Geography Paper 3 June 2023

Answer this when you sit the paper.

Answer this when you sit the paper.

Answer this when you sit the paper.

Answer this when you sit the paper.

Answer this when you sit the paper.

Answer this when you sit the paper.

Answer this when you sit the paper.

1. AReject the null hypothesis: there is a significant association, so age and sex are dependent factors in these cases.
2. BThe test is invalid because a chi-square statistic that small indicates the expected frequencies were calculated wrongly.
3. CNo conclusion is possible without also calculating the correlation coefficient between the age and sex columns.
4. DDo not reject the null hypothesis: there is no significant association, so age and sex are independent factors in these cases.

Answer this when you sit the paper.

1. AClassical tests assume observations are independent, and spatially autocorrelated residuals break that assumption, giving artificially small standard errors and false significance.
2. BIt converts ordinal data into interval data, which is what classical tests such as the t-test are designed to accept as input.
3. CIt removes the need to sample at all, because a spatially autocorrelated variable can be measured once and applied to the whole study area.
4. DIt guarantees that the distribution of the variable is normal across the study area, so a parametric test such as the t-test can always be used in place of a non-parametric one without any further checking of the data.

Tap to enlarge

1. AThey are terminal moraines, ridges of unsorted debris left along the front of an ice sheet that formerly covered the plateau surface and stood still long enough for the material it carried to be dumped in place.
2. BThey are volcanic cones built up by successive eruptions above a hot spot beneath the plateau.
3. CThey are resistant granite masses that survived while more heavily jointed rock around them was weathered and stripped away.
4. DThey are fold mountains raised by compression where two continental plates collided along this part of the craton.

Tap to enlarge

Answer this when you sit the paper.

Tap to enlarge

1. ARadial, indicating a central volcanic cone or dome from which all the streams flow outward down its flanks.
2. BTrellis, indicating alternating bands of hard and soft rock which the tributaries follow along the strike.
3. CDendritic, indicating uniform, structurally undeformed rock of even resistance, which a granite plateau provides.
4. DDeranged, indicating a surface recently disorganised by glacial deposition with no established drainage lines.

Tap to enlarge

1. ANucleated settlement, explained by the need for households to cluster tightly around a single defensible water point that can be relied on through the dry season when the surrounding streams have stopped flowing.
2. BLinear settlement, explained by households strung out along the tarred road to gain access to markets and transport.
3. CDispersed settlement, explained by a subsistence farming pattern in which each household lives on its own arable holding near its fields and water.
4. DPlanned settlement, explained by a resettlement scheme that allocated regularly spaced plots of equal size across the area.

Answer this when you sit the paper.

Tap to enlarge

1. ASeveral spread points allow a transformation to be fitted across the whole sheet and its residual error to be measured, so distortion is corrected everywhere rather than only locally.
2. BThe software requires exactly six points and will not accept any other number, so the count is fixed by QGIS rather than chosen by the analyst.
3. CSpreading the control points to the corners increases the spatial resolution of the scanned raster, so that more ground detail becomes visible after the fit than the original scan was able to record.
4. DEach additional point permanently increases the map's scale, so six points convert a 1:50 000 sheet to a larger scale suitable for measurement.

Answer this when you sit the paper.

Tap to enlarge

1. AA permeable, deeply weathered granite regolith that takes water in as throughflow, combined with gentle relief and a strongly seasonal rainfall regime.
2. BAn exceptionally high annual rainfall total, which has eroded the smaller channels away and left only the largest valleys visible.
3. CRecent glacial scour, which stripped the established drainage lines from the higher ground and left channels surviving only in the sheltered valley bottoms where the moving ice could not reach them.
4. DAn impermeable clay subsoil across the sheet, which sheds almost all rainfall as surface flow within minutes of it landing.

