We have archived this page and will not be updating it.

You can use it for research or reference.

Language selection

Search


Research report summaries 2003–2004

Archived information

Archived information is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

Contractors’ reports are only available in the language in which they are submitted to the Canadian Nuclear Safety Commission (CNSC).

RSP-0164 – Report on performing external dosimetry during non-uniform radiation exposures

In environments where workers may be occupationally exposed to ionizing radiation, there may be situations where their bodies are exposed very non-uniformly. In such situations, wearing one dosimeter is usually insufficient. If two or more dosimeters are worn simultaneously, questions arise as to where the dosimeters are to be worn and how the resulting set of measurements should be interpreted in order to arrive at doses that are to be compared with regulatory dose limits. This document addresses these issues of non-uniform occupational exposures for situations where the torso may be non-uniformly exposed and where the extremities, i.e., hands or feet, may receive higher doses than the torso where a dosimeter is worn.

This report was prepared by the CNSC Working Group on External Dosimetry. This working group is composed of Canadian scientists from the nuclear sector, external dosimetry services and federal government departments. All are experts in external dosimetry.

Top of page

RSP-0164-1 – Report on performing neutron dosimetry in Canada

In Canada, occupational doses due to external sources of strongly penetrating ionizing radiation can arise from exposure to photon, i.e., gamma and x rays, and neutron radiation. Since neutrons interact differently with matter than do photons, different types of dosimeters must be used to measure doses due to these different types of radiation. In addition, neutron energy is an important factor in further determining the best type of neutron dosimeter that should be used to most accurately measure neutron dose. This report addresses these issues and is meant to help organizations, whose staff may be occupationally exposed to neutrons in various types of work environments, in selecting the neutron dosimetry method that is most appropriate.

This report was prepared by the CNSC Working Group on External Dosimetry. This working group is composed of Canadian scientists from the nuclear sector, external dosimetry services and federal government departments. All are experts in external dosimetry.

Top of page

RSP-0165 – Uranium intake – Dose estimation methods

The low annual limit on intake (ALI) and the low abundance of energetic beta or gamma emissions for the naturally occurring uranium isotopes may result in personal bioassay not being sensitive enough to detect intakes in some situations. Because of these limitations, two methods for estimating the committed effective dose resulting from occupational exposures to uranium are proposed for use at uranium mills (after the uranium in ore has been extracted and converted to yellowcake), processing plants and fuel fabrication facilities. These methods, the air monitoring method and the group lung counting method are described in the report. Dose assessment using urinalysis, however, is not addressed in this report. The usefulness of these methods and their limitations are also discussed.

This report was prepared by the CNSC Working Group on Internal Dosimetry under the auspices of the Regulatory Research and Support Program of the CNSC.

Top of page

RSP-0166 – Simulation of the FEBEX experiment as a test case for DECOVALEX III

This final report presents the three phases of the work completed in connection with the CNSC project initiated as a result of the participation of CNSC in DECOVALEX III associated with FEBEX project. The final report is a combination of four separate reports dealing with (i) the documentation and calibration of the FRACON code developed by the research and engineering personnel at the CNSC for the study of the thermo-hydro-mechanical (THM) response of fluid saturated materials, (ii) the extension of the FRACON code through the use of a state space concept to include the mechanical response of bentonitic materials under moisture uptake and the development of the FRACON 3D-UNSAT, (iii) the calibration of the FRACON 3D-UNSAT through experimental data and the modelling of the bentonite response in the FEBEX experiment and (iv) the application of the FRACON 3D-UNSAT code for the modelling of the rock mass response. The four reports presented in the appendices are self-contained in terms of the background, the theoretical developments, the computational results and conclusions.

The FRACON code and the FRACON 3D-UNSAT code are the property of the CNSC and their use is restricted to personnel authorized through written agreements. The FRACON 3D-UNSAT code should not be copied, modified or distributed in any form, or used for any assessment purposes whatsoever without the expressed written consent of the CNSC.

Top of page

RSP-0167 – Proposed plan for scaling analysis

An outline is proposed for a scaling analysis that facilitates the closure of the CNSC’s generic action item (GAI) 00G01. The generic action item is to resolve the issues surrounding the prediction of voiding and voiding rates in individual fuel channels during depressurization in the initial phase of a large loss of coolant accident (LLOCA) in CANDU reactors. Flashing and heating during the initial phase cause coolant voiding and, due to positive reactivity feedback, a fission power pulse.

