem attributes that shape lacustrine biological community types. For example in addition to area, lake depth, shoreline complexity, and maximum length or fetch also interact to determine the likelihood and extent of light penetration and thermal stratification (Silk and Ciruna, 2004). Light penetration is critical in shaping lacustrine aquatic macrophyte and phytoplankton communities and in this way, producing the habitats and resources other lacustrine animal communities depend upon. Lakes typically have three broad light influenced zones with very unique biological communities: 1) the littoral zone where sunlight fully penetrates to the bottom and enables emergent and submergent aquatic plants; 2) the pelagic zone of open water which may be either euphotic where light levels allows photosynthesizing phytoplankton or profuncal where light levels are too low for photosynthesis; and 3) the benthic zone which consists of the lake bottom and its accumulated sediment and detritus. Although the exact extent of these habitats is hard to measure, shallow lakes with high shoreline complexity have much more extensive littoral zones and than deeper lakes of similar surface area. Lakes also exhibit different biological communities based on thermal stratification. During the summer, lakes may exhibit the following major temperature zones 1) the epilimnion where warm water is well mixed, 2) the metalimnion of placid water where temperature drops quickly, and the 3) hypolimnion of cool/cold water at the bottom of the lake. Lakes shallower than 10 meters generally do not develop a stable thermal stratification during the summer as their wave action can stir water to such a depth that the thermal boundary never forms for long. At the other extreme, deeper lakes have a permanent summer thermocline and hypolimnion giving species access to permanent cold-water habitat (Silk and Ciruna 2004). Certain species requiring permanent cold-water habitat, such as lake trout, brook trout, rainbow smelt, burbot, and landlocked Atlantic salmon, will only be found in these deeper summer stratified lakes. In addition to lake depth, local climatic differences in water temperature regime associated with elevation may also explain differences noted in fish, macroinvertebrate, and plant communities between for example high elevation acidic lakes and low elevation acidic lakes (Langdon et al. 1998). In addition to light penetration and temperature stratification, other water properties influencing local lake biological community types include pH, nutrients, minerals, dissolved gases, and turbidity. The acidic or alkaline nature of lake water has been noted is a particularly key structuring variable for lacustrine communities. The water of most lakes has a natural range of variation between 6 a 9. pH below 4 or above 10 is toxic and very few aquatic organisms and even tolerate pH values below 5 or above 9 (Silk and Ciruna 2004). Acid intolerant fish of the northeast include the blacknose dace and creek chub which cannot tolerate pH of less than 6.0-5.5. Acid tolerant fish of the northeast include yellow perch, brown bullhead, and brook trout; however brook trout will not spawn if waters are too acidic (Brown et al. 1990). For example, acid intolerant macroinvertebrates, in this case defined by Odonates include (Gomphus and Basiaeschna) while highly acid tolerant invertebrates include (Cordulia and Leucorrhinia). The pH of the water in lakes is highly influenced by watershed characteristics and specifically the contribution of surface runoff, underlying geology, and the lakes zone of groundwater contribution. The underlying bedrock geology influences the pH as minerals with buffering capacity dissolve into groundwater water as it passes through underlying bedrock and surficial geology. Limestone and other forms of carbonate rock, release molecules have the ability to bind with and reduce the concentration of hydrogen ions in water. Lakes in granite regions tend to be more acidic because granite does not contain much if any buffering minerals. At the other extreme, lakes in some very calcareous bedrock material may contain high concentrations of minerals that buffer acidity thereby maintaining a relatively stable pH in the range of 7 to 8. Water in the atmosphere tends to be highly acidic because it absorbs a wide range of gases in the atmosphere. Pulses of runoff and snowmelt typically thus have low pH. Groundwater typically contain higher concentrations of dissolve minerals that may help temporarily or continuously hold the pH above 8. Concentrations of nitrogen and phosphorous determine the productivity of a lakes. Eutrophic lakes are nutrient rich lakes. They typically have large contributing watersheds with rich organic soils or agricultural soils enriched with fertilizers that yield higher nutrients. These lakes generally have a mucky bottom, cloudy, turbid, warm water and rich plankton and plant life. Oligotrophic lakes are naturally poor in nutrients. They generally have watersheds with infertile soils that release relatively little nitrogen and phosphorus. Many seepage lakes are oligotrophic because their generally small watershed areas contribute smaller amounts of surface runoff into the lake. Oligotrophic lakes generally have clear, cold water, a sandy or gravel bottom, and little plankton. Mesotrophic lakes are intermediate in status between oligotropic and eutropic lakes. The physical location or hydrologic connectivity of a lake within its larger landscape also plays a role in defining the lakes abiotic characteristics and resultant community types. The hydrologic position of lakes has been shown to be correlated with variation in a number of water chemistry attributes such increasing conductivity, Ca2+, Mg2+, alkalinity, dissolved inorganic carbon and pH with increasing downstream lake chain number. These patterns were partly determined by the effect of increasing catchment area; lakes high in the watershed receive a greater proportion of input waters from precipitation than lakes lower in the landscape. The patterns were also partly explained by the systematic processing of materials in lakes and in the stream segments between lakes (Webster and Sorrano 2000, Quinlan et al 2003, Kling et al. 2000). Lakes lower in the watershed may also experience higher flushing rates than lakes of similar surface area/depth ratio located higher. These lakes will often be less thermally stratified and have lacustrine communities with a higher percentage of species typical of more riverine environments (Hunt 2003). Seepage lakes (no inlets/outlets) and lakes positioned at the headwaters of a river drainage with only an outflow to a river have also been shown to be more sensitive to acid rain and snowmelt events because their hydrologies are driven by groundwater flow and precipitation events in a much smaller immediate local watershed (Quinlan et al. 2003) Species composition in lakes is effected by not only the influence connectivity has on current hydrologic and chemical regime, but also by the important role of connectivity for colonization and dispersal. For example, lakes that were formed in isolation and were never part of a larger water system tend to be species poor and fishless. Upper headwater lakes also tend to have a more limited fauna and flora than expected by size and depth due to barriers further down in the watershed that prevent access to higher lakes . These patterns are particularly evident for algal communities, macrophyte communities, macroinvertebrate assemblages, snails, and fish (Quinlan et al. 2003, Lewis and Magnuson 2000, Kratz et al. 1997). Relic lakes that were once part of a large waterbody, such as oxbows which were once connected to the river, share common species assemblages with each other and the former larger lake they were once connected to. Lake classification methods The GIS based lake classification measured key variables related to lake morphometry, local geology, and hydrologic connectivity. Meetings with NH DES staff, previous work implementing a GIS based lake type classification for Maine Department of Environmental Protection, and literature review led to determining the appropriate modeled variables (Olivero and Vaux, 2005). Modeled variables included lake size, depth, acid-neutralizing capacity, elevation, connectivity, shoreline complexity, percent coarse grained sediment in watershed, and the percent shoreline and watershed in wetland cover. The input datasets and classification variables are described below: Lake dataset: The statewide coverage NHWBODY of NH waterbodies (1:24,000 scale, NH Stateplane feet, NAD83) was used for the lake type classification. The hydrography comes from NH GRANIT, while the attributes of Entity_ID, WB_Name, WB_Desc, and WB_Class came from NH Dept. of Environmental Services. Source Katie Callahan December 1, 2004. Because this dataset also included some river waterbodies, lakes were extracted for analysis from this dataset according to the following rules. 1. All lakes with an Entity_ID starting with NHLAK or NHIMP to represent lakes and impoundments tracked and sampled for water quality and fisheries information by NHDES; (SWGINC = 1, 907 polygons, 809 unique Entity_ids (some Enity_ids had more than one polygon representing the lake or reservoir as certain bays were separate polygons). 2. All other waterbodies with a Granitid and Fgen of Lake/Pond or Reservoir and not NHEST (SWGINC = 2, 29380 polygons, 29380 unique Granitids). 3. Lakes were then assigned a new ID for use in SWG called SWGUNIQ. This field assigned a unique number to each Entity_id or Granitid that was in population 1 or 2 from the above queries. This resulted in 30189 unique potential lakes for the analysis. 4. Polygons were then dissolved on this SWGUNIQ to yield combined polygons for those lakes which separately were stored as multiple polygon features. Result: SWGLakes.shp (30189 polygons). 5. Although the original pond and lake polygon dataset included all lakes within the HUC8 watersheds intersecting NH, this analysis is limited to those lakes within or touching the border of the state of NH. The selection results in a population of 15979 lakes totally within NH and 7 that cross state line. This population of 15986 lakes was used in the analysis. Primary classification parameters Lakes are classified by the unique combination of Size, Depth, and Acid Neutralizing Capacity (ANC) class for each lake. These attributes have been used in other lake classifications schemes in New England (Langdon et al 1998, Hunt 2003). In addition, other key parameters influencing lake ecology, such as elevation, connectivity with river systems, sinuosity, shoreline or riparian wetlands, and coarse sediment in buffer parameters are used to display and organize classified lakes. Preliminarily, we organize lake parameters by size, and then display data based on our classification. 1) Size in acres classes (4): 2-9.9, 10-99.9, 100-999.9, 1000+. The difference between a lake and a pond is usually defined by differences in light penetration. In a pond, sunlight reaches all the way to the bottom, whereas in a lake the light does not reach the bottom. The lack of light at the bottom limits plant growth, which affects the distribution of plant-eating organisms and the biological communities present in the lake (Silk and Ciruna, 2004). Although light penetration relationship to waterbody depth, size, and water chemistry/turbidity are complex, New Hampshire Department of Fish and Game use a rule of thumb to define ponds as < 10 acres and lakes as > 10 acres. In this classification, we attempted to classify all waterbodies, however those < 2 likely function more as a wetland ecosystem. Above 10 acres in size, a log scale in size class break was used to define lake classes due to lack of expert opinion or biological justification for further size breaks This same size classification has been used by Maine DEP in their GIS based lake classification. 2) Depth classes (3): Never Stratified < 30ft, Strongly Stratified >= 30ft (9.1 meters). Summer stratification is driven by a complex process driven by interations of lake size, depth, volume, groundwater input, and flushing rate. Due to lake of available data to accurately model flushing rate, groundwater inputs, and volume of lakes, the Maine Department of Environmental Protection suggests a rule of thumb that lakes in the northeast over 30ft deep can be assumed to have strong likihood of thermal stratification and permanent summer cold water habitat. NH DES staff agreed with accepting this general classification break as a useful screening variable to quickly identify the lakes with the strongest likihood of supporting permanent deep cold water habitat and fisheries. Depth of lakes were obtained from Ken Edwardson at NH DES Water Quality Assessment Program 2/3/2005. Lakes were classified into shallow or deep based on their maximum depth being >= 30ft. Lakes with mean and maximum depths both above 30ft were flagged as being particularly noteworthy in their potential larger volume of cold water habitat. 