The map picks the stretch, your eyes pick the spot
Most terrain reading can be done at a desk before you ever reach the stream. You can work out from a map how steep a valley is, where its gradient changes, and where its bends, junctions and constrictions are. That lets you spend your hours on the road and your fuel on the right valley. A day spent on the wrong stream costs far more than an evening spent looking at a map.
But it is worth knowing the map’s limits from the start. A map tells you which stretch is worth walking; it does not tell you which crack to dig in that stretch. Individual boulders, bedrock cracks, the current position of bars and the state of the gravel can only be seen on the ground. The right division of labour is simple: use the map to choose the stretch and your eyes to choose the spot.
The tools in this section complement each other: contours and the long profile show gradient, hillshade shows the shape of the ground, the geological map shows the source, and satellite imagery shows what the bed looks like today. None is enough on its own; read together, they give a fairly reliable picture of a valley.
Contour lines and the long profile
Contour lines join points of equal height. Where a stream crosses them, the contours form a V or U pointing upstream; that is the quickest way to tell from a map which way a stream flows. The spacing of crossings along the stream tells you the gradient: crossings packed close together mean a steep, fast stretch, and crossings spread apart mean the stream is flattening.
To find a slope break, follow the stream downhill and watch the spacing of the crossings. Where several come in quick succession and then the gap suddenly widens, the gradient has dropped. Mark that point; the traps are just below it. If you want gradient as a number, divide the height difference between two points by the distance between them: a stretch that drops thirty metres over one kilometre has a gradient of three percent.
A long profile makes the same information faster to read. It plots the stream’s height against distance from its head. A natural stream profile is usually a concave curve, steep at the top and gentle lower down; the departures from that smooth curve are what is interesting. A sudden flattening after a steep stretch shows up as a shoulder or step in the profile and marks a trap zone. When comparing several streams, profiles are far quicker than contours.
Hillshade, DEMs and the limits of 30 metres
A digital elevation model (DEM) is a grid in which every cell holds the ground height at that point. Everything about gradient, catchment, flow direction and stream shape is computed from it. Hillshade is made from the DEM too: the terrain is drawn as if lit by an artificial light from one direction. Ridges, side valleys, benches and old channels stand out as light and shadow, and flat steps that contour lines alone can hide suddenly become visible.
Hillshade has one catch: the direction of the light can emphasise some shapes while flattening features that run parallel to it. Checking an important valley with more than one light direction, or with a multi-directional hillshade, is a good habit for that reason.
Resolution decides how small a feature a DEM can see. A 30-metre DEM sees the valley: gradient, bends wider than a few cells, junctions and constrictions. It does not see a single boulder, a bedrock crack, or a bar narrower than its cell. In narrow, deep gorges the channel can be narrower than one cell, and the model may blur it into the valley wall. Some models also include vegetation and buildings in the height. So every result from 30-metre data should be read as “this stretch is worth a look”, not “dig this spot”.
Mapped stream lines are approximate for the same reason. Channels move, floods open new branches, and a line computed from a DEM can sit tens of metres from the real bed. Use the line to choose the stretch and its features; confirm the actual channel and bar positions on the ground or with recent imagery.
Catchment and geology: the source question
However good the trap geometry, if there is no gold source in the catchment there is no gold in the stream. So the first question to ask of a map comes before geometry: the source. To read a catchment, trace the ridges around the stream above the point you plan to sample: the divides where water would flow one way or the other. Everything inside that line is the catchment, and everything that erodes upstream passes through your point.
Then look at what is inside it. A geological map shows which rocks are exposed. Quartz veins, zones of hydrothermal alteration and old volcanic rocks are the usual gold sources. A stream draining only young sediments is unlikely to carry gold no matter how good its traps look. Mineral occurrence maps take this a step further: every marked vein, prospect or old working is a possible source, and occurrences inside a catchment are worth far more than ones beside it.
Judge junctions with the same eye. Follow tributaries upstream on the map and see what they drain. A small branch coming from steeper, more mineralised ground is worth more than a bigger one draining gentle farmland. Stream order is a practical filter too: headwater streams with no tributaries are first order, two of them joining make a second-order stream, and so on. For hand prospecting, walkable low- and middle-order streams are usually the most practical.
Reading satellite imagery
Satellite imagery shows something a DEM cannot: what the bed looks like today. Pale, crescent-shaped strips on the inside of bends, trailing downstream, are gravel bars. They stand out clearly in images taken at low water and vanish at high water. So always check the capture date; concluding that a bar “isn’t there” from a high-water image is an easy mistake.
Bedrock is sometimes visible: a darker, blocky texture in the channel, different from gravel, often where the valley narrows or the stream bends sharply. White-water rapids and short steep steps in the image are also common signs of bedrock underneath. These help you mark in advance the stretches where bedrock clean-up may be possible.
Old channels usually give themselves away through vegetation. Because gravel drains differently, plants over an old bed grow in a different colour or density. Look for a curving line of different vegetation running parallel to or across the present stream, or a pale strip of gravel across a meadow. For terraces, use hillshade and imagery together: if a flat shelf in the hillshade also shows up as a different land use or plant pattern in the image, it deserves a closer look.
Reading a valley at your desk
When choosing between two valleys, ask three questions in order. First: is there a source in the catchment? Is there mineralised rock on the geological map, a mark on the occurrence map, any trace of old workings in the area? Second: does the profile show slope breaks, bends and junctions? Third: can you get there? A valley that fails the first question is out, however good the other two look.
In a valley that passes the first question, start at the first good traps below the source. Gold gets finer and sparser with distance from its source, so the first slope break, the first bend or the first junction below mineralised ground is often the best stretch. Mark those points on the profile, then look at each in hillshade and in imagery: is there a bar below the break, does gravel show on the inside of the bend, can you pick out a bedrock texture?
By the end you should have a short list of three to five spots: why each was chosen, what kind of trap you expect, and how far it is from the road. That list is a working plan for the field, not a promise. On the ground you will find some spots are not what you expected; the bed may have shifted, a bar may have moved, bedrock may be buried under gravel. The map reading still pays off then, because you already know where the next candidate is.
- Is there a source in the catchment? If not, drop the valley
- Mark slope breaks, bends and junctions on the profile
- Check each spot in hillshade and in low-water imagery
- Start at the first good traps below the source
- Write the three-to-five spot list with your reasons
Access, route and checking on the ground
Access decides what you can actually work. A rich-looking stretch hours from any track is a completely different plan from one beside the road. Before choosing between two similar stretches, check tracks, stream crossings and where you can leave a vehicle. Measuring the length of the stretch you will walk and its distance from the road lets you estimate realistically how many spots you can work in a day.
Walking upstream along the stream is usually a good route. You see the traps from below, the way the water sees them, and you finish at the upper end, where gold may be coarser. Mark the spots to sample on the way up and work them on the way down. You can read gradient in the field without tools, too: noisy rapids mean a steep stretch, quiet pools and long riffles mean it has flattened. The place where the stream’s character changes may be the very break you marked on the map.
Check protected areas before you go. National parks, nature reserves and heritage sites are usually mapped; GOLDONGO marks them and drops the streams inside them from its lists. Whatever map you use, rules on prospecting and land access vary from place to place; learn the ones that apply where you are before you go. What GOLDONGO does is the geometry reading this section describes — slope breaks, inside bends, junctions, constrictions — across the whole stream network from 30-metre elevation data; a high score points to geometry that could hold heavy minerals, not to gold.
- Check tracks, crossings and parking in advance
- Measure the stretch length and distance from the road
- Walk upstream and mark spots on the way up
- Learn the protected areas and local rules