The scaling analysis must identify the scaling criteria (dimensionless scaling groups) for rapid channel-voiding phenomena and establish whether physical similarity exists between the full-size reactor and reduced-size test facilities, which are used for validating safety-analysis codes. Insufficient similarity would mean that the experimental data from such test facilities do not represent phenomena in the full-scale reactor and are therefore inadequate for the validation of safety analysis codes.

Top of page

RSP-0168 – Review of the coverage limit in the Canadian Nuclear Liability Act – Phase II

The current Canadian Nuclear Liability Act (NLA) [1] limits the liability of nuclear reactor operators in Canada to the first $75 million of any off site damage resulting from an accident. The Government of Canada recently decided to increase this limit to bring it in line with international trends.

The new NLA limit will be based, in part, on the expected higher off site costs of the range of accidents under consideration for nuclear power stations. In preparation for this, the CNSC has commissioned this study to investigate the estimated costs of design basis accidents at Canadian single and multi-units CANDU power stations.

The scope of this project is limited to the evaluation of selected tangible cost components for selected representative accident scenarios and accident conditions, so that an illustrative evaluation of the potential hard component can be made. The scope does not cover a comprehensive evaluation of all conditions at the time of the postulated accident. It does not cover all possibilities in terms of the parameters that can affect cost.

This study is a first step towards understanding, quantifying and documenting the impact of various parameters on the potential cost of a nuclear accident. As such, the analysis presented herein only considers design basis accidents. Severe accidents have for the moment been excluded from the scope of this study.

Top of page

RSP-0169 – Review of Bruce A steam generator and preheater condition assessment and life cycle management plan for research project; Condition assessment and life cycle management of aging steam generators

This project reviewed, for the CNSC, a group of Bruce Power documents and, based on that review, provided expert opinions regarding a series of questions related to the condition assessment and life cycle management of the Bruce Unit 3 and 4 steam generators (SGs) and preheaters. The Bruce Power documents to be reviewed were identified in the contract and are listed in section 6 of this report. The questions to be addressed are listed in the subsequent sections of this report, together with DEI’s detailed responses to these questions. Subsequent to a draft of this document being prepared, Bruce Power provided responses to some preliminary questions (reference i); these responses were considered in this final report. In addition, Bruce Power provided written comments on a draft of the report (reference ii), and additional discussion of their comments was provided at a meeting at the CNSC on October 15, 2003. This final version of the report reflects these comments where they deal with descriptive information regarding the steam generator materials, manufacture, and inspections performed. However, the report does not address Bruce Power’s comments and plans regarding future operational assessments and nor their bases for considering sampling inspection strategies to be appropriate since these topics are outside of the scope of the DEI task.

Top of page

RSP-0170 – Proceedings of the Long Term Management Perspective for Idle Uranium Mines Workshop

The Long-Term Management Perspective for Idle Uranium Mines Workshop was held over October 6 – 9, 2003, in Peterborough, Ontario. This workshop was the final in a set of three workshops on this general topic held by the CNSC. The first two workshops were held in June 2001 and September 2002, respectively. These workshops focused on licensing commitments from government agencies and practical topics for licensees and proponents on technical issues.

The initial workshop involved only those government departments immediately involved in remedial work at legacy uranium mines whereas the second workshop brought together a broader cross-section of stakeholders including current licensees, private sector and public sector proponents.

The third workshop addressed many outstanding issues related to the long-term expectations of regulatory agencies at legacy uranium mine sites. This final workshop added other regulatory bodies including:

  • Environment Canada
  • Health Canada
  • Department of Fisheries and Oceans Canada
  • Ontario Ministry of the Environment
  • Ontario Ministry of Northern Development and Mines
  • Saskatchewan Environment

Through this workshop, the CNSC would put forth the ideas of institutional controls, long-term monitoring and maintenance, financial assurances and other concepts, that require multi-stakeholder commitments.