3) Acid-Neutralzing Capacity (ANC) classes (3). Lakes with calcium carbonate concentrations > 12.5 mg/l are generally considered alkaline while lakes with calcium carbonate concentrations < 12.5 mg/l are generally considered acidic (Hunt 2003). Lakes with calcium carbonate concentrations <2 have pH levels below 5 and no longer support fish and many other forms of aquatic biota. ANC has a theoretical relationship with pH for water in equilibrium with atmospheric CO2, and calcium-carbonate concentrations are often measured as the key acidity/alkalinity variable rather than pH. The buffering capacity or calcium carbonate concentration is particularly critical in estimating how additional inputs of acidity will alter pH. Lakes were placed into the following three major categories of alkalinity 1: ANC< 2mg/l, 2 ANC 2-10 mg/l, 3 ANC>10mg/l. These classes were based on a review of the 6 acid sensitivity classes from EPAs classification of lakes for sensitivity to acid deposition. The classes were also guided by Nortons classification work to identify lakes of the following types: Norton Class 1: low to no acid-neutralizing capacity; Norton Class 2: medium to low acid-neutralizing capacity; Norton Class 3: high to medium acid-neutralizing capacity; Norton Class 4: infinite acid-neutralizing capacity (Norton 1985). Norton notes that sensitivity of aquatic ecosystems to acidic precipitation is based largely on the capacity of the drainage basin bedrock to assimilate acid during chemical weathering (Norton 1980). However any prediction about the vulnerability of a lake to acidification that is based solely on bedrock geology must be tempered with consideration of other factors. For example, small amounts of limestone in a drainage basin exert an overwhelming influence on terrains that otherwise would be very vulnerable to acidification. Although surficial sediments are most usually derived from the underlying bedrock, surficial sediment materials may also have different origin than the bedrock material and may be transported considerable distances. Consequently a particular lake might be underlain by granite and yet have calcareous glacial drift. The results would be a lake with relatively high alkalinity. The local geology of the lake can also play a large effect on buffering capacity, particularly for seepage lakes and small lakes. For example, deep coarse sediment deposits in the watershed or lake buffer area tend to increase the groundwater contribution to lakes. This increased groundwater contribution provides a water source with potentially more buffering chemicals than overland runoff water because it has had more time to contact bedrock and other materials while in residence in the ground water before entering the lake. Decomposing leaf litter also provides significant additional acidic input to lake pH. Classification and regression trees (CART) are a computer intensive alternative to fitting classical multiple regression models. They determine the best possible model for a response of interest using one or more predictors through the use of binary recursive partitioning, classification, and regressions. The Classification and Regression Tree (CART) software used was the commercial product manufactured and sold by Salford Systems ( HYPERLINK "http://www.salford-systems.com" http://www.salford-systems.com). CART models identify splitting variables based on an exhaustive search of all possibilities. Since efficient algorithms are used, CART is able to search all possible variables as splitters, even in problems with many hundreds of possible predictors. Various model runs identified HUC, local geology, and elevation variables as relatively highly correlated with lake ANC level.. The best applied model resulted in a 90-60-90% accuracy rate. This means that 90% if the very acidic lakes were correctly placed in the highest acidity class, 60% of the moderately acidic lakes were correctly placed in its class, and 90% of the neutral/alkaline lakes were correctly placed into its class. Because of the lower accuracy of moderately acidic class, some class 3-4 lakes will spill into both of the other classes. While not entirely accurate, this model provided the best combination of GIS variables to predict ANC across the entire lake dataset. The resulting ANC prediction was used as one of the Primary Classification Parameters to model Lake Types. Table 1: CART model output for lake ANC classes. =================== VARIABLE CODE =================== CODE Key Relative Importance ------------------------------------------------------- PHUCDIST %HUC12 Disturbed 100.000 PHUCFOR %HUC12 Forested 83.779 PHUCDEV %HUC12 Developed 81.306 ELEVZONE Elevation Zone 59.970 NHUC100 %HUC12 low buffering capacity 42.788 B1K_N300 1000m buffer med-hi buffer 36.514 NHUC300 %HUC12 med-hi buffer capacity 33.705 PHG800 %HUC12 coarse sediment / flats 33.260 PHUC_42 %HUC12 conifer 31.964 PHUC_41 %HUC12 deciduous 31.099 PHUCAGR %HUC12 agriculture 25.968 NHUC200 %HUC12 low-med buffer capacity 24.611 ACRESDEC Lake acres 24.079 PHUC_430 %HUC12 mixed forest 21.956 PHUCWET %HUC12 wetland 21.227 Secondary classification parameters Elevation / climate zone classes (4): Elevation can be used as a proxy for major changes in temperature, climate, and allochtonous carbon input (coniferous vs. deciduous): Elevation Zones suggested include the following major elevation zone of New England (Anderson 1999). Lakes were coded with the elevation zone of their centroid. Table 2. Elevation range of lakes. Name Range Example of Characteristic Vegetation Tidal 0-20 ft Tidally influenced ecosystems, tidal wetlands and marshes Very low 20-800 ft Oak, Pine-Oak, Pine-Hemlock, Spruce, Floodplain Low 800-1700 ft Hemlock-N.Hardwoods, N. Hardwoods Mid 1700-2500 ft Northern Hardwoods. Spruce-Hardwoods High >2500 ft Spruce-Hardwoods, Spruce-Fir, Alpine Connectivity classes (5): Lakes were coded as Seepage (not connected to stream network) or Drainage lakes (connected to an order 1 stream or higher). Lakes were coded as connected to the stream network if they intersected a NH 1:24,000k network centerline. In addition lakes were coded with the USGS SPARROW 1:100,000 lake centerline segment code where 1:100,000 centerlines and lakes were available. Not all 1:24,000 scale ponds and lakes were in the 1:100,000k SPARROW dataset. The size of the drainage area upstream of a connected lake helps inform how position in the watershed may influence hydrologic inputs. For example, smaller drainages are relatively higher in the watershed than larger drainage basins, and have hydrologic and chemical characteristics more influenced by local conditions than a lake lower in the basin where water quality and quantity from the incoming river outweigh local terrestrial inputs. Lakes were split into the following categories to stratify lakes by the upstream drainage area of the river output of the lake. Lakes not intersecting the SPARROW 1:100,000k hydrology linework were assumed to have drainage ares < 2 sq.mi. No connection to 1:100,000k hydrology or with drainage areas < 2 sq.mi. as defined by the SPARROW 1:00,000 hydrology Lakes with 2<30 sq.mi. river drainages Lakes with >30 sq.mi. river drainages Finally, we developed and report below a suite of additional descriptive parameters. These include shoreline sinuosity in comparison to perfect circle (sinuosity D = perimeter / 2 " Area * 3.1415926536); percent wetlands in a 200 meter buffer around each lake shore; and percent coarse sediment in 200m lake buffer and watershed. Classification results and discussion Part 1: Primary classification parameters across all lakes: To introduce the data, we first present descriptive information about lakes as they are distributed across the state. To simplify, all parameters are first reported and organized by Lake Size Classes. Part 2 organizes and displays results by Lake Types, as driven by the unique combinations of Size, Depth, and ANC that drives our classification. New Hampshire has nearly 16,000 polygons coded as ponds and lakes in the 1:24,000k hydrology dataset. Most of these are very small water bodies with less than 2 acre of open water (82%) which although coded as lakes/ponds by in the 1:24,000k GIS hydrology coverage are probably not true ponds but rather function as wetlands. Ponds ranging from 2-9.9 acres in size make up 11% of the water bodies, but account for only a small percentage (4%) of the total surface area of the state covered by lakes and ponds. The 1,120 lakes >=10 acres make up only 7% of the total number of lakes/ponds in the state, but account for the vast majority (93%) of the total surface area of the state covered by lakes/ponds. Because waterbodies less than 2 acres function primarily as part of a wetland ecosystem, all analysis and results focus only the ponds and lakes greater than 2 acres (information on the waterbodies < 2 acres is provided in the master dataset table). Table 3: Frequency and acreage of New Hampshire lakes by size class. Size Class in AcresTotal # Lakes and PondsPercent of Lakes and PondsTotal Surface Water AcreagePercent of Total Surface Water Acreage1: < 21306481.725590.52.972: 2-9.9180211.277949.74.233: 10-99.99055.6626599.214.144: 100-999.91961.234972326.435: 1000+190.1298284.552.24Grand Total15986100.00188146.9100.00 Figure 1. Lakes by size class (for water bodies >= 2 acres).  Table 4: Depth Classes by Lake Size Classes Size ClassDepth < 30ftDepth >= 30ft2: 2-9.9179753: 10-99.9827784: 100-999.998985: 1000+19Total # Lakes2722200Percent Lakes93.166.84 Very few lakes have depths >=30ft (Table 4). These lakes represent only 7% of the total population of lakes and ponds, and are by far the largest lakes. Most lakes smaller than 100 acres in size are shallow, and have relatively small amounts of deep, cold water habitats, although there is a huge variety of such habitats across the hundreds of medium and small lakes. Figure 2. Deep Lakes.  Table 5. Acid neutralizing capacity (ANC) from CART model by lake sizes. Size ClassANC < 2ANC 2<10ANC10+Grand Total2: 2-9.958967054318023: 10-99.92274642149054: 100-999.928138301965: 1000+18119Total # Lakes84412907882,922Percent Lakes294427 The results of the CART model for lake acidity show a fairly even distribution across the extremes, with the majority of lakes in New Hampshire in the moderately acidic class (Table 5). Almost all the very large lakes are moderately acidic. Figure 3: Acidification sensitivity class as modeled by CART analysis. Not sensitive lakes are considered neutral.  EMBED MSPhotoEd.3  Secondary classification parameters across all lakes Table 6: Lakes by elevation class and size class. Size Class0-20ft20-800ft800-1700ft1700-2500ft>2500ft2: 2-9.93595972569143: 10-99.974753804034: 100-999.91078635: 1000+1351Total # Lakes421554119611317Total Percent1.4453.1840.933.870.58 Most ponds are in the middle elevation ranges, while the majority of lakes in the highest elevations are ponds or small lakes. Similar to the distribution of elevation across NH, most lakes are found between 20-1700ft. In the lowest elevation, most ponds are smaller than 100 acres. Lakes in the tidal zone and at elevations >1700ft are rarer types and tend to be primarily small lakes <10 acres. Figure 4. High elevation lakes.  Table 7. Lakes by connectivity and size class. Size ClassNot Connected in GIS, Potential Seepage LakesDrainage Area < 2 sq.mi.Drainage area 2<30 sq.mi.Drainage Area >30Grand Total2: 2-9.917312343573818023: 10-99.923570286269054: 100-999.940129271965: 1000+41519Total # Lakes19618447761062,922Percent Lakes6.763.126.63.6 Only 7% of lakes in New Hampshire appear potentially disconnected from rivers. As lakes get larger in size a higher percentage are connected to larger river systems. The size of the river to which the lake is connected influences the biological character of the river based on the size of the connected river and size of the lake. For example, flushing rates will be higher in lakes with large inputs of water from connected rivers than in lakes of similar size and depth with less riverine hydrologic inputs. Smaller lakes connected to larger rivers may function hydrologically like a pool microhabitat of the riverine community and provide refuge for riverine adapted species. Lakes with larger watersheds will also be subject to the cumulative geologic, land use, and other point source pollution delivered from upstream sources. Although it is hard to choose thresholds of lake size and river drainage area size, lakes of a given size class with small upstream basins likely differ biologically than those with a larger upstream basin. Figure 5: Potential seepage lakes, or lakes with no mapped connection to a river system.  Table 8: Wetlands in 200m shoreling buffer by size class. Size ClassShoreline Buffer (200m) < 25% wetlandsShoreline Buffer (200m) >= 25% wetlands2: 2-9.915792233: 10-99.9846594: 100-999.918885: 1000+181Total # Lakes2631291Percent Lakes90.049.96 Ten percent of the states lakes are situated within large wetland complexes having shoreline buffers with >=25% wetlands. These lakes likely have complex habitat diversity and wildlife, with relatively shallow depths particularly near shore. Lake- and pond-shore beaver flowages, fens, and emergent marshes likely provide additional water quality filtering function compared to similar sized lakes without shoreline wetlands. Figure 6. Lakes with 200m shoreline buffer containing >25% wetlands.  Table 9: Coarse sediment in 200m shoreline buffer by size class. Size ClassShoreline Buffer (200m) < 25% coarse surficial sedimentsShoreline Buffer (200m) >= 25% coarse surficial sediments2: 2-9.913534493: 10-99.97351704: 100-999.9163335: 1000+181Total # Lakes2269653Percent Lakes77.6522.35 Twenty-two percent of lakes have shoreline buffers with >= 25% coarse sediment. These lakes are expected to have more hydrology driven more by groundwater and greater buffering capacity due to the increased percolation of water through groundwater minerals. Figure 7. Lakes with 200m shoreline buffer containing >25% coarse sediments.  Table 10. Sinuosity by size class. Size Class1=1-1.242=1.25-1.493=1.5-1.994=2-3.995=4+2: 2-9.948353152825193: 10-99.914221029825324: 100-999.95296081215: 1000+1135Total # Lakes63077088759837Percent Lakes21.5626.3530.3620.471.27 Sinuosity values closer to 1 approach a perfect circle. Lakes in the larger size classes (3, 4,5 ) have more sinuous shorelines with more extensive bay and littoral habitats. Figure 8. Example of sinuosity classes.  