Don L. Stevens

Spatially Balanced Sampling of Natural Resources

The spatial distribution of a natural resource is an important consideration in designing an efficient survey or monitoring program for the resource. Generally, sample sites that are spatially balanced, that is, more or less evenly dispersed over the extent of the resource, are more efficient than simple random sampling. We review a unified strategy for selecting spatially balanced probability samples of natural resources. The technique is based on creating a function that maps two-dimensional space into one-dimensional space, thereby defining an ordered spatial address. We use a restricted randomization to randomly order the addresses, so that systematic sampling along the randomly ordered linear structure results in a spatially well-balanced random sample. Variable inclusion probability, proportional to an arbitrary positive ancillary variable, is easily accommodated. The basic technique selects points in a two-dimensional continuum, but is also applicable to sampling finite populations or one-dimensional continua embedded in two-dimensional space. An extension of the basic technique gives a way to order the sample points so that any set of consecutively numbered points is in itself a spatially well-balanced sample. This latter property is extremely useful in adjusting the sample for the frame imperfections common in environmental sampling.

Variable density grid-based sampling designs for continuous spatial populations

Many environmental resources, such as mineral resources or vegetation cover, or environmental attributes, such as chemical concentration in a stream or benthic community structure, are most appropriately sampled as continuous populations distributed over space, but most applied sampling theory and methodology is concerned with finite, discrete populations. This paper reports sampling methodology that explicitly recognizes the continuous nature of ecological resources. A family of designs are developed to permit control of the spatial dispersion of the sample, variable spatial density, and nested subsampling. The designs have non-zero joint inclusion probability densities, so that rigorous design-based inference and variance estimation are possible.

Spatially Restricted Surveys Over Time for Aquatic Resources

Consideration of the natural characteristics of aquatic resources and available frame material has led to the development of new designs for surveying large-scale regions. This paper illustrates survey designs developed to meet the requirements for surveying various aquatic resources, including a finite, discrete population, such as lakes within one or more states; a continuous linear population within a bounded area, such as wadeable streams within one or more states; and a continuous two-dimensional population within a bounded area, such as coastal waters associated with one or more states. We present a unified approach that addresses the differences of the aquatic resources assuming the availability of frame material, such as Geographic Information System (GIS) coverages of the boundary for coastal waters, stream network, and lake locations from U.S. Environmental Protection Agencys River Reach File 3, derived from U.S. Geological Survey digital line graph data from 1:100,000 scale maps. The basic design methodology distributes the sample over the spatial extent of the resource domain, and a panel structure can be used to extend the sample through time. Key features for the approach are (1) utilizing survey theory for continuous populations within a bounded area, (2) explicit control of the spatial dispersion of the sample, (3) variable spatial density, (4) nested subsampling, and (5) incorporating panel structures for sampling over time.

ASSESSMENT OF WETLAND CONDITION: AN EXAMPLE FROM THE UPPER JUNIATA WATERSHED IN PENNSYLVANIA, USA

The requirement of Section 305(b) of the Clean Water Act (CWA) that all waters of the U.S. be assessed every two years has been historically ignored for wetlands, even though they are included in the definition of “waters of the U.S.” This paper presents the use of a landscape and rapid assessment to describe the wetland resource and assess wetland condition in the Upper Juniata watershed in central Pennsylvania, USA. A Floristic Quality Assessment Index (FQAI) is used to calibrate and refine the landscape and rapid assessments. The landscape assessment defined ecological condition of sites in terms of the degree of departure from reference standard condition (i.e., wetlands in predominantly forested settings). Criteria for condition categories were based on the literature or best professional judgment and resulted in more than half of the area of the resource being rated in high or the highest condition, while about 12% was rated in low condition. The rapid assessment adjusts the landscape assessment by accounting for the presence of Stressors and the ameliorating effects of a buffer. This resulted in a 38% decrease in the proportion of wetland area in the highest and high condition categories and almost quadrupled the area in low condition. Classification and Regression Tree (CART) analysis was used to evaluate 1) whether the results of the landscape and rapid assessments correspond to those from the more quantitative data in FQAI and 2) whether the condition categories established for the landscape and rapid assessments agree with those established using FQAI. CART results indicate that our initial delineation of condition categories for the landscape and rapid assessments should be more stringent. However, it appears that the rapid assessment does a better job of gauging the factors important to wetland condition, as measured by FQAI, than the landscape assessment. This work can serve as a template for wetland monitoring and assessment and reporting as required by the U.S. Clean Water Act. Overall, such monitoring provides information that can be used to target areas for attention or protection, prioritize sites for restoration, design restoration projects, and choose best management practices.

Environmental Monitoring and Assessment Program, EMAP–Surface Waters: A Northeast Lakes Pilot

ABSTRACT In response to a growing need for statistically sound information on status and trends in the condition of the nations ecosystems, the U.S. Environmental Protection Agency is developing and implementing the Environmental Monitoring and Assessment Program (EMAP). One component, EMAP-Surface Waters, will concentrate on lakes and streams. In its focus on lake condition at the regional and national scale, EMAP-Surface Waters will address concerns about the present extent and geographical distribution of lakes, their current ecological condition, the proportion that is degrading or improving, where, and at what rate, and the likely causes of adverse effects. The national design is based on a systematic grid of randomly placed points, so that conditions and trends can be estimated with known uncertainty. Lake condition will be assessed primarily through biological measurements; physical, chemical, and landscape information will be used to support and interpret conditions estimated by the biological in...