Top of page

RSP-0171 – Assessment of radiation doses arising from civilian or military vehicle use of radium luminous devices in the civilian community from operation vehicles, museums and collections

The use of radium in consumer and military products occurred before nuclear materials in Canada were regulated. In particular, radium was used from the early 1900s to approximately the late 1960s in the manufacture of self-luminescent paint. This paint was widely used in the manufacture of avionic components. In addition, other occurrences of radium luminescent devices can be found in various military artifacts, including vehicles, tanks, artillery and radios.

Since the coming into force of the Nuclear Safety and Control Act, there are now many specific regulatory requirements that impose restrictions on possession and use of radium luminescent devices. According to the regulations, a CNSC licence is required for the possession of more than 10 such devices. However, an exemption from licensing for simple possession of more than 10 devices was granted by the Commission in order to allow research and formulation of a regulatory approach.

The objectives of this report were to determine the scope and distribution of radium luminous devices in the public domain: in operational aircraft, static displays and museums, collections, Royal Canadian Legions or other locations where these devices may be located. Specifically, the project involved the measurement of radiation doses and the evaluation of scenarios as received by museum staff and visitors, personnel involved in refurbishing and restoration activities (excluding instrument repair), cockpit or maintenance crew of operating aircraft, or other vehicle operators or restorers.

Twenty four museums, training facilities and commercial airline operations were visited to gather to obtain the sample data. A total of 337 radium luminescent devices were measured at contact. As well, measurements were made of dose rates received to occupants of the pilot and copilot seats, from which dose scenarios were derived for aircraft crew, maintenance and museum staff, and general public members accessing aircraft displays. Measurements are also provided for radium luminous instrument storage areas.

This report also includes recommendations and comments arising from this study.

Top of page

RSP-0172 – International regulatory practices in fuel design qualification

This report has been prepared in accordance with the Articles of Agreement between the Canadian Nuclear Safety Commission, 280 Slater Street, P.O. Box 1046, Station B, Ottawa, Ontario, K1P 5S9, Canada and Davies Associates, Incorporated, 1309 Query Mill Road, Gaithersburg, MD 20878, USA. The contract no. and title is 87055-02-0259, "International Regulatory Practices in Fuel Design Qualification".

Tasks to be performed:

  1. Identify publicly available documentation related to fuel review and approval process such as documents produced by IAEA, NEA, and other international and national nuclear organizations.
  2. Conduct a survey of new fuel design approval practices employed in other countries operating nuclear power reactors. This will include an overview of:
    1. a) the procedures followed by the utilities and/or fuel designers to initiate a regulatory review of a new fuel design
    2. b) the regulatory agency’s fuel design review plan (objectives, areas reviewed, documentation requested, acceptance criteria, etc.)
  3. Review CNSC (AECB) past practices for the approval of new fuel designs.
  4. Recommend a fuel design review plan identifying Canadian regulatory requirements that a new fuel design must satisfy to be allowed for use in power reactors.

The results of our study of these four tasks are summarized in Chapters 1 to 4 in this report.

Top of page

RSP-0173 – Characterization of Northern Transportation Route (Ntr) sites under institutional controls

Since 1992, the Atomic Energy Control Board (AECB) (now the CNSC has been aware of the existence of approximately 30 sites that were potentially contaminated during the transportation of uranium ore form the mine at Port Radium, Northwest Territories (NWT) to the railhead at Fort McMurray, Alberta. The sites were located in populated areas and were, over the course of 10 years, remediated under licence where required. With the coming into force of the Nuclear Safety and Control Act (NSCA), 12 sites required some form of regulatory control. All of these remaining sites were located along the Great Bear River, which drains Great Bear Lake to the Mackenzie River. The information available to CNSC staff indicated that some elevated gamma radiation levels had been observed on site visits to these remote areas. Based on the potential for public health risks if the sites are developed, CNSC staff put in place institutional controls requiring the cooperation of other federal government agencies and the local First Nations Bands. These were reported to the Commission in CMD 01-M78 and an exemption form licensing the possession, management and storage of nuclear substances at these sites was granted. One of the conditions of granting the exemption was that CNSC staff would report on the status of the sites every five years.

This document describes the program undertaken to assemble the characterization information necessary to satisfy this reporting obligation and to present that information to the public.