Part 2: Classification of lake types As an integrated classification, lakes were placed into unique categories based on their size, depth, and ANC parameters. These three were chosen as the primary classification variables to represent significantly different lacustrine habitats and biota across the state. Although 24 types were possible (4 size x 2 depth x 3 acidity classes), only 19 of these combinations occurred in the state. These 19 types were further collapsed into 13 Lake Types, which represent meaningful divisions based on size, depth, and acidity, and seven major Biological System Types, whose characteristics we believe justify unified processes that influence and control general species assemblage patterns (Table 11 & 12). We used the 13 Lake Types to stratify the Condition variables and analysis to ensure that representative habitats were expressed throughout the range of possible combinations and geographies across New Hampshire. While most ponds were shallow (i.e.<30 feet deep), a few ponds had maximum depth > 30ft. We combined these with the shallow pond category because they all had mean depths substantially below 30ft. In other words, even though they had a deep section, overall the pond provides little if any permanent cold water summer habitat for cold water fisheries. In addition, small, medium, and large lakes were combined into a single lake category and split by only depth and acid vs. neutral-alkaline. Table 11: Nineteen unique combinations of lake type based on size, depth, and modeled ANC variables. The subset of 13 Lake Types and 7 Biological System Types are described based on potential habitat and species assemblages. All Potential Lake TypeSize, Depth, Acidity CodeLake TypeTotal # Lakes% of all Lakes in the stateAcidic ponds11200267022.86Very acidic ponds11100158920.05Neutral shallow ponds11300354318.58Acidic shallow lake, small212004a40813.96(very) Acidic shallow lake, small211004a2137.29Neutral shallow lake, small213005a2067.05Acidic shallow lake, medium312004722.46Acidic deep lake, medium322006b662.26Acidic deep lake, small222006a561.92Acidic deep lake, large422006c180.62Neutral deep lake, medium323007b170.58(very) Acidic deep lake, medium321006b150.51(very) Acidic deep lake, small221006a140.48(very) Acidic shallow lake, medium311004b130.44Neutral shallow lake, medium313005b130.44Neutral deep lake, small223007a80.27(very) Acidic pond12100130.10Acidic pond12200220.07Neutral deep lake, large423007c10.03 Table 12: Seven biologically relevant system types, based on likely patterns of species assemblages and relevant ecological processes. Biological System TypeSummary Biological TypesTotal # Lakes% of all Lakes in the stateTotal Acres % Total AcresVery acidic ponds158920.226571.5Acidic ponds267022.929831.6Neutral ponds354318.623091.3Acidic shallow lake470624.23631919.9Neutral shallow lake52197.576654.2Acidic deep lake61695.812235767.0Neutral deep lake7260.982674.5TOTALS2,922182556 Biological system types While the Lake Classification in Table 11 above displays all possible combinations of our Primary Classification Parameters, the Biological System Types suggested here (Table 12) depend more on the lake processes that influence biological communities. In terms of habitat availability, it is more difficult to use lake size to justify specific lake types; size appears to be better expressed as a gradient within which multiple habitat types can exist. Depth, which influences temperature, light penetration, and stratification, and ANC, which influences multiple water quality and chemistry parameters, have much more direct influence on the biota and species assemblages that can live and breed in a given lake. We still distinguish between ponds (i.e. lakes <10 acres) and all other lakes (i.e. => 10 acres) because extreme small size can have direct influence on other important physical components of the pond system. For example, small water bodies (i.e. ponds) are generally not stratified (monomictic) and have relatively lower habitat diversity. Larger water bodies are generally more likely to become summer-stratified (dimictic) and provide much more habitat diversity and refuge for a broader suite of species. In addition, ANC, reported by general acidity categories, has much more direct influence on biological assemblages. Since there is still only emerging information on northeastern lakes and aquatic communities, we rely heavily on classification information developed in other New England states. Most of the species and community assemblage information provided below was developed for Vermont (Langdon et al 1998) and Maine (Hunt 2003) Figure 9. Biological System Types.  EMBED MSPhotoEd.3  Biological system type descriptions Type # 1. Very acidic ponds. The 589 very acidic shallow ponds (Type #2) are distributed throughout the state, but are concentrated in areas of higher elevation (Table 13) and are predominantly associated with acidic bedrock types (Table 15). Type #1 ponds are most likely found higher in their watersheds (Table 14), and only account for 1.5% of all lake acres in New Hampshire (Table 12). These lakes exhibit the highest acidity of any other lake type, and have characteristics such as tannic, brown water, very low biological productivity (dystrophic), shallow depths, and where tannic, low light penetration. They have very low pH and low ANC, with a range of substrates, but most likely mucky peat with organic material that does not decompose over time but rather accumulates, potentially forming bog-like conditions over time. At the highest elevations in the alpine and subalpine zone, this type includes alpine tarns, which are small highly acidic ponds. These ponds are generally non-stratified in summer, and may freeze to nearly the bottom of the pond (depending on its size and overall depth) in winter (monomictic). This type is also likely characterized by low species richness, and fish species may be absent in the most acidic examples. Where fish do occur, they likely are similar to the Brown bullhead-golden shiner assemblage identified by Hunt (2002) and Langdon et al (1998), with brown bullhead in the most acidic ponds, and golden shiner. Based on information compiled for lake types in Maine (Hunt 2002) and Vermont (Langdon et al 1998), biological characteristics of very acidic ponds may also include: Dominant vegetation assemblage: water lilies (Najas, Nuphar, Nymphaea) Potential macroinvertebrate assemblages: Crayfish (Decapoda), odonates (Odonata), bivalve (Sphaeriidae), amphipods (Amphipoda), mayflies (Ephemeroptera), true bugs (Notonectidae), caddisflies (Trichoptera), beetles (Coleoptera), and water strider (Gerris). Acid-tolerant indicators include odonates (Cordulia and Leucorrhinia, Aeshna and Ischnura). Other species characteristic of hard substrates (e.g., shale, cobble, possibly woody debris) including snail (Ferresia californica), and midges (Tribelos and Phaenopsectra). Other species characteristic of soft substrates (e.g., mud, possibly sand) including water beetles (Dytiscidae), true water bugs (Corixidae and Notonectidae). Type #2. Acidic ponds. This is the one of the most dominant lake type in terms of frequency, with 670 ponds (Type #2) distributed throughout the state; however, this type only accounts for 1.6% of all lake acres in the state (Table 12). They are most common in middle elevations (Table 13) and are predominantly associated with acidic bedrock types (Table 15) like Type #1, but they have a broader distribution both elevationally and across bedrock types. Like Type#1, they are primarily found higher in watersheds in which they occur (Table 14), and there are a relatively high proportion of non-connected acidic ponds (Table 14). They are primarily acidic, but not as likely to have pHs low enough to exclude fish as Type #1. They share other characteristics with Type #1, such as low biological productivity (oligotrophic) and relatively shallow depths. They have very low pH and low ANC, with a range of substrates, including sandy and rocky substrates, as well as mucky peat with organic material that does not decompose over time but rather accumulates, potentially forming bog-like conditions over time. These ponds are generally non-stratified in summer, and may freeze to nearly the bottom of the pond (depending on its size and overall depth) in winter (monomictic). This type is also likely characterized by relatively low species richness. Fish assemblages are likely similar to the Brown bullhead-golden shiner assemblage identified by Hunt (2002) and Langdon et al (1998), with brown bullhead in the most acidic ponds, and golden shiner, rainbow smelt, burbot and possibly stocked trout (brook, brown, rainbow) in others. Based on information compiled for lake types in Maine (Hunt 2002) and Vermont (Langdon et al 1998), biological characteristics of acidic ponds may also include: Dominant vegetation assemblage: water lilies (Najas, Nuphar, Nymphaea) and pipeworts (Eriocaulon) Potential macroinvertebrate assemblages: Crayfish (Decapoda), odonates (Odonata), bivalve (Sphaeriidae), amphipods (Amphipoda), mayflies (Ephemeroptera), true bugs (Notonectidae), caddisflies (Trichoptera), beetles (Coleoptera), and water strider (Gerris). Acid-tolerant indicators include odonates (Cordulia and Leucorrhinia, Aeshna and Ischnura). Other species characteristic of hard substrates (e.g., shale, cobble, possibly woody debris) including snail (Ferresia californica), and midges (Tribelos and Phaenopsectra). Other species characteristic of soft substrates (e.g., mud, possibly sand) including water beetles (Dytiscidae), true water bugs (Corixidae and Notonectidae). Type #3. Neutral shallow ponds. Neutral shallow ponds (Type #3) are concentrated in southeastern New Hampshire and along the Connecticut River (Figure 9), and are the least common of all lake acres in the state (Table 12). They are most common in the lowest elevations (Table 13) and are somewhat evenly spread across bedrock types, although they are most prevalent in deep coarse sediments, and relatively more Type #3 ponds occur on calcareous bedrock types (Table 15). Like Type#1, they are primarily found higher in watersheds in which they occur (Table 14), and there are more non-connected neutral shallow ponds than any other lake type. They are primarily neutral, with moderate pHs and higher ANCs, potentially more productive and species-rich biota, with occasional alkaline, nutrient rich ponds. They will exhibit relatively higher biological productivity (meso- and eutrophic) than other types, and they likely have a range of substrates, including silty, sandy, and rocky substrates. These ponds are generally non-stratified in summer, and may freeze to nearly the bottom of the pond (depending on its size and overall depth) in winter (monomictic). This type is also likely characterized by relatively higher species richness. Ponds may have pondshore emergent marshes with higher species richness. Depending on stocking, fish assemblages are likely to include both warmer and cold species, such as chain pickerel, golden shiner, pumpkinseed, large- and small-mouth bass, with lesser abundances of white sucker, longnose sucker, brown bullhead, yellow perch, fallfish, walleye, northern pike. Based on information compiled for lake types in Maine (Hunt 2002) and Vermont (Langdon et al 1998), biological characteristics of neutral shallow ponds may also include: Dominant vegetation assemblage: Floating-leaved aquatics dominate including Potamogeton species, Nuphar variegata, Nymphaea odorata, Brasenia schreberi. Associated plants may include Isoetes echinospora, Utricularia species and submersed Glyceria borealis. Other species may include Scapania sp. and macroscopic green algae. Dominated by low-growing, submergent rosette/mat-forming plants with flowers emergent at low water. Eriocaulon aquaticum is consistently dominant. Common associates are Lobelia dortmanna, Sagittaria graminea, Nymphoides cordata, Myriophyllum tenellum, Potamogeton confervoides, Isoetes echinospora. Vallisneria americana is common but should not occur at high percent cover. Macroinverts: Sowbugs (Isopoda), amphipods (Amphipoda), and Ramshorn Snail (Planorbidae). Acid-intolerant indicators include odonates (Gomphus and Basiaeschna), as well as other odonates (Aeshna and Ischnura) Possibly species characteristic of hard substrates (e.g., shale, cobble, possibly woody debris) and possibly macrophyte beds associated with coarser substrates including snails (Amnicola limosa, Physidae) and mayfly (Stenonoma). Type #4. Acidic shallow lakes. Acidic shallow lakes are distributed state-wide, and are the most abundant lake type in the state, with 707 occurrences, and account for nearly 20% of all lake acres in the state (Table 12). They are most common in middle elevations (Table 13) and are predominantly associated with acidic bedrock types (Table 15) like Type #1 and #2. They are found both high and in the middle of their watersheds (Table 14), and relatively few are non-connected to a river system (TABLE 14). They are primarily acidic, but not likely to have pHs low enough to exclude fish. They share other characteristics with Types #1 and #2, such as low biological productivity (oligotrophic) and relatively shallow depths. They likely have relatively low pH and low ANC, with a range of substrates, including sandy and rocky substrates, as well as mucky peat. These lakes are more likely to be stratified in summer (dimictic), and will freeze each winter as well. With slightly deeper water columns, they likely provide more habitat for