ASSESSMENT OF WETLANDS IN THE UPPER JUNIATA WATERSHED IN PENNSYLVANIA, USA USING THE HYDROGEOMORPHIC APPROACH

This paper reports on the ecological status of wetlands in the Upper Juniata watershed in Pennsylvania, USA, as determined by employing the hydrogeomorphic (HGM) approach. HGM assessment provides a measure of the potential functional performance of a single wetland for up to 11 functions, depending on the subclass. Functional Capacity Index (FCI) scores calculated for each function range between a score of 1 (indicates the site is performing at optimum levels) and a score of 0 (indicates the site is not performing the function). Mean scores for all functions for the wetland resource in the Upper Juniata ranged from 0.48–0.63, except for Long-term Surface-Water Storage (0.39) and Characteristic Hydrology (0.85). Cumulative Distribution Function (CDF) plots were fairly linear over most of the distribution for all functions, indicating that the FCI scores were evenly distributed over the population. Several of the plots flattened at the upper and/or lower ends of the curves, indicating that a very small proportion of the wetland area had the highest and lowest scores. Clustering of the 69 riverine and slope sites using the FCI scores from the three functions with the most well-developed models resulted in the formation of four Functional Status Groups (FSGs). Groups 1 and 2 represented relatively high functioning groups of sites. They were differentiated by an exceptionally high Plant Community Function in Group 1 that differed significantly from the low value in Group 2. FSG’s 3 and 4 represented relatively low functioning groups of sites and were differentiated by a significantly high Vertebrate Community Function in Group 3. We defined three reference domains (Natural, Agricultural, and Developed) based on predominant land cover. Sites of any given FSG were distributed across the reference domains, but there were some differences in distribution. Sites in the Natural Domain were much more likely to be in the higher functioning FSGs, while the Agricultural Domain was dominated by sites with an overall low level of functioning. Sites in the Developed Domain are equally distributed across the four FSGs. In summary, we demonstrated how HGM assessment might be employed to describe the functional status of the wetland resource in a watershed. We also demonstrated how the results of the assessment could be (1) used to evaluate the efficacy of the models comprising the HGM assessment and (2) combined with other data to identify relationships that could be used to develop management approaches.

SAMPLE DESIGN, EXECUTION, AND ANALYSIS FOR WETLAND ASSESSMENT

Probability-based sample surveys are increasingly being used to assess natural resource condition, yet many of the techniques commonly used in survey sampling are not well known to environmental scientists. Here we discuss several techniques from survey sample methodology that can substantially increase the efficiency of a sample (lower the variance for the same sample size), make the survey more cost-effective, or adjust the analysis to accommodate difficulties in implementation, such as lack of access to all sample sites. The techniques are illustrated with two surveys that were designed to assess wetland condition: one in the Juniata watershed in Pennsylvania, USA, and one in the Nanticoke watershed in Maryland and Delaware, USA.

COMBINING HGM AND EMAP PROCEDURES TO ASSESS WETLANDS AT THE WATERSHED SCALE — STATUS OF FLATS AND NON-TIDAL RIVERINE WETLANDS IN THE NANTICOKE RIVER WATERSHED, DELAWARE AND MARYLAND (USA)

The hydrogeomorphic (HGM) approach to wetland assessment was combined with the Environmental Monitoring and Assessment Program (EMAP) survey design procedures to evaluate the condition of non-tidal riverine and flats wetlands in the Nanticoke River watershed (Delaware and Maryland, USA). We found degradation of wetland functions below reference standard levels for the majority of wetlands in both classes. Wetland condition was also related to the level of disturbance in both wetland classes. In flats, the most common disturbances were associated with hydrologic and vegetation modifications. Flat wetlands with low HGM function scores for the Plant Community and Habitat functions had almost all been converted from hardwood forest to Loblolly pine plantations. Most modifications associated with riverine wetlands were associated with stream channelization. Results of this study demonstrate that a site-specific and reference-based approach to assessment (i.e., the HGM method) can successfully be applied at the scale of an entire watershed if it is combined with a sampling approach that allows sites to be selected without geographic bias. The approach can also be used to determine if wetland functions vary from one sub-basin to another, and results of this project can be used by managers to begin to develop strategies for restoration of wetland functions at the watershed scale.

Using a Master Sample to Integrate Stream Monitoring Programs

The need for aquatic resource condition surveys at scales that are too extensive to census has increased in recent years. Statistically designed sample surveys are intended to meet this need. Simple or stratified random sampling or systematic survey designs are often used to obtain a representative set of sites for data collection. However, such designs have limitations when applied to spatially distributed natural resources, like stream networks. Stevens and Olsen proposed a design that overcomes the key limitations of simple, stratified random or systematic designs by selecting a spatially balanced sample. The outcome of a spatially balanced sample is an ordered list of sampling locations with spatial distribution that balances the advantages of simple or stratified random samples or systematic samples. This approach can be used to select a sample of sites for particular studies to meet specific objectives. This approach can also be used to select a “master sample” from which subsamples can be drawn for particular needs. At the same time, these individual samples can be incorporated into a broader design that facilitates integrated monitoring and data sharing.

Response designs and support regions in sampling continuous domains

In many environmental samples, the target population is distributed over space in a more or less continuous manner, e.g., the waters of a lake or the trees in a forest. Attributes of such a population can be conceptualized as a continuous function defined on the spatial domain of the population. Some attributes (e.g., water temperature) can be observed at a point; others (e.g., species diversity) can only be determined over a finite extent or support region. A fixed-shape support with uniform weights leads to an unbiased estimator of the population total; however, it may be impossible to maintain a fixed shape near domain boundaries. From a purely formal standpoint, unbiasedness can be maintained by using differential weights or by changing the shape of the support region near the boundary. Both of these procedures raise some issues of interpretation that often are overlooked. We derive estimators that account for edge effects under several support strategies, and identify some interpretation issues, using examples from forestry and limnology. Copyright