Top of page

RSP-0174 – Comprehensive review of the effectiveness of waste rock management and decommissioning practices

Waste rock management at mining operations represents one of the major challenges that face the industry, particularly in the context of increasingly stringent standards for the protection of the environment. Predictions of future behaviour of mine rock effects on nearby water quality continue to have uncertainties. The uncertainties related to potential future impacts represent a challenge not only to mine operators but also to regulators that have the responsibility to approve mine rock management practices while upholding environmental protection legislation. The purpose of this report was to provide an overview of current approaches and practices related to the prediction of mine rock behaviour and of mine rock management.

The main objective of this review was to compare approaches or processes and possibly methods that are being applied to manage mine rock from uranium (and other) mining operations in Canada to those in other jurisdictions around the world. The focus of this review was on the environmental aspects such as effects on water and air quality as well as protection of humans and the natural environment. Issues related to physical concerns such as stability and slopes, for example, were not considered here. Contributions from Australia, Germany, Russia and South Africa were solicited and incorporated in this review to complement the North American experience, and sufficient information was available to compare at an overview level similarities and differences in approaches being applied to mine rock management and permitting.

In a very general sense, mine rock environmental issues arise because the process of mining uranium and other types of ore extracts rock from zones where it was formerly isolated from weathering and for the most part from biological processes. Mine rock is broken and the reactive surface area increases as it is placed in an active weathering environment. This increases the weathering rates and the rates of release of trace elements and radionuclides to the adjacent environment. This increase in release rates can cause unacceptable environmental or public health concerns.

It is evident that there are various possible approaches to mine rock assessment with a focus on water quality. One approach that is developing as a dominant theme is based on an assessment of potential release rates from rock with subsequent transport through the near-field environment, and finally with transport into the receiving environment where the resulting concentrations are estimated. In this approach, it is necessary to estimate release rates from the rock, if any potentially significant quantities are anticipated, and to determine if or how the loadings may be modified before discharge to the receiver. In general, the largest uncertainties in the predicted outcome using this approach are associated with the estimates of release rates, but there are also uncertainties associated with the modifying chemical reactions as affected drainage migrates through the near-field environment toward the receiver. Much of the effort associated with mine rock assessment has focused on methods and procedures to determine mine rock reactivity, and the potential to release metals and other chemicals of concern, including acid.

An extensive series of test methods and procedures are used to assess mine rock. These generally fall into the following broad categories: static tests (and other \ chemical characterization), kinetic tests (that focus on acid generation and release rates of chemicals of concern) and modelling (that, here, refers to any calculations to estimate the outcome for projected full-scale mine rock facilities as a function of time). One of the greatest technical challenges to date has been to make use of laboratory data to scale up to mine rock stockpiles or other rock repositories. In general, models are required for this purpose but there is no general agreement on a modelling approach and many models remain in the realm of research. Most practitioners’ models tend to rely on combinations of empirical data and more sophisticated models or spreadsheet calculations to assess long-term water quality evolution associated with mine rock facilities.

Some generalizations can be discerned from the perspectives of the various countries included in this review. All countries have a long history of mining of metals but uranium mining in Canada, Germany and Russia dates from the 1940s and in Australia from the 1970s. South Africa has not experienced uranium mining but has uranium associated with gold deposits that are similar in origin to the uranium deposits in the Elliot Lake region of Canada.

Current regulations that address uranium mining are relatively recent in most of the countries. The regulations have common themes for protection of humans from excess radiological dose that generally included consideration of dust and radon emissions in addition to aqueous releases. Most of the regulations also address water and air quality criteria for the protection of the natural environment from chemical and radiological toxicity.

All countries have altered the focus of mine rock management within the final decades of the 1900s from physical stability issues and radiological protection from direct exposure to more chemically and environmentally focused issues. With the exception of Russia which has been challenged by major social reorganization issues in the 1990s, the other jurisdictions share similar approaches to the evaluation of mine rock for planning and permitting. The awareness of acid drainage and metal leaching associated with sulphide mineral oxidation in mine rock stockpiles, has driven assessments to address water quality issues in rock drainage as a primary concern for rock management.

Another common theme across most countries is the scarcity or absence of fully reclaimed and closed mines (either uranium or other non-uranium mines with other environmental challenges) that have been returned to the respective government jurisdictions. Only Germany has taken significant steps toward large scale reclamation that was driven by reunification in the early 1990s and the legacy of uranium mining in the former East Germany where many operations were in close proximity to populated areas. The commitment of more than €6 billion by the federal government in Germany and the development of the mine reclamation organization known as WISMUT provided the means to conduct significant reclamation activities. The reclamation activities in Germany will represent an important learning experience as ongoing monitoring provides a performance assessment for different strategies that have been applied in that country.