cold water fisheries than acidic ponds. This type is likely characterized by moderate species richness. Fish assemblages are likely similar to be dominated by chain pickerel, golden shiner, pumpkinseed, largemouth bass, with brown bullhead, yellow perch, white sucker as well. Cold water fisher in smaller lakes will also likely include brown bullhead in the most acidic ponds, and golden shiner, rainbow smelt, burbot and possibly stocked trout (brook, brown, rainbow) in others. Larger lakes will have more characteristic cold water species, such as lake trout (Salvelinus namaycush), rainbow smelt, burbot, landlocked Atlantic salmon, and brook trout. Common exotic species include rainbow trout and brown trout. Based on information compiled for lake types in Maine (Hunt 2002) and Vermont (Langdon et al 1998), biological characteristics of acidic shallow lakes may also include: Dominant vegetation assemblage: water lilies (Najas, Nuphar, Nymphaea) and pipeworts (Eriocaulon) Floating-leaved aquatics dominate including Potamogeton species, Nuphar variegata, Nymphaea odorata, Brasenia schreberi. Associated plants may include Isoetes echinospora, Utricularia species and submersed Glyceria borealis. Other species may include Scapania sp. and macroscopic green algae. Dominated by low-growing, submergent rosette/mat-forming plants with flowers emergent at low water. Eriocaulon aquaticum is consistently dominant. Common associates are Lobelia dortmanna, Sagittaria graminea, Nymphoides cordata, Myriophyllum tenellum, Potamogeton confervoides, Isoetes echinospora. Vallisneria americana is common but should not occur at high percent cover. Macroinverts: Cayfish (Decapoda), odonates (Odonata), bivalve (Sphaeriidae), amphipods (Amphipoda), mayflies (Ephemeroptera), true bugs (Notonectidae), caddisflies (Trichoptera), beetles (Coleoptera), and water strider (Gerris). Acid-tolerant indicators include odonates (Cordulia and Leucorrhinia), with (Aeshna and Ischnura). species characteristic of hard substrates (e.g., shale, cobble, possibly woody debris) and possibly macrophyte beds associated with coarser substrates including snails (Amnicola limosa, Physidae) and mayfly (Stenonoma). species characteristic of soft substrates (e.g., mud, possibly sand) including water beetles (Dytiscidae), true water bugs (Corixidae and Notonectidae). Type #5. Neutral shallow lakes. Neutral shallow lakes (Type #5) are similar to neutral shallow ponds in many respects. For example, they are also concentrated in southeastern New Hampshire and along the Connecticut River (Figure 9), and are the second least common lake type in the state (Table 12). They are most common in the lowest elevations (Table 13) and are somewhat evenly spread across bedrock types, although they are most prevalent in deep coarse sediments (Table 15). They are found both high and in the middle of their watersheds, and only one example is non-connected to a river system (Table 14). They are primarily neutral, with moderate pHs and higher ANCs, potentially more productive and species-rich biota, with occasional alkaline, nutrient rich ponds. They will exhibit relatively higher biological productivity (meso- and eutrophic) than other types, and they likely have a range of substrates, including silty, sandy, and rocky substrates. These lakes are more likely to be stratified in summer (dimictic), and will freeze each winter as well, however monomictic lakes (unstratified) likely occur in the smaller sizes. With slightly deeper water columns, they likely provide more habitat for cold water fisheries than acidic ponds. This type is also likely characterized by relatively higher species richness. Shallow lakes may have lakeshore emergent marshes with higher species richness. Depending on stocking, fish assemblages are likely to include both warmer and cold species, such as chain pickerel, golden shiner, pumpkinseed, large- and small-mouth bass, with lesser abundances of white sucker, longnose sucker, brown bullhead, yellow perch, fallfish, walleye, northern pike. Based on information compiled for lake types in Maine (Hunt 2002) and Vermont (Langdon et al 1998), biological characteristics of neutral shallow ponds may also include: Dominant vegetation assemblage: Floating-leaved aquatics dominate including Potamogeton species, Nuphar variegata, Nymphaea odorata, Brasenia schreberi. Associated plants may include Isoetes echinospora, Utricularia species and submersed Glyceria borealis. Other species may include Scapania sp. and macroscopic green algae. Dominated by low-growing, submergent rosette/mat-forming plants with flowers emergent at low water. Eriocaulon aquaticum is consistently dominant. Common associates are Lobelia dortmanna, Sagittaria graminea, Nymphoides cordata, Myriophyllum tenellum, Potamogeton confervoides, Isoetes echinospora. Vallisneria americana is common but should not occur at high percent cover. Macroinverts: Sowbugs (Isopoda), amphipods (Amphipoda), and Ramshorn Snail (Planorbidae). Acid-intolerant indicators include odonates (Gomphus and Basiaeschna), as well as other odonates (Aeshna and Ischnura) Possibly species characteristic of hard substrates (e.g., shale, cobble, possibly woody debris) and possibly macrophyte beds associated with coarser substrates including snails (Amnicola limosa, Physidae) and mayfly (Stenonoma). Type #6. Acidic deep lakes. Acidic deep lakes are also distributed state-wide, and while not the most abundant lake type in the state, with only 169 occurrences, they account for nearly 67% of all lake acres in the state (Table 12). These lakes provide the majority of lake habitat state-wide, and include most, if not all, of our regionally famous and economically important Lakes Region lakes, as well as Lake Sunapee, Nubanusit Lake, and Newfound Lake to name a few. They are most common in middle elevations (Table 13) and are predominantly associated with acidic bedrock types (Table 15). They are spread across their watersheds with the largest lakes sitting within some of the larger watersheds (Table 14). They are primarily acidic, but not likely to have pHs low enough to exclude fish. They share other characteristics with Types #1 and #2, such as low biological productivity (oligotrophic) but all are deeper than 30 feet, with some providing extensive deep water habitats. They likely have relatively low pH and low ANC, with a range of substrates, including sandy and rocky substrates, as well as mucky peat. These lakes are stratified in summer (dimictic), and will freeze each winter as well. With extensive deeper water columns, they provide an abundance of habitat for cold water fisheries. This type is characterized by high-moderate species richness. Fish assemblages are likely similar to be dominated by chain pickerel, golden shiner, pumpkinseed, largemouth bass, with brown bullhead, yellow perch, white sucker as well. Cold water fisher in smaller lakes will also likely include brown bullhead in the most acidic lakes, and golden shiner, rainbow smelt, burbot and possibly stocked trout (brook, brown, rainbow) in others. Larger lakes will have more characteristic cold water species, such as lake trout (Salvelinus namaycush), rainbow smelt, burbot, landlocked Atlantic salmon, and brook trout. Common exotic species include rainbow trout and brown trout. The largest lakes in New Hampshire (Lake Winnipesaukee, Squam Lake, Lake Wentworth), may have lake whitefish, round whitefish (Lake Winnipesaukee, Newfound Lake), and landlocked Atlantic Salmon. In addition, lake trout has been introduced into First and Second Connecticut Lakes, Squam, Winnipesaukee, Squam, Winnisquam, Newfound, and Crystal Lakes. Based on information compiled for lake types in Maine (Hunt 2002) and Vermont (Langdon et al 1998), biological characteristics of acidic deep lakes may also include: Dominant vegetation assemblage: water lilies (Najas, Nuphar, Nymphaea) and pipeworts (Eriocaulon) Floating-leaved aquatics dominate including Potamogeton species, Nuphar variegata, Nymphaea odorata, Brasenia schreberi. Associated plants may include Isoetes echinospora, Utricularia species and submersed Glyceria borealis. Other species may include Scapania sp. and macroscopic green algae. Dominated by low-growing, submergent rosette/mat-forming plants with flowers emergent at low water. Eriocaulon aquaticum is consistently dominant. Common associates are Lobelia dortmanna, Sagittaria graminea, Nymphoides cordata, Myriophyllum tenellum, Potamogeton confervoides, Isoetes echinospora. Vallisneria americana is common but should not occur at high percent cover. Macroinverts: Cayfish (Decapoda), odonates (Odonata), bivalve (Sphaeriidae), amphipods (Amphipoda), mayflies (Ephemeroptera), true bugs (Notonectidae), caddisflies (Trichoptera), beetles (Coleoptera), and water strider (Gerris). Acid-tolerant indicators include odonates (Cordulia and Leucorrhinia), with (Aeshna and Ischnura). Species characteristic of hard substrates (e.g., shale, cobble, possibly woody debris) and possibly macrophyte beds associated with coarser substrates including snails (Amnicola limosa, Physidae) and mayfly (Stenonoma). Species characteristic of soft substrates (e.g., mud, possibly sand) including water beetles (Dytiscidae), true water bugs (Corixidae and Notonectidae). Type #7. Neutral deep lake. There are only 26 total occurrences of this lake type in New Hampshire, and in addition to being the rarest type, this type only accounts for 4.5% of all lake acres (Table 12). They are similar to neutral shallow lakes in many respects. For example, they are also concentrated in southeastern New Hampshire and along the Connecticut River (Figure 9). They are most common in the lowest elevations (Table 13) and are somewhat evenly spread across bedrock types, and are likely most influenced by runoff, groundwater, and/or sedimentary influences (Table 15). They are primarily found in watersheds between 2-30 square miles, with only one example is non-connected to a river system (Table 14). They are primarily neutral, with moderate pHs and higher ANCs, potentially more productive and species-rich biota, with occasional alkaline, nutrient rich ponds. They will exhibit relatively higher biological productivity (meso- and eutrophic) than other types, and they likely have a range of substrates, including silty, sandy, and rocky substrates. These lakes are more likely to be stratified in summer (dimictic), and will freeze each winter as well, however monomictic lakes (unstratified) likely occur in the smaller sizes. With slightly deeper water columns, they likely provide more habitat for cold water fisheries than acidic ponds. This type is also likely characterized by relatively higher species richness. They may have lakeshore emergent marshes with higher species richness. Depending on stocking, fish assemblages are likely to include both warmer and cold species, such as chain pickerel, golden shiner, pumpkinseed, large- and small-mouth bass, with lesser abundances of white sucker, longnose sucker, brown bullhead, yellow perch, fallfish, walleye, northern pike. Based on information compiled for lake types in Maine (Hunt 2002) and Vermont (Langdon et al 1998), biological characteristics of neutral shallow ponds may also include: Dominant vegetation assemblage: Floating-leaved aquatics dominate including Potamogeton species, Nuphar variegata, Nymphaea odorata, Brasenia schreberi. Associated plants may include Isoetes echinospora, Utricularia species and submersed Glyceria borealis. Other species may include Scapania sp. and macroscopic green algae. Dominated by low-growing, submergent rosette/mat-forming plants with flowers emergent at low water. Eriocaulon aquaticum is consistently dominant. Common associates are Lobelia dortmanna, Sagittaria graminea, Nymphoides cordata, Myriophyllum tenellum, Potamogeton confervoides, Isoetes echinospora. Vallisneria americana is common but should not occur at high percent cover. Macroinverts: Sowbugs (Isopoda), amphipods (Amphipoda), and Ramshorn Snail (Planorbidae). Acid-intolerant indicators include odonates (Gomphus and Basiaeschna), as well as other odonates (Aeshna and Ischnura) Possibly species characteristic of hard substrates (e.g., shale, cobble, possibly woody debris) and possibly macrophyte beds associated with coarser substrates including snails (Amnicola limosa, Physidae) and mayfly (Stenonoma). Similar to Deep Acidic, more productive, meso-, eutrophic Macroinverts: Acid-intolerant indicators include odonates (Gomphus and Basiaeschna). Dominant widespread indicators include odonates (Aeshna and Ischnura). Not as well correlated with trophic state as with acidity/alkalinity. In areas of low fish predation, leeches abundant. sowbugs (Isopoda), amphipods (Amphipoda), and Ramshorn Snail (Planorbidae). (Mesotrophic) , leech (Hirundinea). low oxygen-tolerant, low light-tolerant organisms including mayfly (Hexagenia) and bivalve (Pisidium). Table 13: Biological System Types by elevation zone. Type #Tidal Elevation: 0-20ftLow Elevation: 20-800ftModerate Elevation: 800-1700ftHigh Elevation: 1700-2500ftVery High Elevation: 2500ft+Grand Total185458379589245918521567033541582115434314361283706571743082196857861697222226Grand Total4215541196113172,922 Table 14: Biological System Types by connectivity class. Type #Potential seepage lakes: not connected to a riverUpper headwater lakes, drainage area < 2 sq.mi. Headwater lakes, drainage area 2 - 30 sq.mi.River influenced lakes, drainage area >30 sq.mi.Total126445107115892674531381267038033611215543418413245307065112383122196365772416971914226Grand Total19618447761062,922 Table 15: Biological System Types by 200m shoreline buffer area dominant geology class (local geologic setting). Type #Acidic GraniticAcidic Sed. / Metased.Mafic/ Interm. GraniticModerately Cal. Sed. / Metased.Calc. Sed. / Metased.Deep Coarse Sediment SurficialDeep Fine Sediment SurficialGrand Total12302921461442589236814816114113106703111133403941417554343422771116158170656969151213815219695511157169761241326Total122198211190114041032,922Lake condition background Six major classes of stressors are largely responsible for the degradation of lake ecosystems: nutrient enrichment, riparian disturbance, non-native species, hydrological changes, acidification, and contamination. (National Research Council 1992, Whittier et al. 2002) The impacts of these stressors on lake ecosystems vary by the type of disturbance and specific characteristics of the affected lake, but some responses are common across several types of stressors. For example, disturbed lakes tend to lose sensitive native species, support simplified food webs dominated by tolerant and non-native-invasive species, and experience frequent nuisance conditions such as blue-green algae blooms (National Research Council 1992). Although many of the above types of degradation are hard to measure across entire populations of lakes, some surrogate related variables can be measured comprehensively in GIS as a proxy. We suggest using a limited number of GIS attributes to evaluate the level of ecological, shoreline, and landscape integrity of lake ecosystems. Below we also briefly describe general impacts of these different stressors. Watershed and lakeshore development Watershed impervious surface is recognized as being highly correlated with Index of Biotic Integrity measures for aquatic ecosystems (Center for Watershed Protection 2003). Significant negative impacts on aquatic ecosystems are noted when watersheds reach more than 10% imperviousness and stream are considered unsupporting when their watershed gets > 25% impervious (Center for Watershed Protection 2003). Although most research on the effects of watershed impervious cover on aquatic ecosystems has focused on riverine systems, lakes are expected to experience negative impacts similar to riverine waters (Center for Watershed Protection 2003). For example, lakes in watersheds with higher impervious surfaces are expected to be negatively impacted by increased nutrients, sediment loading, higher bacteria and pathogen levels, metals, and chlorides. Of these pollutants, phosphorus is regarded as the highest lake management concern. Urban and residential land uses are known to yield a higher export of phosphorus and nitrogen per unit area to waterbodies than forested or agricultural cover types (Downing and McCauley 1992). Because of increased nutrient levels, waterbodies associated with highly urban and residential landscapes are likely to experience and an upward shift in trophic status (Center for Watershed Protection 2003.) Although precise thresholds that trigger eutrophication are unique to each lake given its internal geometry, current phosphorous concentration, and the size and types of development in its contributing watershed, Siver et al. (1999) note in a study of Connecticut lakes that the majority of lakes in watersheds with over 80% forest cover have not significantly changed in trophic state, pH, conductivity, or total nitrogen in the last hundred years, yielding relatively intact algal communities as compared to historical core samples (Siver et al. 1999). Lakes in watersheds with 25% or more residential development experienced the greatest amount of change (Siver et al. 1999). The Environmental Monitoring and Assessment Program (EMAP) study of lakes in the northeast U.S. estimates that about half of natural lakes would be expected to be oligotrophic and about half should be mesotrophic, with naturally eutrophic lakes extremely rare. Results of the EMAP study of the lake current condition yielded 22% of lakes in the northeast as eutrophic or hypereutropic, with 17% of upland lakes being eutrophic and 39% of lowland lakes being eutrophic. Whittier suggests that most of these eutrophic lakes are the result of cultural eutrophication due to development which occurred within the last 150 years. In addition to cumulative watershed impacts, local shoreline development also contributes to cultural eutrophication through both impervious cover runoff and septic system leeching. Most lakefront developments are serviced by septic systems because of their seasonal use or distance from wastewater treatment plants. Because of their proximity to lakes, septic systems can become a source of subsurface phosphorus seepage to the lake. Poorly functioning waterfront septic systems have been shown to be an important source of phosphorus and nitrogen in a wide range of lake systems (Harper 1995, Robertson and Harman 1999, Arnade 1999). Although the relative impact of shoreline and shoreline buffer development vs. watershed development to overall lake biotic integrity has not been well studied (Whittier et al. 2002), shoreline development has been associated with many other negative impacts on lake ecosystems. For example, a number of studies have noted declining fish abundance or diversity with increasing shoreline development (Hinch and Collins 1993, Hinch et al 1994, Bryan and Scranecchia 1992). Fish foraging and spawning have also been shown to decline as a direct function of cottage or home density around the lakeshore (Engel and Pederson 1998). Alteration of the littoral habitat is particularly noted as a critical concern because most fish species spend at least part of the lifecycle in the littoral zone of the shoreline. Maintaining shade, leaf litter, woody debris, complexity of emergent and submergent plants, and water quality components of the littoral habitat becomes increasingly difficult with shoreline development. Many birds, such as eagles, loons, and songbirds have also been found to avoid developed lakes. Whether due to loss of nesting sites, changes in prey base, or lack of tolerance for noise or other disturbances, their avoidance has been noted at a relatively low rate of cottage development (Johnson and Brown 1990, Heimberger et al 1983). Similar relationships have been discovered for amphibians and reptiles which utilize the lakeshore to bask, feed, nest, and overwinter (Engel and Pederson 1998). Since lakefront property is so desirable, it is quite common to have intense lakefront development in otherwise lightly developed watersheds (Capiella and Schueler 2004). These shorelines are often increasingly developed as additional owners build summer homes or cottages and seek both good access to the water and an unobstructed view of the lake. The greatest density of homes is usually found within 500 ft (150m) of the lake and less density further away (Capiella and Schueler 2004). Nonindigenous species Increasing numbers and abundances of nonindigenous aquatic species were associated with higher lake development levels in the EMAP study of northeastern lakes (Whittier and Kincaid 1999). Nonindigenous aquatic species are defined as a species that enters a body of water or aquatic ecosystem outside of its historic or native range (USGS 2003). Introductions of non-native species have been increasingly recognized as resulting in local extirpations and even extinction of native species, as well as regional and global homogenization of species assemblages which results in a reduced broad-scale biodiversity (Whittier et al. 2002). Nonindigenous species have a number of negative impacts such as competition with indigenous species for food and habitat, reduction of natives by predation, transmission of diseases or parasites, hybridization, and habitat alteration (USGS 2003). Unlike chemical pollutants that can be eliminated at their source or physical habitats that can be restored, species introductions are usually extremely difficult if not impossible to undo. Natural enemies are often lacking and once alien species are established in a new environment, they are often capable of reproducing and dispersing far beyond the point of origin. Native lake fish assemblages of the Northeast are naturally species poor, in comparison with those of other areas of the eastern U.S. This ecological simplicity makes native northeastern lake communities highly vulnerable to perturbations from altered community composition due to introduced species (Abell et al. 2000). Due to the recent glaciation and post-glacial freshwater recolonization pathways which were cut off from the species rich glacial refuges in the Mississippian drainages, naturally northeast lowland lake fish assemblages generally include only one or two sunfish, one top predator (chain pickerel), one or two bullheads, yellow perch, white sucker, and a few other species. Northeast small upland lakes naturally have brook trout and numerous minnow species (Tonn et al. 1990). Certain types of introduced species, particularly predators, are more likely to eliminate native species (Moyle & Light 1996). Strong circumstantial evidence suggests that species that are small or soft fined are at greater risk of local extirpation by introduced species. The most susceptible species in northeastern lakes are currently found in cold-water or cool water lakes. Of the non-native predators, the smallmouth bass has the greatest potential to spread further into these cooler systems because of its temperature tolerances and its appeal to sport fisheries (Whittier et al. 2002). A number of formerly common minnows, burbot, and four species of stickleback are particularly noted as declining or extirpated in many northeast lakes due to introduced fish. Brook trout appear to be intolerant of smallmouth bass in lakes and are maintained only by stocking in lakes where smallmouth bass populations have developed (Whittier et al 2002). Whittier suggests that native lake fish assemblages in the northeast have already been greatly altered due to widespread introduction of game fish and transfers of lowland fish to upland settings (Whittier et al. 1997, 1999, Whittier and Kincaid 1999). Whittiers study of the EMAP lakes found only 21% of lakes in their study of 235 lakes in the northeast had all native fish assemblages, with the majority in Maine or western New York. 15% of lakes had more introduced species than native species and introduced individuals outnumbered natives in 31% of lakes (Whittier and Kincaid 1999). Four nonindigenous centrarchids, largemouth bass, bluegill, smallmouth bass, and black crappie, are particularly noted as having been so widely introduced in the that they are now often the most common species in northeast lakes (Whittier et al. 2002). In addition to these four centrarchids, other widespread introduced fish species in New England include the bluntnose minnow, cutlips minnow, fathead minnow, pearl dace, pumpkinseed, rainbow smelt, rainbow trout, brown trout, burbot, lake trout, rock bass, round-whitefish, and trout-perch (USGS Nonindigenous Aquatic Species database http://nas.er.usgs.gov). In addition to fish, a large number of nonindigenous plants, amphibians, reptiles, mammals, mollusks, crustraceans, and sponges have also entered aquatic systems. Some of the most problematic and invasive species among these introductions include the asiatic clam, zebra mussel, purple loosestrife, common reed grass, Eurasian water-milfoil, water-chestnut, yellow iris, curly pondweed, two-leaf water-milfoil, European water-clover, Carolina fanwort, watercress, Brazilian waterweed, dotted duckweed, pond water-starwort, and hydrilla. These species have significantly altered physical and biological functions of aquatic systems. For example, the water chestnut is a highly invasive species that can out-compete native plants, choke the waterbodies it invades, and reduce oxygen levels increasing the potential for fish kills. Similarly, Eurasian water milfoil, a stringy submerged plant, can quickly proliferate and aggressively compete with native plant communities to form large dense mats that clog waterbodies. Curly pondweed, a submerged perennial, can tolerate low light and low water temperatures, making it competitively superior especially early in the season as it forms new plants under ice cover. Mid-summer die-offs of this plant may result in a critical loss of dissolved oxygen and decaying plant matter can increase water nutrients and contribute to subsequent algal blooms. Hydrological change EMAP estimated that 47% of northeastern lentic waterbodies >1 hectare (2.471 acres) were human made. Of these human lakes, 93% were impoundments and 7% were quarries. In the Uplands 69% of lakes were natural whereas in the Lowlands only 27% were natural (Whittier et al. 2002). Dams alter aquatic ecosystem structure and functioning by causing barriers to upstream and downstream migration and by setting up a series of changes upstream and downstream from the impoundment. These changes include altered flow, temperature, water clarity, and severing of terrestrial/aquatic linkages critical for maintaining the riparian and floodplain communities. As barriers, dams have clearly resulted in substantial reductions in the amount of stream and lake habitat available to diadromous species such as alewife, shad, salmon, and eel. Widely potomodromous (freshwater migratory) species such as brook trout or white suckers have also been impacted. Dams can also alter riparian, emergent, and submergent floras by disrupting the dispersal of plant species whose spores or seeds are waterborne. For example, Jansson et al. (2000) showed that adjacent impoundments in similar environmental settings developed different riparian floras because species with poor floating capacity become unevenly distributed among impoundments. Such discontinuities were not found