1. AThey fix the coordinate reference system of a project, since a GIS derives its projection from the units in which the first measurement is taken.
2. BThey are the raw material of derived indices such as drainage density and population density, they allow change to be quantified between dates, and they underpin buffering, overlay and site suitability analysis.
3. CThey are needed only to print a scale bar in the map layout, since every other GIS operation works on attribute tables rather than on geometry.
4. DThey convert a raster layer into a vector layer automatically as soon as a length or area is read off it, which is the only route by which features can be stored in a GIS database and later queried by attribute.

1. AAnalogue processing uses satellite imagery while digital processing uses aerial photography, the difference being the platform the image came from.
2. BAnalogue processing works visually and manually on a continuous hard copy image, while digital processing manipulates numerical pixel values by computer, so it is quantitative and repeatable.
3. CAnalogue processing can only be applied to black and white images, since the eye cannot judge tone reliably in colour, while digital processing is what makes any interpretation of colour or false colour composites possible at all.
4. DAnalogue processing is carried out before the image is captured and digital processing afterwards, so the two happen at different stages of the same workflow.

Tap to enlarge

1. AEnhancement is performed by the sensor at the moment of capture while restoration is performed by the analyst afterwards.
2. BEnhancement is subjective and improves appearance for human interpretation without adding information; restoration is objective and removes known defects using a model of the degradation.
3. CEnhancement is applied to multispectral images and restoration only to single band panchromatic images.
4. DEnhancement compresses the image so that it can be stored and transmitted economically, and restoration decompresses it again for display, so the two stages are simply the opposite halves of one file handling operation.

Tap to enlarge

Answer this when you sit the paper.

Answer this when you sit the paper.

1. ARadiometric resolution, allowing 4096 rather than 256 brightness levels to be recorded, so subtle differences in reflected energy can be separated.
2. BSpectral resolution, allowing several hundred narrow wavelength bands to be recorded instead of four broad ones.
3. CTemporal resolution, allowing the same place on the ground to be revisited and imaged daily instead of once every sixteen days, so that rapid changes such as flooding or crop growth can be tracked through a season.
4. DSpatial resolution, allowing objects half a metre across to be identified where previously 30 m was the limit.

Tap to enlarge

1. AA frontal inversion: warm air riding up over a wedge of denser cold air is at its thickest above the centre of the built up area, so the trapped layer beneath it is warmest there and cools away towards the edges.
2. BA katabatic effect: cold dense air drains off the surrounding higher ground at night and ponds at the city margins, leaving the centre relatively warm.
3. CA rain shadow effect: the built up area lies in the lee of high ground, so descending air warms adiabatically over the city centre.
4. DAn urban heat island: dense hard surfaces absorb and slowly release solar radiation, evapotranspiration is minimal, anthropogenic heat is added and street canyons trap outgoing longwave radiation.

Tap to enlarge

1. AWater reflects almost all incoming solar radiation, so the river has a far higher albedo than any built surface in the city.
2. BThe river channel lies well below the general level of the city, so cold air draining into it is compressed and cools further.
3. CFlowing water generates no friction with the air above it, so no sensible heat can be transferred from the surface at all.
4. DOpen water has a high specific heat capacity so it warms slowly, and evaporation from its surface consumes latent heat, cooling the air above.

Answer this when you sit the paper.

1. AA dispersion diagram: plot every one of the ninety yearly totals as a dot on a single vertical axis, so that the spread of the record, its median and its interquartile range can all be read directly from the column of dots.
2. BA departure from the mean bar graph: calculate the long term mean, plot each year's total minus that mean as a bar above or below a horizontal zero line representing the mean.
3. CA pie chart: convert each year's rainfall to a percentage of the ninety year total and draw it as a segment, so that wet years occupy the larger sectors.
4. DA scatter graph: plot rainfall against year and fit a best fit line, then read the correlation coefficient to measure how variable the record has been.

