The sensitivity and magnitude matrix in an LVIA

GLVIA3 does not give you a matrix. It gives you a method for reaching a judgement and an obligation to explain it, and the matrices that appear in almost every Landscape and Visual Impact Assessment are a presentational device that practitioners added. Used well, a matrix makes the reasoning auditable. Used badly, it replaces the reasoning, and that is the weakness an advocate will go for at inquiry.

The distinction matters because the guidance is explicit that thresholds should not be applied mechanically and that professional judgement cannot be delegated to a table.

Why GLVIA3 resists mechanical matrices

The third edition of the Guidelines for Landscape and Visual Impact Assessment, published jointly by the Landscape Institute and the Institute of Environmental Management and Assessment, moved deliberately away from prescription. It sets out the components of the judgement, insists that criteria are stated and that reasoning is transparent, and declines to supply a formula.

The reason is that the same combination of inputs produces different answers in different places. A medium magnitude of change to a landscape receptor of high sensitivity may be a major adverse effect on a chalk escarpment and a moderate one in a fragmented urban fringe, because the character of the receiving landscape and the nature of the change are not fully captured by two labels.

So a matrix is a summary of a judgement already made, not the mechanism that makes it. Reports that present the matrix first and the reasoning afterwards get the order wrong, and it shows.

Value and susceptibility combining into sensitivity

For landscape receptors, sensitivity is a function of the value attached to the landscape and its susceptibility to the particular change proposed. Both halves have to be argued separately before they are combined.

Value draws on designation where it exists, but designation is not the whole of it. The guidance expects value to be assessed for undesignated landscapes too, drawing on landscape character assessments, local plan designations, recreational use, perceptual qualities such as tranquillity or relative wildness, scenic quality, rarity and any cultural or historical associations. State the evidence for each.

Susceptibility is specific to the development proposed. A landscape may be highly susceptible to vertical infrastructure and barely susceptible to low-rise development in the same location, so a susceptibility judgement that does not name the type of change is not a judgement at all.

For visual receptors the components differ. Susceptibility turns on the occupation or activity of the people affected and their expectations: residents at home and users of a promoted recreational route sit at one end, people at work in a commercial setting or passing on a trunk road at the other. Value attaches to the view itself, evidenced by designation, guidebook or interpretation board references, inclusion in local plan policy or provision of viewing infrastructure.

Size, scale, geographic extent and duration of change

Magnitude has three components, and reports routinely collapse them into one.

Duration is where wind and solar schemes differ from housing. A temporary consent with a decommissioning condition supports a different magnitude judgement from a permanent residential allocation, and the guidance expects that difference to be reasoned rather than assumed.

Geographical extent is the one most often conflated with the size of the effect. A change experienced across an entire character area is not necessarily a large change, and a severe change confined to one field is not necessarily a small one.

Reaching a reasoned significance judgement

The combination step is where the report earns its fee. What a reviewer, or an inspector, is looking for is a short passage of reasoning for each receptor that says: the sensitivity is this, for these reasons, the magnitude is this, for these reasons, and the resulting effect is this, for this reason.

The last clause is the one that is usually missing. Where a report gives high sensitivity, medium magnitude and "moderate to major adverse" with no explanation of why it lands between two categories, the author has recorded an output rather than made a judgement.

Three practices improve the reasoning:

1. State the criteria for each category before applying them, so the reader can see the scale being used.
2. Give the reasoning per receptor in prose, with the matrix as a summary table afterwards.
3. Where the judgement departs from what the matrix would suggest, say so and explain why. That is not a weakness in the assessment. It is the guidance being followed.

In an assessment forming part of an Environmental Impact Assessment, significance has a statutory function under the Town and Country Planning (Environmental Impact Assessment) Regulations 2017, which require likely effects of that order to be identified. Say plainly which effects meet that test. An Environmental Statement must be prepared by competent experts, which is a legal requirement rather than a preference.

Separating landscape effects from visual effects

The two are assessed separately, and merging them is the most common structural fault. Landscape effects are effects on the landscape as a resource: its elements, its character, its aesthetic and perceptual qualities. Visual effects are effects on specific views and on the visual amenity of identified people.

A development can have a limited landscape effect and a serious visual effect, or the reverse. A building tucked into a valley may barely alter the character of the area while dominating the outlook of four dwellings on the valley side.

Keep separate receptor schedules, separate magnitude judgements and separate conclusions. Where the assessment is of a townscape rather than a rural landscape, the equivalent framework sits in a Townscape and Visual Impact Assessment, with character areas and views treated in the same two-track way.

Mitigation, year 1 and year 15 residual effects

Mitigation that is part of the scheme design is assessed as part of the proposal rather than as a reduction applied afterwards. Additional measures are then assessed for what they achieve and when.

Landscape mitigation takes time, which is why residual effects are reported at more than one point. The convention is an assessment at year 1, meaning completion with planting newly established, and at year 15, by which time structure planting is assumed to have grown to a stated height and density.

Two things are worth stating explicitly. First, the assumed growth rates and the management regime the year 15 position depends on, because unmanaged planting does not deliver the modelled screening. Second, whether the planting is secured, by condition or obligation, with a replacement provision for failures. An inspector asked to accept a year 15 judgement will want to know what guarantees it.

Where screening is the whole mitigation case, say so. A scheme that is acceptable only once a hedge has grown for fifteen years is a different proposition from one that is acceptable on completion.

Where matrices are challenged at inquiry

Four lines of attack recur, and all four are avoidable in the drafting.

The first is that the author cannot explain the criteria. If "medium susceptibility" is not defined anywhere in the report, cross-examination will establish that in two questions.

The second is selective receptor choice: viewpoints positioned where the effect is least, or residential receptors grouped so that the worst affected disappear into an average. Residential amenity assessments are a related but separate exercise, and conflating them with viewpoint assessment is a recognised fault.

The third is the use of "moderate" as a resting place. Where a report reaches moderate for every receptor, the scale has stopped discriminating and the conclusion carries little weight.

The fourth is inconsistency between the matrix output and the narrative conclusion, which happens when a report is edited late. Check every cell against every paragraph before issue.

Landscape and visual work of this kind should be prepared by a suitably qualified landscape professional, and councils usually expect that; professional review is strongly advised, and the professional directory is the route to a reviewer. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.