along free-flowing systems suggesting an effect of dams on the dispersal of aquatic plants (Jansson et al 2000). In addition to barrier impacts, dams impact on the natural hydrology depends, in large part, on the management policies that determine patterns of water release from dams. Water flow may be either stabilized or destabilized, relative to the undammed situation. Severe water level fluctuations in impoundments or formerly natural lakes can negatively impact both plant and animal communities. For example, Cameron found that species richness in macrophyte communities was much lower in Flagstaff and Graham Lakes than might otherwise have been expected, presumably as a result of substantial drawdowns in these systems (Cameron, cited in Vaux 2005). Jiffry (1984) documented mortality of mussels in Lake Sebasticook as a result of lake drawdowns. Many impoundments are also former river channels which were deepened and widened. The slowed water flow in these impounded river sections now promotes a lentic species assemblage in place of the natural lotic assemblage which would otherwise exist in these former river sections. In addition to water levels and flow, impounded lakes also often experience altered temperature, oxygen, and sedimentation regimes. Being deeper, impounded lakes often stratify where previously no cool water summer habitat was available. In some cases, water releases may disrupt stratification resulting in warmer temperatures that may be too warm for the original native aquatic species. Fine sediments also settle out of the water in impoundments as dams slow the movement of incoming water and store the water. By trapping fine sediments (and associated nutrients and food resources) in impoundments, water in impoundments is unnaturally augmented with nutrients, which promotes eutrophication. Fine sediment can also more quickly accumulate on the lake bed substrate, sometimes requiring dredging of reservoirs to maintain their depth which disturbs lake bed biota. Water released downstream of the dam is thus clearer, less nutrient rich, and floodplains can end up being starved of the sediments that would normally be dispersed to the floodplain during floods (Doppelt et al. 1993). Dam impacts to lake and stream ecosystems also vary by the type of dam. Hydroelectric dams are particularly noted as a type of dam altering the natural hydrologic regime as they store water for release to meet specific energy demands that vary seasonally and throughout a 24 hour period. Daily fluctuations in energy demands usually cause operators to restrict water flow through the turbines from mid-morning through early evening. Irrigation dams store as much water as possible during the rainy season for release during the growing season. Water withdrawal from irrigation ponds and lakes can be particularly stressful to lacustrine communities in the summer. Flood control reservoirs maintain only a small permanent pool in order to maximize storage capacity in case of a flood event and often have very unnatural lacustrine communities that exist in these pools. Recreational and water supply dams usually store a certain amount of water during the rainy season to sustain reservoir capacity, and they have a variety of release management practices. Run-of-the-river dams are usually of low height and are thought to have small adverse effects on hydrology. These dams do block access to rivers and lakes and enlarge impounded areas, but because they release water at the same rate it enters the reservoir these impoundments are not usually subject to large drawdowns or other significant anthropogenic water level alterations throughout the year (Allen 1995). Lake condition methods Land use Land use around lakes was calculated using two basic landscape metrics. To capture shoreline development impacts, we assumed that most roads, septic systems, and shoreline housing development would have the majority of influence within a 200 meter buffer from the lakeshore. To capture the influence of lake-watershed or -basin land use impacts, we chose the HUC12 basin surrounding the given lake. For lakes spanning more than one HUC12, the land use in all HUC12s associated with the lake was summarized. The percentage of all types of land use as defined by the NH Land Cover Assessment 2001 dataset was calculated for each lake for a 200 meter buffer distance and the HUC12 watershed basin of each lake. Data was also summarized into the following classes: Developed Residential, commercial, or industrial (100); transportation (140); Agriculture (200s); Forested (all types of forest, 400s), Wetland (600s), Water (500), Other Natural (bedrock/vegetated 720, tundra 810); and Disturbed (Barren Distrurbed quarries other areas where the earth and vegetation have been altered or exposed, 710 and clear cut or agriculture fields reverting to forest, 790). A value for total percent natural was derived by adding the per cent Forest, Wetland, Other Natural, and Water. In addition to the above categories, a summary index of shoreline buffer land use alteration was calculated to aid in a finer ranking of lakes. For example, although many lakes have shoreline buffers of >= 90% developed, the varying land use impacts in small areas that are developed are important to consider. Some lakes may have 9% residential/commercial/ transportation development around the lake, while others have only 1% development and maybe 4% pasture land around the lake. We would expect the lake with 9% residential/commercial/ and transportation land use to be more impacted. The following index (Table 16) was developed to represent, in a coarse way, the levels of impact of different land uses in the 200m shoreline buffer local area based on the NH Land Use 2001 Data. Table 16: Land use model attributes for 200m shoreline buffer around lakes. Total Impact = sum of below classes and multipliers. 1. % Class 100: Residential, commercial, or industrial x 3 2. % Class 140: Transportation x 3 3. % Class 211: Row Crops x 2 4. % Class 212: Hay and Permanent Pasture x 1 5. % Class 221: Orchard x 1 6. % Class 710: Disturbed x 1 7. % Class 790: Clear cut, and other open reverting to forest x 1 Isolation from roads Access to a lake by road or trails is correlated with the presence of non-native species in a lake ecosystem (Silk and Ciruna 2004). Although state sponsored stocking has accounted for many introductions, nonindigenous species have also been widely introduced by private citizens seeking to create a local sport fishery, inadvertently transporting species that are attached to their boats, or through discarding of excess bait. Locals in New England have noted a long history of private introductions, explaining that it is not unlikely for fisherman to hike into even remote ponds with buckets of fish to create desired fisheries in lakes and ponds (Vaux personal communication). As a standard measure of potential introduction access, the minimum distance from a mapped road or trail to each lake or pond was calculated. It is assumed that the more difficult access is to the lake, the less likely the probability of it being stocked with nonindigenous fish. The source road and trails used included state of New Hampshire DOT roads which included state, local, and some private roads as provided by DOT road coverage9/29/04, all additional private roads from the DOT anchor road cover obtained 12/04, all additional private roads from DOT private road cover obtained 12/04, and all trails from USGS roads (class 211, 212). Dams The New Hampshire Department of Environmental Services Dams data layer from January 2004 was used in this analysis, which includes attributes and locations of 5009 dams. Dams within 100 meters of the shoreline of a lake were queried. This buffer distance was used to locate dams on or very near lakes to account for 1) spatial inconsistencies between the mapped dams and the lakes upon which they were suppose to be located and 2) the adverse effect a nearby dam might have on a lake upstream or downstream from it. The number of total dams within the 100m buffer of the lake, the distance of the lake to nearest dam, and the identification # (id) of the dam nearest to the lake was added to the lake attribute table and used in further reporting of dams by type. Conservation lands Conservation lands, like land use parameters, was calculated using two basic landscape metrics. We defined a 200 meter buffer envelope, and used the HUC12 boundary, and calculated conservation lands in each landscape area. For lakes spanning more than one HUC12, the land use in all HUC12s associated with the lake was summarized. Managed area polygons included all NH Conservation and Public Lands dataset February 2005 and additional conservation land parcels collected and coded by TNC during as of 5/1/2005. Gap status codes were reviewed by TNC staff to correct errors in existing NH Conservation and Public Lands datasets for parcels > 25 acres. No attempt was made to review GAP status codes to parcels < 25 acres and a few new parcels of unknown conservation status that were added after the completion of the TNC review of GAP status do exist in the dataset. The GAP status codes that TNC used included: Status 1: An area having permanent protection from conversion of natural land cover and a mandated management plan in operation to maintain a natural state within which disturbance events (of natural type, frequency, intensity, and legacy) are allowed to proceed without interference or are mimicked through management. Status 2: An area having permanent protection from conversion of natural land cover and a mandated management plan in operation to maintain a primarily natural state, but which may receive uses or management practices that degrade the quality of existing natural communities, including suppression of natural disturbance. Status 3: An area having permanent protection from conversion of natural land cover for the majority of the area, but subject to extractive uses of either a broad, low-intensity type (e.g., logging) or localized intense type (e.g., mining). It also confers protection to federally listed endangered and threatened species throughout the area. Census and housing density projections Change in housing density within the local census block of each lake was calculated as a measure of current and future threat. The 1960, 1990, 2020, and 2050 housing density for the census block the lake is within was obtained from the dataset by David M. Theobald "Mapping historical, current, and future housing densities in the US using Census block-groups." (2001). For lakes totally within one census block, we report the housing unit density of that block. For lakes intersecting more than one census block, we created an average housing density value for the lake by summing the total census block areas and total housing units from all census blocks intersecting the lake edge to generate an overall average housing density for that lake. Although these methods are an over simplification by spreading out the housing units per total census block area, this information provides useful information on general pressure trends throughout the state at a relatively fine scale. New Hampshire Fish and Game requested we focus on the 1990 to 2020 prediction, so we focused on reporting trends in the % of area/blocks around the lake that fell in Rural, Exurban, Suburban, or Urban categories and the rate or change in housing density over that time-frame. Information on the full time series of housing density predictions are available in the classification data table that can be linked to each lake. Housing densities are summarized into four general classes: urban, suburban, exurban, and rural. Urban densities are typically defined as areas with greater than 1,000 people per square mile (1.6 people per acre). Assuming an average of 2.5 people per housing unit, this translates to roughly 0.7 units per acre. Data was thus classified according to Theobald into the following four categories: urban housing density as greater than 1 units per acre. Suburban densities as from 0.1 to 1 units per acre. Exurban densities from 0.025 to 0.1 units per acre. Rural density as housing density below 0.025 units per acre (1 per 40 acres and more). We also looked at the rate of housing increase regardless of density class. We calculated the average rate of housing unit density increase between 1990 and 2020 (# housing units/census block). This rate will more finely depict how population density is changing over time. Lake condition results Part 1: Results across all lakes Table 17: Lake types by isolation from road (500m, .5mi, 1mi). Lake TypeBiological System TypeNearest Road < 500m (.3m)Nearest Road > 500m (.3 mile) but < .5 mileNearest Road >= .5 mile but < 1 mileNearest Road >= 1 mileGrand TotalVery acidic ponds149859266589Acidic ponds259344303670Neutral ponds3512238543Acid shallow lake, small4a57026214621Acidic shallow lake, medium4b8585Neutral shallow lake, small5a20132206Neutral shallow lake, medium5b1313Acidic deep lake, small6a6341270Acidic deep lake, medium6b8181Acidic deep lake, large6c1818Neutral deep lake, small7a88Neutral deep lake, medium7b1717Neutral deep lake, large7c11Total Lakes266015988152,922Total Percent91.05.43.00.5 Isolation from roads. No lakes > 100 acres (size class 4) are remote as defined by >500m (.3mi) from a road or trail, > .5 mi from road or trail or > 1 mile from road or trail. Of all lakes, only 9% are greater than .3 miles from a road or trail. Ponds that are shallow and very acidic are the most common type of remote pond or lake using the > .3 mile from a road parameter. Figure 10. Isolated lakes.  