Soil Biodiversity and Ecosystem Functioning


Global Litter Invertebrate Decomposition Experiment

Study Objectives

The GLIDE project will attempt to address the following four questions:

  1. Are similar taxonomic groups involved in decomposition irrespective of biomes and latitude?

  2. What are the effects of excluding animals to the rate of litter decomposition across biomes and latitudes?

  3. Are patterns of animal succession involved in decomposition the same across biomes and latitude even though the rate of succession varies?

  4. How does the succession of taxa vary with latitude and decomposition rate?

We will address these questions by using litterbags to determine whether a standard amount and type of organic "litter" at sites in different biomes decays differently than could be predicted by abiotic factors alone. Linked to this we will compile the first ever database of global patterns in succession of litter biodiversity and its relation to rates of decomposition, making these data accessible on the web. Research sites will be assigned Core or Satellite status. Both site types will require volunteer time and coordination; materials and taxonomic sorting are provided by GLIDE benefactors for Core sites, while Satellite sites need to provide their own funding for these services.

Core Sites: Thirty-six confirmed sites have been selected from pre-established international networks and International Long Term Ecological Research (ILTER) cooperators. These sites will fill gaps in our knowledge of decomposition and diversity for certain biomes and latitude. For information on how you can add your site click here.

Satellite Sites: Additional sites will be encouraged to participate. However, cooperators will need to provide their own funding for litterbags, shipping, and taxonomic sorting. These sites should complement biomes and latitude represented in Core site locations. Back to the top.

Litterbag Construction

Approximately 2000 20 cm x 20 cm 1.0 mm fiberglass mesh bags were constructed at Colorado State University, Fort Collins, CO in May 2001. Litterbags were filled with 10 g (+/- 0.5 g) of grass hay (Agropyron cristatum) that had been air-dried for 2 years at <20% rH, pre-processed through a 1.0 mm screen (to remove lose material prior to shipping) with all florets removed, and sterilized by gamma irradiation at IBA in Tustin, CA. Ratio of stem: leaf material was kept as similar as possible for all bags, but still could be a source of variation in decomposition rates. Hay is a local stock from Fort Collins, CO and initial litter C:N analysis (43.12:1.361) was determined at Colorado State University. Back to the top.

Litterbag Handling

Twenty-seven litterbags (4 plots X 6 bags per plot, plus 3 "traveler" bags) were shipped to each of the 31 core sites in June 2001. Care was taken to minimize loss of material before transport; however, it is highly likely that some material was lost in travel and transport. Consequently, all litterbags should be weighed to the nearest 0.001 g before placing them in the field. Following the recommendation of Harmon et al. (1999), 3 "traveler" bags were sent to each site; these bags should be handled as other bags, retrieved immediately after placement in the field and then reweighed to the nearest 0.001 g. This will allow us to determine variation of litter loss caused by transport and handling. Back to the top.

Experimental Design for Core Sites


Each site will have four plots with three replicates of the assigned treatment (litter only) and control (litter plus mothballs) in each plot, for a total of 24 bags per site. This design will allow for random removal of 8 bags (4 treatment and 4 control) per sampling date (see Time of Bag Removal below). One treatment and control bag will be removed per plot per sampling date.

Field Placement of Litterbags:

Areas where litterbags are placed will be cleared of all vegetation and debris. Bags should be placed in contact with mineral soil, flagged, and tethered by tent stakes in areas for which there is historical information. The tent stake will be driven into the corner of each bag to hold it in place and on the surface of the mineral soil. Bags must be placed well off game trails to minimize damage from large mammals.

Six bags will be placed in each of 4 plots (Figure 1); all plots should be at least 10 m apart. Back to the top.

Figure 1. Arrangement of plots and litterbags in the field

A plot is defined as two paired 20 m transects with 3 treatment and 3 control bags along each transect. Bags should be placed 10 m apart on these two parallel transects (Figure 2). Transect orientation within a plot will be randomly assigned (0-360 degrees) on-site by a random number generator. The treatment and control transects will be at least 10 m apart and parallel. This distance will address three concerns, 1) the control bags (mothballs) will be far enough from the treatment bags to avoid contamination, 2) reduce impacts of disturbance by small mammals, and 3) reduce the effects of spatial autocorrelation. Back to the top.

Figure 2. Distance between litterbags within each plot

Mothball Application:

A standardized formulation of mothballs will be added to bags at sampling time 0 in crystalline form at all sites. Mothball formulas vary internationally, so mothballs will be shipped from BioTrack in Australia. Two mothballs should be added around each control bag at each sampling time (Table 1). Back to the top.

Table 1. Addition of mothballs to treatment litterbags

Sampling time

# of mothballs required/plot (site)

Addition of mothballs


6 (24)

Bag 1--add 2 mothballs

Bag 2--add 2 mothballs

Bag 3--add 2 mothballs


4 (16)

Bag 1--remove litterbag

Bag 2--add 2 mothballs

Bag 3--add 2 mothballs


2 (8)

Bag 2--remove litterbag

Bag 3--add 2 mothballs


0 (0)

Bag 3--remove litterbag

Labeling Bags:

Barcodes to identify litterbags and specimens taken from each site will be provided by BioTrack (Mark Dangerfield, Key Centre for Biodiversity and Bioresources, Department of Biological Sciences, Macquarie University, NSW 2109 Australia). Metal id tags for barcodes will be placed underneath the bags to reduce the chance of attracting birds and mammals. Back to the top.

Time of Bag Removal:

There will be three sampling times (Table 2), and sampling interval starting points will vary with local climate. Tropic sampling (1, 2, and 3 months) will begin at the end of the wet season when plant production is at its maximum. Temperate, boreal, and arctic sampling (2, 4, and 12 months) will begin midsummer. Ideally, weather conditions are similar for sampling times within a site, but this is not probable or realistic due to seasonal variation. Site managers will record climatic data as often as possible over the sampling period. Site managers will also be responsible for site security (mammalian damage, human disturbance, etc.). Back to the top.

Table 2. Sampling intervals for tropical, temperate, arctic, and boreal climates.


 Sampling Time (months from bag placement)












































Litterbag Collection and Extraction Protocols:

Control bags will be photographed prior to collection with id tag in clear view. The surface of bags will be carefully cleaned of adhering soil, living plant parts (roots or moss), rock fragments, etc., immediately before collection. Retrieved bags will be placed gingerly in a plastic ziplock bag, sealed, and carefully (without much disturbance) returned to the laboratory. It is crucial to ensure that decomposing materials do not fragment and fall out of the bags during retrieval or before processing. Any fragmentation or loss of material must be documented on the data sheet for each sample (Harmon et al. 1999). All bags will immediately be weighed to a thousandth of a gram and then placed gently on a Tullgren funnel. Equipment (bulb wattage, screen size), processing time (minimum of 5 days), and extractant information for dry heat extraction have been standardized by M. Dangerfield and disseminated with barcode and protocol information by mail and on the website. The recommended heating regime will achieve a residual moisture content of 10% within 48 hours. Dry faunal extractions will be followed by on-site specimen preservation with a barcode label in 10 ml vials of 95% alcohol. Vials will be provided by BioTrack. Specimens are to be shipped to BioTrack in box labeled "dead insects for scientific research - no commercial value-preserved in 95% alcohol" where they will then be identified to morphospecies for all taxa. Earthworms, termites, and water-based animals (nematodes and tardigrades) will not be sampled with this design. Following each faunal extraction, treated litter will then be oven dried, weighed, and shipped to the NREL for appropriate chemical analyses as outlined in Harmon et al. (1999) and archived on-site due to quarantine restrictions. Back to the top.


This worldwide webpage will be used as a centalized database for data entry and analysis, discussion topics/papers, site-specific sampling times as well as bag handling, data collection/handling/ management, and shipping protocols. Site managers will be able to download standardized data sheets from this website in Excel format. These raw data sheets will be posted on the web, along with final morphospecies identification from BioTrack. Final data collation will be in BIOTA database format at GLIDE headquarters. This webpage will also facilitate other communications between site cooperators, Committee members, and the Project Coordinator at CSU. Back to the top.


Harmon, M.E., K.J. Nadelhoffer, and J.M. Blair. 1999. Measuring decomposition, nutrient turnover, and stores in plant litter. Pages 202-240 in G.P. Robertson, C.S. Beldsoe, D.C. Coleman, and P. Sollins, editors. Standard soil methods for long-term ecological research. Oxford University Press. Back to the top.



What is GLIDE?



Study Design



2005 Meeting

2003 Meeting


Image Gallery



This webpage is funded by the Soil Science Society of America.

Please contact the GLIDE headquarters (email: if you have any comments or questions.

GLIDE was a project of the International Biodiversity Observation Year 2001-2002

This material is based upon work supported in part by the National Science Foundation under Grant No. 98 06437 Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.


Search this site powered by FreeFind


This page was last updated on February 1, 2005

Apply to CSU