You need to avoid data collection bias due to different sampling methods of capture. The first way to do this is by ensuring your sampling is randomised so that every fish has an equal chance of being sampled. Then, if you cannot use the same method of capture across all river habitats, you will have to 'oversample'. Oversampling can be used to avoid sampling bias in situations where members of defined groups are underrepresented, as may happen in your situation :)

https://www.alchemer.com/resources/blog/how-to-avoid-sampling-bias-in-research/#:~:text=Use%20Simple%20Random%20Sampling,in%20the%20study%20at%20hand.

https://www.scribbr.com/methodology/sampling-bias/

Kindly take help of local fishermen community as they have more indigenous knowledge of fish population and its availability in vicinity

Why did you specify 10 sampling sites in a 40 km stream? What are your criteria? You can determine the sampling areas in streams according to zoning (water temperature, oxygen status, flow rate, etc.). In general, rivers are divided into 4 zones in the area from the source to the point where they adding into the sea. There is a distribution of fish according to these regions. I don't know if the properties of the water you are examining are other than these. However, the important thing is that you can obtain examples from these areas. Whatever method you use, you can calculate its abundance and distribution based on unit catch effort (CPU). However, you should aim to catch different-sized fish species and different-sized individuals of these species. For this, you can design a gill-net that you can create by designing nets of equal length together, using from the smallest mesh-sized to the largest mesh-sized, in a way that completely closes the water column (vertical). You can sample at each station you specify, by leave this gill-net in equal time. Thus, you can sampled demersal and benthic fish species. You can have an idea about the future of the population according to the rate of fry you will catch.

My experience is in marine systems, not riverine, but I think he problems are the same: Almost every fish-sampling method we have is selective to some (unknown) degree. Catches taken by different methods are not directly comparable and it is almost impossible to inter-calibrate them. If the same sampling method cannot be used at every sampling site, then the answer to your question is probably "no", sampling effort cannot be standardized.

Some of the work by Corey Dunn could be useful to you.

(PDF) Assessment and diversity of fish communities in non-wadeable tributaries of the Missouri and Mississippi rivers (researchgate.net)

(PDF) Accounting for dispersal and local habitat when evaluating tributary use by riverine fishes (researchgate.net)

(PDF) A flexible survey design for monitoring spatiotemporal fish richness in nonwadeable rivers: optimizing efficiency by integrating gears (researchgate.net)

are you sampling a single (maybe migratory) species? if so, mark recapture methods may be your best bet for comparing abundance between upstream and downstream reaches. If you are sampling multiple species, then I would refer you to these studies:

Article Fish zonations and guilds as the basis for assessment of eco...

Article Fish environmental guilds as a tool for assessment of ecolog...

Welcome to the wonderful Brazilian palaeogeomorphological heterogeneity! A reality that few researchers from countries other than the Neotropical region understand.

I sampled in similar conditions in the past. I suggest:

Article Fish community response to environmental variations in an im...

Article Idade e crescimento de peixes de riacho: métodos e desafios ...

I also suggest the special issue of Oecologia Australis: Ecologia de Peixes de Riachos (Ecology of Stream Fishes). During its production, all authors discussed about sampling standardization and produced some chapters that could be interesting for you:

https://revistas.ufrj.br/index.php/oa/issue/view/1746

Andrew Paul McKenzie Pegman thank your for the answer. I'll keep that in mind.

Estevan Luiz da Silveira thank your for sharing the literature. I'll take a look at it.

For all, thanks for contributing with your answers. I will consider them all.

Standardazing the abundance of fish will be restricted to the areas that were sampled with the same fishing gears.

I would be careful about how you compare counts of fish captured along a longitudinal gradient that may include variation in temperature or other factors that might effect capture probability. You can standardize by gear but if you can, try to use a method where you can account for capture probability. Attached is a recent analysis we completed that has some good background.

This paper on multigear standardization may be useful for your analysis: Article Canadian Journal of Fisheries and Aquatic Sciences

Large rivers are really tough to sample. They're big, dangerous, and diverse both physically and ecologically. Their habitats also scale with stream size making it tough to keep absolute sampling effort proportional to river size.

We developed a standardized multi-gear approach for sampling rivers, but the focus was on detecting richness versus composition. The community data, however, could be relativized to be an informative index of community composition. We attempted to create a multi-gear version of the most common status quo approach to sampling rivers (electrofishing-only).

The straightforward takeaway from my team's work across the nine different rivers that we sampled is that rivers have far more species locally present than even high-effort with the single-most effective gear would detect. Some recent meta-barcoding approaches have come to similar conclusions.

This is our methods paper for sampling rivers: Dunn & Paukert. 2020. A flexible survey design for monitoring spatiotemporal fish richness in nonwadeable rivers: optimizing efficiency by integrating gears. Canadian Journal of Fisheries & Aquatic Sciences: Article A flexible survey design for monitoring spatiotemporal fish ...

Good luck