Benthic Macroinvertebrate Protocol

The described sampling protocol was designed to generate data sufficient to characterize the condition and trend of aquatic macroinvertebrate assemblages including quantifying the impact of anthropogenic disturbances and/or restoration actions. This protocol is only applicable to wadable, perennial streams; individuals seeking to sample intermittent, wetland, spring, or other systems should consult the NAMC Director.

This protocol is compatible with most aquatic macroinvertebrate monitoring programs targeting wadable streams throughout the Western United States (e.g., State environmental quality programs [OR, WA, UT, CO], BLM Lotic Assessment Inventory and Monitoring Program (Lotic AIM), USDA Forest Service Aquatic and Riparian Effectiveness Monitoring Program [AREMP], US EPA’s National Rivers and Streams Assessment [NRSA]), therefore slight variations to the quantitative protocol are permissible.

Sampling Procedure Recommendation

• Sampling period: July 1st – October 15th; earlier for extremely arid regions.
• Sampler type: Surber, kick net (D-frame), or Hess sampler
• Mesh size: 500 μm
• Sample reach length: 40 × wetted width or 20 x bankfull (minimum of 150 m)
• Target habitat: Fast-water (riffle) habitats
• Number of fixed area samples to composite: eight 1 ft² (8 ft²) fixed-area samples:
• Targeted riffle: one to two (1 ft²) samples collected from 4 to 8 different fast water habitats
• Reachwide (no riffle/fast-water habitats): one (1 ft²) sample collected from eleven evenly spaced transects.

Placement of Sampling Device

Sampling locations must be randomly determined to avoid bias:

1. Riffle/fast-water: generate eight random four-digit numbers between 0 and 9999. The first two numbers represent the percent upstream along the habitat unit’s length. The second two numbers represent the percent of the stream’s width from the left bank. Sample where the length and width intersect.
2. Reachwide approach: take one sample (1 ft²) from each of eleven transects. Collect samples alternately from the left quarter, center, and right quarter of each transect, with the first location randomly selected.

Sample Collection

Collecting the sample at each of the eight or eleven sample locations, orient the mouth of the sampler into (perpendicular) the flow. Collect invertebrates from within the area delineated by the net frame. If using a kick net, carefully delineate the 1 ft² sample area. Thoroughly wash all rocks, fine sediment, and organic debris to a depth of ~ 10 cm.

Field Processing

Place all contents in a bucket filled with water and decant invertebrates and organic matter into a 500 μm sieve or net (optional step). Repeat this process until only sand and gravel remains in the bucket. Inspect the remaining gravel for cased caddisflies, snails, or other invertebrates. Sample preservation 75-95% ethanol; 3:1 preservative to sample by volume.

Field Equipment (*required)

• 0.09 m²; Surber, kick net, or Hess sampler with 500 μm mesh net*
• Buckets (2), plastic, 8-10 qt. capacity
• Sieve with 500 μm mesh
• Squirt bottle (2), 0.5 – 1 L capacity*
• White plastic wash tub*
• 500 mL HDPE plastic with screw caps sample jars*
• Small spatula, scoop, or spoon
• Forceps
• Funnel with large bore spout
• 95% ethanol*
• Rubber gloves
• Cooler
• Labeling materials*
• GPS

Detailed Sampling Instructions

Sample Design

This protocol focuses exclusively on field procedures for benthic macroinvertebrate collection; it does not explicitly address study design or data analysis. In general, stream reaches for benthic macroinvertebrate monitoring are selected using either a targeted or probabilistic design depending on monitoring objectives and associated scope of inference. Sites selected using a targeted design generate data that is relevant for measuring impacts from a known source or answering other site-specific questions. Sites selected using a probabilistic design provide information on the overall condition or trend of the watershed, basin, or region. The number of sampling locations and sampling frequency are two other important considerations that should be directly tied to explicit monitoring objectives. The described protocol applies to both targeted and probabilistic sampling designs.