There are examples of reclaimed and closed uranium mines in other countries, including Elliot Lake mines in Canada that did not have mine rock issues and Rum Jungle in Australia where the rock stockpiles were covered in the early 1980s. In almost all cases of reclaimed mines, however, there is a need for ongoing care and maintenance, including ongoing water treatment. And as discussed in the review from South Africa, the discharges from existing mine rock facilities (and mine sites in general) do not currently conform to regulatory requirements in the surrounding environments and therefore there are very few examples of fully closed and reclaimed mines with removal of liability from former operators.

It is evident that all countries are in a learning mode for mine rock reclamation and closure planning in this first decade of the 21st century. There are a variety of approaches for mine rock management that have been suggested or are being carried out. In Canada, in-pit and underwater disposal of reactive rock is favoured at uranium and non-uranium mines. Examples of soil covers on reclaimed reactive rock (including non-uranium) are scarce and there are no known successes walk-away for cover systems involving reactive rock.

There are also differences in details related to evaluation methods among the countries reviewed. It is also evident that there are differences in interpretation of the test results within Canada and that these differences reflect the state of our knowledge and ongoing learning and research that are needed to address water quality issues associated with rock drainage. Overall, however, it is clear that the approaches to mine rock management in Canada are as technically advanced as those in the other countries reviewed.

It is clear that mine waste management, with respect to water quality issues, is not a straightforward engineering issue. Much has been learned over the past two decades about the behaviour of sulphide mine rock subjected to weathering and oxidation and we, as practitioners, researchers, mining companies and regulators are continuing to learn and adjust our thinking and approaches to managing mine waste.

Because we do not have a full understanding of the issues and because we cannot, with certainty, predict outcomes for water quality generated by mine rock in all environments over long time scales, we need to be able to adapt our management of mine rock to new findings and observations, and ensure that alternatives and contingencies are in place to address the uncertainties. It is evident that we will continue to learn, by experience, to refine mine rock management strategies into the foreseeable future. We do not have the option of waiting for researchers to answer all of the questions and therefore need to take practical steps in planning and implementing good practice for mine rock management that provide options for cost effective mitigation as required.

In conclusion, it can be safely stated that the Canadian experience has not lagged behind those in other countries with respect to state-of-the-art practices for mine rock management. Although Germany has invested a large collective (and monetary) effort to address legacy issues associated with uranium mine wastes, and has succeeded in the primary goal of isolation of potentially hazardous materials from exposure to humans, there has not yet been sufficient time to assess the long-term effectiveness of reclamation activities for the protection of the environment.

Nonetheless, the German experience will provide ongoing feedback for various reclamation strategies and should be a valuable source of information into the future. The mine-environment community in Canada is very active and many of the world experts in mine waste management are acknowledged to be part of that community. With the continuation of mining as an important component of the Canadian economy, it is very likely that we will remain at the forefront of knowledge-based solutions to the challenges associated with mine rock management.

Top of page

RSP-0175 – Uranium concentrations in Port Hope soils and vegetation and toxicological effect on soil organisms

ECOMatters Inc. and subcontractors conducted a study in the Port Hope area funded by the CNSC specifically to obtain site-specific data relevant to environmental and human health assessments. Three aspects were identified by the CNSC as particularly important: data to verify models of long-term fate of uranium (U) in Port Hope soils, bioavailability of soil U especially for the soil-to-plant pathway, and effects of soil U on soil organisms. There was also a need to support a long-term monitoring plan. There were three approaches followed. The first was to sample in detail soil profiles where U concentrations were relatively high but where contamination was thought to be solely atmospheric. These profiles were useful to investigate the general mobility of U in soil. The second approach was to gather site specific plant/soil concentration ratios and this involved sampling vegetation and surface soil from a range of sites. A broader range of soil U concentrations could be included for these sites. The third approach was to evaluate potential ecotoxicity, and after consideration that few if any soils in Port Hope would have U concentrations high enough to show toxicity, a series of aged spiked soils including a soil from Port Hope were used. The bioassays were chosen from those developed and promulgated by Environment Canada.