Land use. 23% of lakes and ponds in the state have >=10% of the shoreline 200m shoreline buffer in developed land use. Neutral-alkaline lakes generally have more shoreline development. It appears that regardless of depth or acidity, a higher percentage of the medium and large sized lakes have >10% of their shoreline buffer in development. Table 18: Lake types by percent developed in 200m shoreline buffer (developed = residential, commercial, industrial, transportation). Lake TypeBiological System Type# lakes with <10% of buffer in developed land# lakes with 10% or more of buffer in developed land% of Lakes of this type with 10% of more of buffer in developed landGrand TotalVery acidic ponds15177212589Acidic ponds252914121670Neutral ponds332621740543Acid shallow lake, small4a5338814621Acidic shallow lake, medium4b63222685Neutral shallow lake, small5a1436331206Neutral shallow lake, medium5b853813Acidic deep lake, small6a57131970Acidic deep lake, medium6b59222781Acidic deep lake, large6c1263318Neutral deep lake, small7a53388Neutral deep lake, medium7b984717Neutral deep lake, large7c101Total Lakes2262660232,922 Table 19: Lake types by percent developed in 200m shoreline buffer and HUC12 Watershed. Lake TypeBiological System TypeBuffer < 10% Developed and HUC < 10% developedBuffer < 10% developed, HUC >= 10% developedBuffer >= 10% developed, HUC < 10% developedBuffer >= 10% developed, HUC >= 10% developedGrand TotalVery acidic ponds1513724589Acidic ponds2514134157670Neutral ponds321181115136543Acid shallow lake, small4a52380108621Acidic shallow lake, medium4b61202285Neutral shallow lake, small5a97204643206Neutral shallow lake, medium5b642113Acidic deep lake, small6a56111270Acidic deep lake, medium6b57212181Acidic deep lake, large6c125118Neutral deep lake, small7a5218Neutral deep lake, medium7b95317Neutral deep lake, large7c11Total Lakes20654551972052,922Total Percent70.715.66.77.0 Over 70% of lakes have both shoreline buffer and local HUC12 watershed with < 10% developed. These lakes appear least impacted by development in their local buffer and larger watershed context. Seven per cent (7%) of lakes have both shoreline buffer and local HUC12 watershed with >= 10% developed. These lakes are likely to experience negative impacts from development. Figure 11. Lakes with development in 200m shoreline buffer and HUC12 watershed.  Dams. Overall 23% of lakes currently have a dam within 100m of their shoreline. By size class, the larger lakes have a higher percentage of dammed occurrences, with 63% of the lakes > 1000 acres having a dam nearby. Other types of lakes with high percentage of dammed examples include medium sized 100-1000 acre shallow lakes (65% dammed) and medium sized 100-1000 acre deep neutral alkaline lakes (60% dammed). Within a size class of lakes, deep lakes appear to be more dammed, such as small deep acidic lakes where only 57% are undammed vs. small shallow acidic lakes where 70% are undammed. Table 20: Lake types by presence or absence of dams. Lake TypeBiological System TypeLakes without dams% of Lakes without damsGrand TotalVery acidic ponds151286.9589Acidic ponds256183.7670Neutral ponds345884.3543Acid shallow lake, small4a43670.2621Acidic shallow lake, medium4b3035.385Neutral shallow lake, small5a13867.0206Neutral shallow lake, medium5b753.813Acidic deep lake, small6a4057.170Acidic deep lake, medium6b4049.481Acidic deep lake, large6c738.918Neutral deep lake, small7a562.58Neutral deep lake, medium7b741.217Neutral deep lake, large7c00.01Total Lakes22411Total Percent76.72,922 Table 21: Lake types by dam type. Lake TypeBiological System Type# lakes with Dams of Type Agriculture, Conservation, Fire Protection, Detention, Lagoon, or Mill# lakes with Dam of Type: Recreation# lakes with Dam of Type: Flood Control# lakes with Dam of Type: Hydro electric# lakes with Dam of Type: Water SupplyVery acidic ponds12645321Acidic ponds23467215Neutral ponds3265612Acid shallow lake, small4a161461166Acidic shallow lake, medium4b441433Neutral shallow lake, small5a114719Neutral shallow lake, medium5b411Acidic deep lake, small6a52212Acidic deep lake, medium6b133142Acidic deep lake, large6c2612Neutral deep lake, small7a3Neutral deep lake, medium7b10Neutral deep lake, large7c1Total Lakes125480222232Total Per cent of dammed lakes18.972.63.33.34.8 Recreational dams are the most common types of dams with 73% of dammed lakes having a recreational dam. 11% of dams are flood control, hydroelectric, and water supply dammed lakes which are expected to be larger dams and actively managed. 19% of dams are other types of dams that are expected to be primarily small. Figure 12: Dammed lakes by dam type.  Conservation land. Forty four percent of lakes have some conservation land in their shoreline buffer. 31% of lakes have 1-80% conservation land in their buffers. Only 12% of lakes have a large percentage (80%+) of protected land in their shoreline buffer. Nine percent of these lakes have shoreline buffers in over 80% conservation land of any Gap Status. Only 3% of lakes include 80% or more of the buffer area in Gap status 1 or 2, a high level of permanent protection. Table 22: Lake types by percent conservation land in 200m shoreline buffer. Lake TypeBiological System TypeZero > 0 but < 80% in GAP Status 1, 2, 3, or 9>= 80% in GAP Status 1, 2, 3, or 9>= 80% in GAP Status 1, 2Grand TotalVery acidic ponds13831126232589Acidic ponds24041736825670Neutral ponds3335187165543Acid shallow lake, small4a3172107717621Acidic shallow lake, medium4b304311185Neutral shallow lake, small5a11177162206Neutral shallow lake, medium5b75113Acidic deep lake, small6a36249170Acidic deep lake, medium6b26459181Acidic deep lake, large6c214218Neutral deep lake, small7a3328Neutral deep lake, medium7b51217Neutral deep lake, large7c011Total Lakes1659906273842,922Total Percent56.831.09.32.9100.0 Figure 13: Lakes with >=80% conservation land within 200m shoreline buffer.  Census and housing density projections. Just over 16% of lakes are expected to change in the overall housing unit density in their surrounding census blocks. Of these lakes, the most common change is from Rural to Exurban with 9% of lakes in the state expected to undergo this change. One can also review the rate of housing unit density increase per census blocks of the lake over the time period 1990-2020. The rate of housing density increase is shown below (Figure 15). Table 23: Lake types by census blocks housing density class in 1990 and in 2020. Codes reflect change over time. For example, RR means the status starts rural and stays rural. RE means the status changes from Rural to Exurban, or one step more developed. R=Rural, E=Exurban, S=Suburban, U=Urban. Lake TypeBiological System TypeRRREEEESSSSUUUGrand TotalVery acidic ponds1125953227391589Acidic ponds21005437143984670Neutral ponds36414122692371126543Acid shallow lake, small4a83803633065621Acidic shallow lake, medium4b885051485Neutral shallow lake, small5a21456199736206Neutral shallow lake, medium5b1131713Acidic deep lake, small6a1710345470Acidic deep lake, medium6b12449411181Acidic deep lake, large6c462618Neutral deep lake, small7a2211118Neutral deep lake, medium7b231217Neutral deep lake, large7c11Total Lakes439272137818959114392,922Total Percent15.09.347.26.520.20.51.3100.0 Figure 14. Lakes expected to experience housing density change from 1990-2020. Codes reflect change over time. For example, RR means the status starts rural and stays rural. RE means the status changes from Rural to Exurban, or one step more developed. R=Rural, E=Exurban, S=Suburban, U=Urban.  Figure 15: Lakes embedded in a census block by rate of housing unit increase 1990-2020.  Part 2: Most natural and least natural lakes As an exercise inform how to prioritize conservation action, we developed a preliminary sorting model. The quantitative data developed for this project can be used to highlight lakes meeting relatively intact, or more natural conditions relative to lakes that are more impacted or least natural. To highlight lakes in these two groups we queried lakes for combinations of (1) remoteness; (2) buffer condition; (3) HUC12 watershed condition, and; (4) presence of dams. While this combination of parameters and the thresholds we chose requires additional review, we suggest these as a starting point. As the science becomes more definitive, and as we add in biological data, managers should refine the model parameters and results over time. Most natural lakes In order to identify most natural lakes, we started by sorting Lake Types based on the following thresholds: 1. Lakes with >90% natural land cover in the 200m shoreline buffer AND no dams 2. Level of remoteness 3. Presence within a HUC12 watershed >= 90% natural cover Table 24. Most natural lakes by lake type. Data represent only those lakes within HUC12 watersheds with =>90% natural land cover that also have >=90% natural cover buffers and no dams. Lake TypeBiological System TypeNearest road or trail is >1 mileNearest road or trail is .5-1 mileNearest road 500m (.3mile)-.5 mileNearest road <500m (.3mile)Total Lakes% All LakesVery acidic ponds16203514720835.3Acidic ponds2315118311216.7Neutral ponds31121234.2Acid shallow lake, small4a48610812620.3Acidic shallow lake, medium4b333.5Neutral shallow lake, small5a12125.8Neutral shallow lake, medium5b00.0Acidic deep lake, small6a1131420.0Acidic deep lake, medium6b889.9Acidic deep lake, large6c3316.7Neutral deep lake, small7a1112.5Neutral deep lake, medium7b115.9Neutral deep lake, large7c00.0total lakes13455340051117.5total percent2.58.810.478.3100 29% of all lakes (865 of 2,922) occur in HUC12 watersheds with land cover >= 90% natural. Of these 865 lakes, 59% (511) (or 18% of all lakes in the state) also had intact shoreline buffers >= 90% natural cover and no dams. These 511 lakes again represented 11 of the 13 lake types and varied in their isolation from roads as measured above. They are less spatially distributed across the state than lakes with buffers >= 90% natural and no dams, regardless of watershed context, due to the more limitation distribution of very intact watersheds. Figure 16: Most Natural lakes by absence of dams, natural land cover, and proximity to roads. Least natural lakes In order to identify least natural lakes, we started by sorting Lake Types based on the following thresholds: 1. Lakes with <75% natural land cover in the 200m shoreline buffer AND road or trail within 500m 2. Presence of Dams 3. Presence within a HUC12 watershed <80% natural cover Table 25: Least natural lakes by lake type. Data represent only those lakes within HUC12 watersheds with <80% natural land cover. Lake TypeBiological System Typeroad/ trail within 500m, buffer natural < 50%, no damroad/ trail within 500m, buffer natural < 50%, damroad/ trail within 500m, buffer natural 50-75%, no damroad/ trail within 500m, buffer natural 50-75%, damTotal Lakes% All LakesVery acidic ponds1516152.5Acidic ponds2136119395.8Neutral ponds387261021523042.4Acid shallow lake, small4a3145223.5Acidic shallow lake, medium4b11244.7Neutral shallow lake, small5a19826187134.5Neutral shallow lake, medium5b3323.1Acidic deep lake, small6a1234.3Acidic deep lake, medium6b1122.5Acidic deep lake, large6c115.6Neutral deep lake, small7a11225.0Neutral deep lake, medium7b113529.4Neutral deep lake, large7c0.0total lakes132451685239713.6total percent33.211.342.313.1100.0 397 or 49% of the 817 lakes were also in watersheds of <80% natural cover. We would expect these lakes that also have local buffers with <50% natural cover (177) to be significantly impacted, with those dammed examples (45) being the least natural. Lakes with 50-75% natural cover are also likely to be heavily impacted, again with those dam examples being potentially more severely impacted. Figure 17: Least Natural lakes by presence of dams, natural land cover, and proximity to roads.  Top 10 lakes In addition to the above most natural and least natural general categories, we generated a smaller set of potentially most intact or least intact lakes within each lake type. This summary could be useful to focus management and potential protection or restoration activity on a set of representative lakes. The most and least intact ten lakes in each of the 13 types were identified by first ranking lakes using a similar method from above, and adding in the finer scale 200 meter shoreline buffer condition model by land cover (Table 16), as below. We assessed the results and, based on subjective opinion, occasionally added or removed lakes from the list. For example, if a lake was in excellent condition based on the categories below but also had a single dam, it may have been upgraded compared to a lake in poorer condition with no dam. Top 10 Lake condition categories: 1. 13 Lake Types 2. Local Condition (200m shoreline buffer) Categories 1buffer >=90% natural/ no dams/ nearest road or trail is >1 mi2buffer >=90% natural/ no dams/ nearest road or trail is .5 1 mile3buffer >=90% natural/ no dams/ nearest road 500m -.5 mile4buffer >= 90% natural/no dams/ nearest road < 500m5buffer < 90% natural/no dams/ any remoteness6Dams 3. Watershed (HUC12) Condition Categories 1HUC12 Watershed Very Intact: >= 90% Natural Cover2HUC12 Watershed Lightly Impacted: 80-90% Natural Cover and <10% developed 3HUC12 Watershed Impacted: All Others 4. By Buffer Land Use Impact Index (continuous variable) (From Table 16). 