Recommended Sampling Period

Macroinvertebrate sampling should be conducted between July 1st and October 15th. However, sampling may begin earlier in the southern, arid regions of CO, UT, NV, AZ, NM, and CA. This sampling timeframe is designed to minimize sampling variability related to seasonality and macroinvertebrate phenology (i.e., developmental timing). Furthermore, this time is optimal because the stream benthos has stabilized following spring runoff events, many macroinvertebrates have attained body sizes that can be readily identified, and macroinvertebrate species richness is generally maximized.

Type of Sampler

A variety of samplers exist for collecting quantitative, benthic macroinvertebrate samples. The two most important considerations for choosing a sampler are the ability to collect a fixed-area sample (i.e., standardize area of stream bottom to be sampled) and mesh size. Both Surber (1 ft x 1 ft / 0.305 x 0.305 m) and Hess (0.6 ft / 0.18 m radius) samplers are optimal because the net’s metal frame delineates and standardizes the area to be sampled. In contrast, D-frame or kick nets require the user to carefully standardize the area to be sampled; depending on the area of the kick net, this will be either 1 ft² (0.093 m²) or 1ft x 2ft (0.305 m x 0.61 m). While Hess and kick nets are all compatible with this protocol, the use of a Surber sampler is recommended. Mesh size refers to the size of the openings in the net of the sampling device. A 500 μm mesh size is recommended herein, regardless of sampler type, as this mesh size is consistently used by a majority of state and federal biological assessment programs. If a sieve is used to clean samples, ensure the sieve mesh sizes matches that of the sampler.

Sample Reach Length

Sample reaches need to be long enough to incorporate local habitat-scale variation and
represent average conditions. In meandering alluvial channels, pool-riffle sequences generally alternate every 5 – 6 times bankfull width. Therefore, to incorporate multiple riffle habitat units sample reaches should be scaled in proportion to stream size, with the sample reach encompassing forty (40) times the low flow wetted width or twenty times bankfull width and a minimum reach length of 500 ft (150 m). Note that other monitoring parameters such as physical habitat or fish assemblages may influence the length of the sample reach. At a minimum, four (4) different fast-water habitats, if present, should be sampled.

Target Sampling Habitats

Macroinvertebrate samples should be collected from fast-water habitats (i.e., riffles), if available. Riffles are characterized by relatively fast currents, moderate to shallow depth, cobble/gravel substrates, and generally have the most diverse macroinvertebrate assemblages. Furthermore, standardizing sampling to a fixed habitat type simplifies sampling methodologies and facilitates comparisons among sites. See instructions below for sampling reaches with no fast-water habitats.

Compositing Subsamples

Because of the patchy distribution of benthic macroinvertebrates, multiple samples should be collected and composited into a single sample. Specifically, sampling a minimum of eight (8) replicates (total of 8 ft² [0.74 m²] if using a Surber sampler) is recommended. Given the sampling area of the samplers described above and the number of samples needed to adequately characterize benthic heterogeneity, macroinvertebrate samples should be taken from either: 4 different fast-water habitats (two separate 1 ft² fixed-area samples taken from each of 4 fast-water units for a total of 8 samples (8 ft² of stream bottom sampled)); or 8 different fast-water habitats (one (1 ft²) fixed-area sample taken from 8 fast-water habitats for a total of 8 samples).

If no fast-water habitats are present, locate eleven equally spaced transects (perpendicular to the thalweg) along the study reach. Take one 1 ft² sample from each of the eight transects. Samples should be alternately taken from the left quarter, center, and right quarter of each transect, with the first location being randomly selected. For example, if the first sample was collected from the middle of the first transect, subsequent samples would be collected from the right quarter of the second transect, the left quarter of third transect, the middle of the fourth transect and so on.

Placement of Sampling Device

Once the target stream reach has been delineated and the riffle/fast-water habitats identified, the locations of the eight individual samples need to be determined. To avoid bias, individual samples should be located such that each square foot of riffle habitat has an equal probability of being selected. Net placement is most easily determined by generating eight (8) random four-digit numbers between 0 and 9999. The first two numbers represent the percent upstream along the habitat unit’s length. The second two numbers represent the percent of the stream’s width from the left bank. Take samples where the length and width distances intersect (estimate by eye). If it is not possible to take a sample at the locations (e.g., log in the way, too deep, etc.), draw additional random numbers until you can. Be sure to first sort the eight, four-digit numbers from lowest to highest to facilitate working from downstream to upstream.