1. AIt requires the underlying data to be normally distributed before the bars can be plotted, and long rainfall records are almost never normally distributed, so the technique cannot lawfully be applied to them.
2. BIt hides the absolute totals and depends entirely on the base period chosen, so changing the years used for the mean changes every bar on the graph.
3. CIt cannot show years that were drier than average, since a bar has no way of representing a negative quantity.
4. DIt cannot show more than about ten years of record, because beyond that the horizontal axis becomes mathematically undefined.

Tap to enlarge

Answer this when you sit the paper.

Tap to enlarge

1. AIsochrones must be drawn so that they pass exactly through the plotted point values themselves and are never interpolated between two of them, since any interpolated position would be an estimate rather than a measurement.
2. BEach isochrone must be labelled with the number of data points it encloses rather than with the travel time it represents.
3. CEach isochrone must be drawn as a perfect circle centred on the hospital, since travel time increases equally in every direction.
4. DAn isochrone must never cross another isochrone, and it either closes on itself or runs off the edge of the frame rather than ending inside the map.

1. AWidely spaced isochrones show where the data points were most numerous, since more points force the lines further apart.
2. BThe spacing is fixed by the isoline interval chosen and so carries no information about the ground at all.
3. CClosely spaced isochrones show where travel time rises steeply over a short distance, so they identify the poorly served directions.
4. DClosely spaced isochrones show where the population density is highest, since more people are recorded in those areas.

1. AQuota sampling, whose advantage is that it allows the researcher to choose the most typical site within each sub-area by judgement.
2. BSystematic sampling, whose advantage is that it gives completely even coverage of the whole study area for the least fieldwork, since the interval is fixed in advance and no random numbers have to be generated.
3. CSimple random sampling, whose advantage is that it removes all need for prior knowledge of the study area before sampling begins.
4. DStratified random sampling, whose advantage is that every sub-group is represented in proportion to its size, so a small but important stratum cannot be missed.

1. AIt requires prior knowledge of the sub-groups within the population, which is rarely available before fieldwork begins.
2. BIt can be badly biased where the sampling interval coincides with a periodic pattern in the ground, such as ridge and furrow at the same spacing.
3. CIt cannot be used in the field at all, because fixed intervals cannot be measured accurately without surveying equipment.
4. DIt gives uneven coverage, since regular intervals leave large parts of a study area with no samples at all.

Answer this when you sit the paper.

1. AFires occur in one sharply defined late dry season, rising from July to a peak in September and October, which together hold 55 per cent of the total, then falling abruptly in November.
2. BFires are spread fairly evenly through the whole year, with only a slight rise towards the middle months, so no single season stands out as more dangerous than any other and control effort must be maintained continuously.
3. CFires rise steadily from January to a single peak in December, following the onset of the rainy season.
4. DFires occur in two separate seasons of equal size, one in the early part of the year and one in the later part.

Answer this when you sit the paper.

1. AThey are surface fires burning the cured grass layer and litter in the late dry season, spread rapidly with the wind and uphill, and are largely human ignited.
2. BThey are crown fires burning through a closed canopy, occur mainly in the wet season when fuel is abundant, and are almost always started by lightning.
3. CThey burn slowly downhill against the wind, occur only at night when humidity is highest, and are confined to areas of dense evergreen forest.
4. DThey are subsurface peat fires that smoulder underground for months at a time, are unaffected by wind direction because no flame reaches the surface, and occur uniformly through the year wherever organic soils have dried out.

1. ABurning the strip only once a wildfire is already approaching it, so that the two fires advance towards one another and meet head on at the boundary, where each removes the fuel the other needs and both go out.
2. BBurning the strip at the height of the dry season in October, when the grass is fully cured so that the strip burns completely in a single pass.
3. CBurning the strip at the start of the dry season in June or early July while the grass is still partly green, so the fire burns cool and controllable and the strip is already bare before the dangerous season arrives.
4. DBurning the strip immediately after the first rains in November, when the new growth is green enough to carry a fire that cannot escape.