1. % Class 100: Residential, commercial, or industrial x 3 2. % Class 140: Transportation x 3 3. % Class 211: Row Crops x 2 4. % Class 212: Hay and Permanent Pasture x 1 5. % Class 221: Orchard x 1 6. % Class 710: Disturbed x 1 7. % Class 790: Clear cut, and other open reverting to forest x 1 The most intact lakes usually shared Local Condition Categories 1-4 within Watersheds of Condition Summary Categories 1 or 2. If too many lakes shared these characteristics, the very best lakes were limited to those sharing the best combined local condition category, watershed land use and buffer land use index. Lakes with dams were never chosen in the best category, even if they had more intact buffers and watersheds. In some cases lakes with more intact buffers but in more developed watersheds were chosen over lakes with much less intact buffers that occurred in watersheds that were more developed. The least intact lakes usually fell in Watershed Condition Summary Category 3, were dammed, and had high buffer land use impact indices. In some cases, undammed lakes in Watershed Summary Category 3 had extremely high buffer land use impact indices. Where these examples occurred and were significantly greater in buffer land use impact than dammed lakes within Watershed Summary Category 3, these highly developed yet undammed lakes were also selected. Thus, in many lake types the 1-5 least intact examples were dammed and the next 5-10 least intact lakes had similar land use and watershed impacts buffer indices but were undammed. We felt these were likely more impacted than dammed lakes with very intact buffers and watersheds. (See Appendix I for a list of lakes per type. For lake types with <20 examples in the state it was not possible to select 10 lakes and examples are instead sorted from least and most intact accordingly). Figure 18: Top 10 lake occurrences for each lake type.  Table 26. Statistics of top 10 lakes by lake type and conservation status within 200m shoreline buffer. Lake TypeBiological System TypeZero1-80%>= 80% in GAP 1,2, 3, 9>= 80% in GAP 1 or 2Grand TotalVery acidic ponds114510Acidic ponds2323210Neutral ponds362210Acid shallow lake, small4a214310Acidic shallow lake, medium4b17210Neutral shallow lake, small5a22610Neutral shallow lake, medium5b4217Acidic deep lake, small6a423110Acidic deep lake, medium6b43310Acidic deep lake, large6c27110Neutral deep lake, small7a224Neutral deep lake, medium7b358total lakes34333111109total percent31.230.328.410.1 Table 27. Statistics of Top 10 lakes by lake type and housing density projected changed 1990-2020 within census blocks. Codes reflect change over time. For example, RR means the status starts rural and stays rural. RE means the status changes from Rural to Exurban, or one step more developed. R=Rural, E=Exurban, S=Suburban, U=Urban. Lake TypeBiological System TypeRRREEEESSSGrand TotalVery acidic ponds11010Acidic ponds28210Neutral ponds3324110Acid shallow lake, small4a 61310Acidic shallow lake, medium4b 22610Neutral shallow lake, small5a 51410Neutral shallow lake, medium5b 13127Acidic deep lake, small6a 42410Acidic deep lake, medium6b 41510Acidic deep lake, large6c 44210Neutral deep lake, small7a 2114Neutral deep lake, medium7b 2338total lakes50113648109total percent45.910.133.03.77.3 Lake condition discussion The quantitative condition data developed for this project measured GIS variables related to remoteness, riparian intactness, watershed intactness, and dam impacts to lakes. These variables can be used to highlight trends across the state and highlight groups of lakes that may be more impacted by particular kinds of stressors (e.g. dams or agriculture). We used the data to conduct a set of multi-variable queries to highlight groups of lakes sharing characteristics of high or low shoreline and basin integrity that the literature indicate are correlated with specific lake threats. We used quantitative data on current conservation lands within the shoreline buffer and predictions of future development pressure to suggest which lakes are protected from further development and/or most threatened with further development. The queries we present are a good place to start when considering priorities for future conservation action. However, in the absence of specific biological data and water quality information for specific lakes, our analyses should be used with caution. For example, lakes may have characteristics we could not model that may indicate much better (or worse) condition. In addition, research is needed to more fully understand if the weighting of variables at local and watershed scales truly define most ecologically significant thresholds, and determine how these variables interact to influence lakes across biological gradients. It will also be important for a future lake conservation strategy to assess additional aspects of lake uniqueness. These factors might include lakes with breeding native brook trout or lake trout, lakes supporting populations of rare species, or lakes surrounded by exemplary natural terrestrial communities. A statewide conservation strategy should also seek to distribute conserved lake examples across the range of their distribution in the state to ensure protection across geographical gradients. Our goal was to raise questions that would lead to more focused water quality monitoring, biological inventory, and data analysis to assess lake threat. For example, lakes we highlighted as being in high condition may warrant a strategy of shoreline protection to maintain ecological integrity. On the other hand, highly threatened lakes may require strategies focusing on ecological restoration or more regular monitoring. We also hope that the parameters and queries we present start a constructive dialog about what further work is required; how to model parameters better over time; and what new information to incorporate to fine-tune these types of analyses. Literature cited Abell, R.A., D.M. Olson, E. Dinerstein, P.T. Hurley, et al. 2000. Freshwater Ecoregions of North America A conservation assessment. World Wildlife Fund. United States. Island Press ,Washington, D.C. Allen, J. D. 1995. Stream Ecology: Structure and function of running waters. Kluwer Academic Publishers. Dordrecht, The Netherlands. Anderson, M.A. and Barbour, H. Lower New England Ecoregional Plan. The Nature Conservancy, Boston MA 2000. Arnade, L.J. 1999. Seasonal Correlations of Well Contamination and Septic Tank Distance. Ground Water 36(6):920-923. Bryan, M.D., and D.L. Scranecchia. 1992. Species Richngess, Composition, and Abundance of Fish Larvae and Juveniles Inhabiting Natural and Developed Shorelines of a Glacial Iowa Lake. Environmental Biology of Fishes 35. Capiella, K. and T. Schueler, 2004 Crafting a Lake Protection Ordinance. Technical Article in Urban Lake Management. p751-768. Center for Watershed Protection (CWP). 2003. Impacts of Impervious Cover on Aquatic Systems. Watershed Protection Research monograph No. 1. Doppelt, B., M. Scurlock, C. Frissel, J. Karr. 1993. Entering the Watershed: A New Approach to Save Americas River Ecosystems. Island Press, Washington, D.C. Downing, S.S., A.S. Dixit, and J.P. Smol. 1996. The nitrogen:phosphorous relationship in lakes. Limnological Oceanography 37: 936-945. Eadie, J.McA; Keast, A. 1984. Resource heterogeneity and fish species diversity in lakes. Canadian Journal of Zoology. 62: 1689-1695. Engel S., and J.L. Pederson. 1998. The Construction, Aesthetics, and Effects of Lakeshore Development: A Review. Wisconsin Department of Natural Resources. Harper, H.H. 1995. Effects of Groundwater Seepage from Septic Tank Areas on Nutrient Loadins and Bacteriological Inputs to Clear Lake. Heimberger, M., Euler, D, and J. Barr. 1983. The Impact of Cottage Development on Common Loon Reproductive Success in Centeral Ontario. Wilson Bulletin 95:431-439. Hinch, S.G. and Collins, N.C. 1993. Relationships of Littoral Fish Abundance to Water chemistry and Macrophyte Variables in Central Ontario Lakes. Canadian Journal of Fisheries and Aquatic Sciences 50. Hinch, S.G., Somers, K.M., and N.C. Collins. 1994. Spatial Autocorrelation and Assessment of Habitat-Abundance Relationships in Littoral Zone Fish. Canadian Journal of Fisheries and Aquatic Sciences 51:701-712. Janson R., C. Nilsson, B Renofalt. 2000. Fragmentation of riparian floras in rivers with multiple dams. Ecology 81(4):899-903. Jiffry, F. 1984. Loss of freshwater shellfish and some ecological impacts after water drawdown in Lake Sebasticook. MS thesis. University of Maine, Orono Johnson, W.N. Jr., and P.W. Brown. 1990. Avian Use of a Shoreline buffer Strip and an Undisturbed Lakeshore in Maine. Northern Journla of Applied Forestry 7: 114-17. Kling, G.W., Kipphut, G.W et al. 2000 Integration of lakes and streams in a landscape perspective: the importance of material processing on spatial patterns and temporal coherence. Freshwater Biology43:3, 477-497 Kratz, T.K., Webster, K.E., et al. 1997. The influence of landscape position on lakes in northern Wisconsin. Freshwater Biology 37. 209-217 Lewis, D.B. and J. J. Magnuson. 2000 Landscape spatial patterns in freshwater snail assemblages across Northern Highland catchments. Freshwater Biology43:3, 409-420 Minns, C.K. 1989. Factors Affecting Fish Species Richness in Ontario Lakes. Transactions of the American Fisheries Society: V. 118, No. 5, pp.533-545. Moyle P.B., Light T. 1996. Biological invasions of freshwater: Empirical rules and assembly theory. Biological Conservation 78: 149-161. National Research Council U.S. 1992. Restoration of aquatic ecosystems. National Academy Press, Washington, D.C. USA. Quinlan, R, Paterson, A.M., et al. 2003 A landscape approach to examining spatial patterns of limnological variables and long-term environmental change in a southern Canadian lake district. Freshwater Biology48:9, 1676-1697 Robertson, W.D. and J. Harman. 1999. Phosphate Pluem Persistence at Two Decomissioned Septic System Sites. Gound Water 37 (2): 228-236. Siver, P.A. et al. 1999. Century Changes in Connecticut USA Lakes as Inferred from Siliceous Algal Remains and Their Relationships to Land-Use Change. Limology and Oceanography. 44(8): 1928-1935. Silk, N. and K. Ciruna ed. 2004. A Practitioners Guide to Freshwater Biodiversity Conservation. The Nature Conservancy. 419p. Theobald, D.M. 2001. Technical description of mapping historical, current, and future housing densities in the US using Census block-groups. Natural Resource Ecology Laboratory Colorado State University. 2001. Tonn, W.M. and J.J. Magnuson. 1982. Patterns in the Species Composition and Richness of Fish Assemblages in Northern Wisconsin Lakes. Ecology 63(4) 11491166. Tonn W.M., Magnuson, J.J., et al. 1998. Isolation versus Extinction in Assembly of Fishes in Small Northern Lakes. Ecology 79(8) 2941-2956. Tonn, W.M., J.J. Magnuson, M. Rask, and J. Toivonen. 1990. Intercontinental comparison of small-lake fish assemblages: The balance between local and regional processes. American Naturalist 136: 345-375. USGS 2003. The USGS Nonindigenous Aquatic Species database (http://nas.er.usgs.gov) Vaux, Peter. 2005. Maine Aquatic Biodiversity Project. Unpublished report, in progress. Webster, K.E., Soranno, P.A., et al. 2000. Structuring features of lake districts: landscape controls on lake chemical responses to drought. Freshwater Biology. Volume 43, Issue 3, Page 499-515, Mar2000 Whittier T.R., Kincaid T.M. 1999. Introduced fish in northeast USA lakes: Regional extent, dominance, and effect on native species richness. Transactions of the American Fisheries Society 128:769-783 Whittier T.R., Halliwell D.B., Paulsen S.G. 1997. Cyprinid distributions in northeast USA lakes: Evidence of regional-scale minnow biodiversity losses. Canadian Journal of Fisheries and Aquatic Sciences 54:1593-1607 Whittier T.R., Larsen D.P., Peterson S.A., Kincaid T.M. 1999. A comparison of impoundments and natural drainage lakes in the Northeast USA. Hydrobiologia. Whittier, T.R., Paulsen, S.G., Larsen, D.P., Peterson, S.A., Herlihy, A.T., Kaufmann, P.R. 2002. Indicators of Ecological Stress and Their Extent in the Population of Northeastern Lakes: A Regional-Scale Assessment. BioScience 52(3):235-247. Yeardley R.B. Lazorchak JM., Paulsen S.G. 1998. Elemental fish tissue contamination in northeastern US lakes: Evaluation of an approach to regional assessment. Environmental Toxicology and Chemistry 17: 1875-1884 Author Biographical Information: Arlene Olivero is acting GIS Manager and an Aquatic Ecologist in The Nature Conservancys (TNC) Eastern Region Conservation Science Office. Arlenes areas of expertise include raster and vector GIS analysis with a focus on hydrologic modeling, freshwater ecosystem classification and condition assessment, multivariate statistics, and conservation planning. Doug Bechtel has been Director of Conservation Science with the NH Chapter since 1999. Previously he was Associate Ecologist for the NH Heritage Program. Doug oversees all of TNCs New Hampshire science programs, including ecoregion 56P     - . B I T Z [ i j s { |   Ǿ{sh+CJaJhahK|CJaJhF=CJaJh+vCJaJhahq}-CJaJhah)CJaJh7d5CJaJhtf5CJaJhk5CJaJhk5>*CJaJhk>*CJaJhk6CJaJhk hU5 hk5 h%I5\ hk5\-6PQRa      . B U gda $da$gdkdgdk $da$gdkdgdk<U | ( j " > V v  > ^ }  2 `gda^gdagda   # $ ' : ; I J T U [ \ _ ` d i x y      ! ) * 0 1 5 = G H L U ` a h i m r s u v h+CJaJh[CJaJh+vCJaJhahq}-CJaJhF=CJaJT    }  * 1 H R u } $%*+348?FGNOT\delmqyҿʿʿʿʿʿhahaCJaJh+CJaJhaCJaJhah)CJaJhF=CJaJhahq}-CJaJh+vCJaJK2 S ~ @]z'(8uMgd_gd p^pgda ^`gda^gda ^`gdagda &(8  '-HIš hP@h_ hlwh_hU h{h_ha hRh_ h_Yh_ h{{h_ hh_h_h ha5hoFCJaJh CJaJhaCJaJhF=CJaJh+CJaJhahaCJaJ3M .I |44z%U^Vgd_V4iL{$'0+/48;A $da$gdkdgdk d`gdk$a$gdl5gdagd_-24567<LMNUV]oy%&&& &%&&&*&E3S3t33333333ȿh1Oh1O6 h1Oh1Oh1Ohh h7d5 hF=5 hk5hkhahk5CJaJhk5CJaJhtf5CJaJh_CJaJ hXh_h_ hlwh_h_CJOJQJ^JaJ hh_4333333444<=+=F=H=L=N======>>>?>?D?E?F?L?R?Y?c????̴̩̿̿yrnrnrg` hDKhk hDKhoh h ho h hgh hDKB*phh hoB*phh h B*phhB*phh hB*phhDKhoB*phhDKhB*H*phhDKhB*ph hDKhghghkh1Oh1O6 h1Oh1O h1Ohkh1O#??????K@N@i@u@x@@@A?AAABB(C?CKCLCCCCCCCCCCCCD DD#D;DADXDYD`DjD{DDDDFEGELEEEȺ㳨 hk5 hIh1Oh B*phhIhIB*phhIhDKB*ph hIhDK hIhk hIhI hIh/MlhIh/MlhDKhoh hhk hDKhk hDKho hDKhDK4AEEEAHBH*J|K}KLL MMMM;O