Collecting the Sample

At each of the eight randomly located sample locations, place the sampler so the mouth of the net is facing into and perpendicular to the flow. Collect invertebrates from the area delineated by the net frame. If using a kick net, carefully delineate the 1ft² or 2 ft² area to be sampled. Work from the upstream edge of the sampling plot backward and carefully pick up and rub stones directly in front of the net to remove attached animals. Quickly inspect each stone to make sure you have dislodged everything and then set it aside. If a rock is lodged in the stream bottom, rub it a few times concentrating on any cracks or indentations. After removing all large stones, disturb small substrates (i.e., sand or gravel) to a depth of about 10 cm by raking and stirring with your hands. Continue this process until you can see no additional animals or organic matter being washed into the net. After completing the sample, hold the net vertically, with the mouth up, and rinse material to the bottom of the net. If a substantial amount of material is in the net, empty the net into the bucket for processing before continuing to the next sample location. Otherwise, move to the next sample location and repeat the above procedure.

Field Processing

After taking the eight samples, empty the net contents into the 500 μm sieve, and rinse thoroughly to remove fine sediment. Spoon or spatula the material in the sieve (invertebrates and organic matter) into the sample jar and then wash any remaining material in the sieve into the jar with a squirt bottle. You may do this over a plastic bucket to ensure that no material spills from the sieve. During this process, all organic and inorganic materials should be retained and placed into sample jars unless thoroughly washed. Algae cannot be thoroughly washed, for example, and should be retained in the sample jar. If material is too large to fit into jars, thoroughly wash and inspect for organisms such as cased caddisflies, snails or other animals that might remain prior to discarding. Remove any remaining animals by hand using tweezers and place in the sample jar. 

Once the sample has been processed, add 95% ethanol to fill the jar or vial. Be careful not to add too much organic matter relative to ethanol recommended dilution is three (3) parts of preservative to one (1) part sample by volume. Immediately label the jars inside and out with the number of jars (e.g., 1 of 5, 2 of 5), date, stream name, a unique site and sample identifier if possible, and state. Recording the date using the following format assists us with ensuring data quality: use letters rather than numerals for the month and to use four digits for the year (e.g., 06Sep2014). Labels must be on Rite in the Rain paper and filled out with a pencil. Labels that are not Rite in the Rain paper will decompose from the ethanol, while ink from pens will leach and become illegible. Place the inside label loose in the sample and place the outside label on the side of the jar with clear packing tape making sure the tape is wrapped completely around the jar. Tighten the lid of the sample jar, and secure it in place with electrical tape wrapped clockwise around the jar and lid. Do not use clear packing tape to secure the lid to the sample jar. In a spreadsheet, record the following data to be submitted electronically to NAMC:

• Site ID -The name/code you are calling a site.
• Visit ID - A concatenation of site and date or similar value (optional)
• Sample Date - Date of sampling effort for this site.
• Jar Count - The number of jars containing this sample.
• Field Split (%) - The total % of sample being sent to NAMC. Usually 100 unless the sample was subsampled in the field.
• Latitude (DD) - Latitude (in decimal degrees, not DMS)
• Longitude (DD) - Longitude (in decimal degrees, not DMS)
• Waterbody Name - The name of the water body. If the waterbody has no given name, put NA
• Ecosystem: Habitat - The type of habitat sampled (typically riffle or multiple habitat)
• Sample Design - The style of sampling (targeted, transect, proportional)
• Sample Effort Standardized - If the sampling was standardized by area (i.e. the above protocol), by time, or not standardized.
• Area (m²) or Time (min) sampled
• Sampler Type - The type of net or apparatus used to sample.
• Mesh Size (µm) - The size of the mesh of the net or apparatus used to sample.
• NHD ID - A unique identifier from the National Hydrography Dataset (COMID) that corresponds to your site location. This assists us with watershed delineations and bioassessment index computations (optional)