After careful planning and consultation with local experts and landowners, sampling sites were identified and samples were collected in September 2002. Soil samples were solubilised and measurements of about 50 elements, including U, Th, Pb, As and Sb, known contaminants in Port Hope, were made by ICP-MS. Plant samples were ashed to improve detection limits and were similarly analysed by ICP-MS. Soil properties such as texture, pH, carbonate content and organic matter content were determined. Soil solid/liquid partition coefficients, Kd, were determined using centrifugally expressed soil pore water. Plant/soil concentration ratios, CR, were computed for about 70 different plant samples. The plants were chosen to represent both native and cultured plants, with emphasis on those that might be consumed by humans.

In general, the concentrations of U were positively correlated to those of Ag, As, B, Ba, Bi, Cd, Co, Cu, Mo, Ni, Pb, Sb, Sn, Tl and Zn with depth in the soil profiles, whereas there was a weak negative correlation with Th concentrations. This suggests the Th was native as opposed to a contaminant, and that the processes that influenced U distribution also influenced many other metals and potential inorganic contaminants. It is not clear if all these elements are from the same source, but they were considered co contaminants in the soil profiles. Several profiles showed monotonic decreases in U concentration with depth, and these were interpreted as being the result of migration of U deposited from the atmosphere. However, because the distribution of U in the soil profiles was similar to those of less - and more mobile -co contaminants, it is suggested that at least some of the migration may have resulted from physical processes such as particle migration. Profiles of soil solid/liquid partition coefficients (Kd) were obtained, and, in general, there were progressive changes in Kd with depth. However, sometimes Kd increased with depth and sometimes it decreased with depth. The relationship of Kd to soil pH conformed well to expectation from the literature.

Plant/soil dry weight concentration ratios (CR) for U were computed, and the geometric means (n, geometric standard deviations) were overall: 0.0068 (63, 4.9), fruit: 0.00076 (15, 3.5), vegetables 0.0041 (2, 2.0) and edible roots: 0.0093 (4, 4.9), trees, shrubs, non edible annuals and forages: 0.014 (43, 2.4). These conform well to expectation. It is probable that some portion of this U was as surface deposition on the plants that resisted washing with detergent, but the majority was probably present by root uptake. The key soil properties to influence plant uptake of U are pH and U concentration. Soil pH is quite uniform in Port Hope and an effect of soil pH could not been shown. Soil U concentration may be a useful parameter to account for a small portion of the variation in CR.

The soils used for the bioassays were those used previously by Sheppard et al. (1992). They had been stored moist, outdoors in covered, in-ground containers for over 10 a. This makes them ideal soils for study because the U was well aged and was not confounded by the presence of co contaminants. There were three soils, a fine sand forest soil (limed in 1992 to allow plant growth and earthworm survival), an organic rich garden soil, and a loam from Port Hope. Analysis of the soils showed that U concentrations had not changed, and ranged from background to ~1000 mg U kg 1 dry soil. The bioassays used were northern wheatgrass (Elymus lanceolatus) early seedling growth; earthworm (Eisenia andrei) acute (14 d) survival and a chronic (56 d) reproduction bioassay; and two soil arthropod (Onychiurus folsomi and Folsomia candida) reproduction tests. Preliminary screening showed no effect at ~1000 mg U kg 1, so a small aliquot of each soil was spiked to achieve ~3000 mg U kg 1. An additional series of bioassays with the soil arthropods was done to compare aged versus recently spiked soils. Only O. folsomi proved to be especially sensitive to U, in both the screening and definitive bioassays. The plant and earthworm bioassays were not affected by U concentrations of ~1000 mg U kg 1, and F. candida was not affected by U concentrations below about 350 mg kg 1. Survival and reproduction of O. folsomi were diminished to 20 percent less than controls (EC20, an interpolated value in this case between the no observed effect concentration and the lowest observed effect concentration) at U concentrations of 92 to 190 mg kg 1 in the limed sand soil, 710 mg kg 1 in the Port Hope soil and 480 mg kg 1 in the garden soil. The limed sand was an exceptional soil, very low in organic matter content and clay materials. In general, these data support the expected no effect value (ENEV) of 250 mg kg 1 derived from the literature for most soils and endpoints.

Page details

Date modified: