Registration Dossier

Ecotoxicological information

Sediment toxicity

Currently viewing:

Administrative data

Link to relevant study record(s)

Description of key information

Key value for chemical safety assessment

Additional information

PNEC Sediment: 109 mg Ni/kg dry wt. (Reasonable Worst Case PNEC)

A multi-year Technical Conclusion i) testing program for nickel in sediments was completed in December 2011 upon delivery to the Danish Environmental Protection Agency (DEPA) of all research reports from the laboratory testing program, field testing program, and conclusion i) group discussions. Delivery of the information package to DEPA fulfilled the requirements outlined in the Nickel EU Risk Assessment conclusion i) that was established in 2008 upon completion of the Existing Substances Risk Assessment of Nickel. On December 19, 2012, DEPA completed the evaluation of the nickel research reports and sediment effects assessment documents and submitted to ECHA a report entitled “Summary report regarding risk management of nickel in relation to chronic toxicity to freshwater sediment organisms”, as well as a summary entitled “Conclusion of substance evaluation for transitional dossiers”. This multi-year study had the goals of deriving Predicted No Effects Concentrations for sediment-associated nickel (PNECsed), and for identifying relationships between important sediment parameters and the toxicity of nickel to sediment-dwelling organisms. The first phase of the program focused on developing spiking procedures to create a more realistic exposure and reduce the diffusional loss of soluble nickel from the sediment phase into the overlying water in laboratory sediment toxicity tests (Brumbaugh et al, 2013). The purpose of the second phase of the study was to provide the ecotoxicity dataset necessary to populate a species sensitivity distribution (SSD) and to derive a PNECsed(Besser et al, 2013). The third phase of the study aimed to develop a bioavailability model to normalize nickel effects in sediment (Vangheluwe et al, 2013). Finally, field studies were conducted to validate laboratory results by examining the short-term toxicity and long-term recolonization of sediments (Costello et al, 2011). In 2013-2014, four additional species were tested in the laboratory. Toxicity data from these tests were added to the SSD and used to develop bioavailability models for the additional benthic species. The full Sediment Effects Assessment can be found in Vangheluwe and Nguyen, 2014.. 

 

Laboratory Results

The Technical Conclusion i) program evaluated the chronic toxicity of eight field collected Ni-spiked freshwater sediments (spanning the 10P to 90P of sediment parameters in the EU) to nine species of benthic invertebrates to create a robust database of chronic toxicity tests for nickel. The nine benthic organisms tested represent a variety of feeding strategies and taxonomic groups. The chronic toxicity tests covered different endpoints such as abundance, survival, growth, biomass, reproduction, fecundity and hatching, and resulted in four EC10values and four unbounded NOEC values (one species was excluded from evaluation). The data were used in a first phase to populate a SSD and to derive a reasonable worst-case PNEC (RWC-PNEC) for nickel. The 2013-2014 sediment testing program evaluated nickel toxicity to four species in three additional sediments (which also showed broad ranges of sediment parameters), resulting in an SSD with a total of eight EC10values. 

 

Next, the results of the chronic toxicity testing were used to characterize relationships between nickel toxicity and sediment characteristics. Bioavailability models were subsequently derived based on EC20values for the selected endpoints. The research identified Acid Volatile Sulfide (AVS) as the key parameter controlling Ni toxicity. Bioavailability models based on AVS were developed for three benthic organisms (H. azteca,G. pseudolimnaeusandHexagenia species) using sediments that represented broad ranges of AVS and other sediment parameters thought to affect metal bioavailability in sediments. In 2014, bioavailability models forT. tubifex,C. ripariusandS. corneumwere developed.

Regression models relating the toxicity of nickel to AVS in sediment.

 

Species

Model

R2

H. azteca

Log EC20total Ni (mg/kg dry wt) = 2.65 + 0.492 Log AVS (µmol/g dry wt.)

0.74

G. pseudolimnaeus

Log EC20total Ni (mg/kg dry wt) = 2.8 + 0.358 Log AVS (µmol/g dry wt.)

0.62

Hexagenia sp.

Log EC20total Ni (mg/kg dry wt) = 2.35 + 0.175 Log AVS (µmol/g dry wt.)

0.59a

(p = 0.07)

S. corneum

Log EC20total Ni (mg/kg dry wt) = 2.73 + 0.478Log AVS (µmol/g dry wt.)

0.99

T. tubifex

Log EC20total Ni (mg/kg dry wt) = 3.05 + 0.125Log AVS (µmol/g dry wt.)

0.99

C. riparius

Log EC10total Ni (mg/kg dry wt) = 2.85 + 0.1798Log AVS (µmol/g dry wt.)

 

0.99

E. virgo

Log EC20 total Ni (mg/kg dry wt) = 2.21 + 0.218 Log AVS (µmol/g dry wt.)

0.95

 

 

 

anon-significant

 

Most of the toxicity models show a significant relationship with AVS, with EC10/EC20values increasing with increasing AVS concentrations. Only the relationship between AVS andHexageniawas not statistically significant; however, the trend of decreasing toxicity with increasing AVS was still apparent. This may be a result of organism behavior, as the strongest mitigating effects are observed for those species with an epibenthic lifestyle such asH. azteca,S. corneumandG. pseudolimnaeus, with slopes ranging from 0.358 to 0.492. The relationships are less pronounced for the benthic speciesT. tubifex,C. ripariusandHexagenia sp(i.e. slopes ranging from 0.125-0.180) that exhibit more burrowing activity and subsurface feeding.

Chronic ecotoxicity data were normalized to an AVS concentration of 0.8 umol/g, which represents the 10thpercentile of its distribution in EU sediments. The non-normalized data measured from actual RWC sediments and the species mean values for all data after normalization to 0.8 umol AVS/g are shown in Table Y. The normalized data were included in the SSD for HC5-50derivation. Unbounded NOECs (> value) were not included in the analysis.

Non-normalized EC10values for the most sensitive endpoint for benthic invertebrates tested with nickel-spiked low binding sediments and EC10values normalized to 0.8 µmol/g dry wt AVS.

 

Organism

Most sensitive endpoint

Non-normalized EC10value (mg Ni/g dry wt.)

Normalized EC10value (mg Ni/g dry wt.) to RWC AVS (0.8 µmol/g dry wt.)

H. azteca

Biomass

149.1

203.5

E. virgo

Biomass

194

141.1

G. pseudolimneaus

Biomass

228

348.4

Hexagenia sp.

Biomass

236.7

188.7

L. variegatus

Abundance

554

529.8

C. riparius

Development

762

673.5

S. corneum

Biomass

388

322.1

T. tubifex

Biomass

1,103

1000,3

C. dilutus

 

> 762

NA

Lampsilis siliquoidea

 

> 762

NA

 

Field results

 

Multiple field studies were conducted to examine nickel toxicity under field conditions (e.g., Costello et al, 2011, Nguyen et al, 2011). These studies examined arange of different sediment types, were conducted during different seasons, represented different geographical locations (Europe and North America) indifferent types of systems (lotic and lentic), with varying water quality and abioticparameters. The field studies examined recolonization of the deployed spiked sediments over time as well as acute toxicity exposures.

 

The 2010-2011 field recolonization study (Costello et al. 2011) resulted in a NOEC of 230 mg Ni/kg dry wt, using the same sediment types that were used for laboratory testing. In this study, the effects on recolonization were measured after 28 and 56 days. Effects attributable to Ni exposure were only observed at the 28 day sampling period, regardless of the fact that Ni concentrations at day 56 remained greater than 4,500 mg Ni/kg in some cases. No effects on the composition of the benthic communities were measured at the Day 56 sampling period. Additionally, no acute effects were observed during the field deployment.  

 

The outcome of the field deployments provide a range of effects concentrations that are protective of the toxicity results seen in the laboratory sediment testing. No evidence exists to suggest that field data are more sensitive than laboratory-based HC5-50values. To the contrary, these data and similar data for benthic macro-invertebrates and pelagic communities show that field/mesocosm data are less sensitive than results of laboratory tests.

 

PNECsed Derivation

 

Results of the whole sediment toxicity test were normalized to the RWC AVS (0.8 µmol/g dry wt.). The resultingHC5-50value of 109 mg Ni/kg dry wt. was estimated from the log-normal distribution. TheHC5-50of 109 mg Ni/kg dry wt. is a robust estimate of nickel toxicity in sediments at the 10th percentile of parameters controlling nickel bioavailability and represents a RWC ecotoxicity value in the absence of the necessary data for performing bioavailability correction (i.e. AVS data).

 

The sediment bioavailability models were used to normalize the toxicity dataset to nine different bioavailability scenarios (eight sediments ranging from low to high AVS content and one hypothetical RWC condition) (Vangheluwe et al, 2013). The modeledHC5-50values obtained for the nine bioavailability scenarios range from 109-305 mg Ni/kg dry wt.

 

HC5-50values (mg TR-Ni/kg dry wt.; with 90% confidence limits) calculated using the organism specific AVS models (log-normal distribution).

Bioavailability scenario

Measured AVS

(µmol/g dry wt.)

Calculated HC5-50

(mg TR-Ni/kg dry wt.)

RWC

0.8

109 (40-182)

SR

0.9

115 (43-191)

DOW

1.0

121 (46-201)

STJ

3.8

185 (75-296)

RR2

6.1

210 (85-337)

RR3

8.0

225 (91-336)

P30

12.4

249 (99-403)

STM

24.7

284 (108-470)

WB

38.4

305 (111-515)

 

 

The PNECsedis calculated by dividing the HC5-50by the Assessment Factor (AF). The derivation of the RWC PNECsedtakes into account several lines of evidence, including sediment ecotoxicity data, pelagic ecotoxicity data in combination with the equilibrium partitioning approach, field studies, natural Ni background concentrations, and bioavailability correction. For instance:

·       Novel sediment spiking techniques were employed (Brumbaugh et al, 2013);

·       The robust single species sediment toxicity database (10 species: 8 EC10s, 2 Unbounded NOECs) represents the largest sediment database for a metal;

·       Sensitive groups were included in the database;

·       A wide range of relevant taxonomic groups was covered;

·       Sediments tested spanned the 10th to 90th percentile of the distributions AVS and TOC in the EU;

·       Direct relationships between toxicity and relevant sediment phases were established;

·       Quantifiable AVS relationships were established for 7 out of 8 species;

·       None of the geometric means of the single species toxicity tests were below the HC5-50value;

·       The protectiveness of the HC5-50value (benthic SSD) was further confirmed with field studies.

Based on the outcome of the Technical Conclusion i) testing program, the 2013-2014 Nickel Sediment Testing Program, and corroborating field studies, nickel toxicity and behavior in freshwater sediments has been well characterized and the remaining uncertainty is low. Given the robust sediment toxicity database, an AF of 1 is justified. The resulting RWC PNECsedis 109 mg Ni/kg dry wt.