Clearance screening

Clearance screening compares a radioactive inventory against per-nuclide clearance levels. The clearance index CI = sum_i A_i / CL_i adds up each nuclide’s activity divided by its clearance level, and the sum-of-fractions rule treats CI <= 1 as satisfying the screening criterion (the boundary included). Nucleide’s nucleide.alara helpers run that arithmetic over plain {nuclide: activity} dicts, and nucleide.fispact reads the clearance-bearing inventory table from FISPACT-II output. This tutorial shows both; the implementations live in crates/alara-io and crates/fispact-io.

Reading the FISPACT-II clearance block

FISPACT-II prints a wide inventory table with a per-nuclide clearance column when the input deck sets the HAZARDS and CLEAR keywords. fispact_parse_clearance reads that table from the output text — one dict per (nuclide, time step) row:

from pathlib import Path

from nucleide.fispact import fispact_parse_clearance

text = Path("fixtures/fispact/clearance.out").read_text()
rows = fispact_parse_clearance(text)
print(len(rows))

step1 = [r for r in rows if r["interval"] == 1]
for r in step1:
    print(r["nuclide"], r["flags"], r["activity_bq"], r["clearance_index"], r["half_life_s"])

7
V-55 > 0.0 0.0 -1.0
Fe-56 #> 0.0 0.0 -1.0
Co-60 > 4000000.0 1000000.0 166360000.0
Rb-86m & 8000000000.0 4000000000.0 61.0

Each row carries the 1-based interval, the step time in seconds (time_s) with its verbatim time_label, a cooling flag (true for COOLING TIME IS steps), the nuclide name in the shared dialect spelling (Co-60), the verbatim flag characters, the activity_bq column, the half-life in seconds, and the code’s own clearance_index column — FISPACT-II’s per-nuclide A_i / CL_i against its internal clearance library, not against a caller table. Stable nuclides are stored with half_life_s == -1.0 (V-55 and Fe-56 above). The flags keep FISPACT-II’s markers verbatim: # stable, > present before irradiation, & gamma spectrum approximately calculated, ? convergence not reached.

Screening against the default clearance table

alara_clearance_index computes the clearance index over an inventory dict. With limits=None (the default) the comparison uses Nucleide’s vendored transcription of EU 2013/59/Euratom Annex VII Table A — 260 activity-concentration limits in Bq/g for solid material. Because that table lists concentrations, the inventory must be massic activities (Bq/g), not total activities (Bq):

from nucleide.alara import (
    alara_clearance_eu_table,
    alara_clearance_index,
    alara_sum_of_fractions,
)

eu = alara_clearance_eu_table()
print(len(eu), eu["Co-60"], eu["H-3"])

inventory = {"Co-60": 0.05, "Cs-137": 0.2}  # Bq/g
print(alara_clearance_index(inventory))

out = alara_sum_of_fractions(inventory)
print(out["sum"], out["class"], out["max_nuclide"], out["max_fraction"])

260 0.1 100.0
2.5
2.5 exceeded Cs-137 2.0

alara_sum_of_fractions returns the same sum plus the screening class ("satisfied" when sum <= 1, otherwise "exceeded") and the dominant contributor — the nuclide with the largest single fraction A_i / CL_i (here Cs-137 at 0.2 Bq/g against a 0.1 Bq/g limit). Nuclide keys accept any shared-dialect spelling ("Co60", "Co-60", and "co-60" resolve to the same nuclide); results report the canonical spelling.

Spanish conditional tables

For Spanish conditional NORM landfill screening, alara_clearance_es_table selects one of the three CSN Tables 1–3 explicitly by landfill type; the caller picks the governing table and there is no cross-table logic:

from nucleide.alara import alara_clearance_es_table, alara_sum_of_fractions

es = alara_clearance_es_table("non_hazardous", "slags")
print(len(es))

inventory = {"Ra-226": 0.2, "K-40": 5.0}  # Bq/g
print(alara_sum_of_fractions(inventory, es)["class"])

40
satisfied

The class boundary

The boundary sits exactly at sum == 1 and belongs to the satisfied side: an inventory sitting exactly at its combined limits classifies as satisfied, and one float step above flips the class:

toy_limits = {"Co60": 10.0, "H3": 5.0, "Fe55": 2.0}

at = alara_sum_of_fractions({"Co60": 10.0}, toy_limits)
print(at["sum"], at["class"])
above = alara_sum_of_fractions({"Co60": 10.000000000000002}, toy_limits)
print(above["sum"], above["class"])

1.0 satisfied
1.0000000000000002 exceeded

A custom limit table

Pass a {nuclide: limit} dict as limits to screen against site-specific or national values instead of the EU table. Activities and limits must share one unit basis — here both sides are total activities in Bq, so the comparison is consistent even though the numbers look nothing like Bq/g concentrations:

print(alara_clearance_index({"Co60": 10.0, "H3": 2.5}, toy_limits))

1.5

Every inventory nuclide needs a table entry, and every activity must be finite and non-negative; violations raise a clear error instead of being skipped:

try:
    alara_clearance_index({"Mn54": 1.0}, toy_limits)
except ValueError as exc:
    print(exc)

nuclide Mn54 has no clearance limit (table holds 3 entries)

From a parsed inventory to a screening result

Tying the two halves together: take one time step from the parsed FISPACT-II block and screen it against a caller-supplied limit table in the matching basis (total Bq on both sides):

step1_inv = {r["nuclide"]: r["activity_bq"] for r in rows if r["interval"] == 1}
limits = {"V-55": 1.0, "Fe-56": 1.0, "Co-60": 4.0e6, "Rb-86m": 8.0e9}
out = alara_sum_of_fractions(step1_inv, limits)
print(out["sum"], out["class"], out["max_nuclide"], out["max_fraction"])

2.0 exceeded Co-60 1.0

Screening arithmetic, not a compliance decision

These helpers are screening arithmetic, not a compliance decision. Real clearance calls for the governing regulatory table for the material and pathway, complete material bookkeeping, and the national transposition of the underlying directive. The vendored EU table is a transcription of one directive’s annex for solid materials, and the Spanish tables transcribe the CSN conditional NORM landfill levels; a sum below 1 marks an inventory as passing this arithmetic check — not as cleared for release.