solar powered GPRS anemometer

Hello.
Im trying to develop solar low-maintenance anemometer that sends data trough GPRS to my website. I'm a self learner coming from medical background and have no formal training in electronic field. I did however learn quite much software programming in last few years so i think software part of the project should be fine is't the hardware I struggle with.

For my project I would use:
arduino board - probably nano
gprs module SIM800L - is this the best option ?
battery and solar cell
anemometer with wind vane - Davis 6410

As explained in LINK connecting anemometer to arduino should be easy

My question is mostly about battery and solar cell. I watched some tutorials on youtube and online but most of them seems to be overly complicated.
What kind of combination of solar cell and battery would be best for this project ? Im confuesd with so many options. Should I go for 6V, 12V or 4.7v ? What about battery capacity and sollar cell size?
Any help is appreciated.

One question to start. How often must the data be transmitted ? Do you need to see the results in more or less real time or can say data collected over the last hour be sufficient ? A GPRS module has a relatively high power consumption and this has an impact on the design of your power supply.

Thank you for your response. To clarify, arduino would be in sleep mode, then woken up to read the data from anemometer lets say every 10 seconds. Average wind speed, wind gust and direction would be saved in RAM and then POST request would be made to my website in ether 5 or 10 min (haven't made final decision) intervals and 15-30 min during night time.

One additional question regarding battery. Should I go with li-ion or nimh? I was planing on using li-ion but i read somewhere that if battery is constantly charged it is better to use nimh?

For low power projects the Nano, Uno etc are not good choices because these have a lot of power hungry components which limits the benefits of sleep mode. Use a "barebones" ATMEGA328P (the chip used by the Uno, Nano etc.) or use an Arduino Mini and disconnect the LED and voltage regulator. If the GPRS module is 3 volt, a 3 volt mini would be OK.

For the choice of battery technologies, maybe someone else can answer. I did, however, find this which could help: Arduino and GSM Module running on Li-primary batteries for years - Electrical Engineering Stack Exchange

What kind of combination of solar cell and battery would be best for this project?

To do this correctly with minimal expense IS complicated, and can't be done at all until you know how much power your system will consume on average. That depends on how often you send data as well as standby current consumption.

However, battery and panel calculations can be done after the system is put together and running. Then, you measure the average current draw and plan accordingly. There are web sites that help you with this, depending on your location and average weather patterns.

Or, if you demand simplicity, buy a big lead acid battery, a charge controller, a step down converter for the electronics, and "big enough" solar panel, say 50 Watts. If the system stops working during long cloudy periods, buy a bigger solar panel.

For a truly minimal system, see this excellent tutorial: Gammon Forum : Electronics : Microprocessors : Solar powered Arduino

Thank you both for your replies.
@6v6gt I was also thinking of using arduino pro mini and turn off LED. GPRS modul works from 3.7-4.4V. Would it be possible to use it with 3.3V pro mini or only 5v? Would using 3.3V change anything in regards of the first picture in the post with the scheme that shows how to connect anemometer ? More specific regarding yellow wire and and 4.7 resistor. Please forgive my ignorance and I apologize to everyone that is reading this nonsense but where it says +5V on the yellow wire it doesn't mean that battery should be plugged there right? Also would it work on 3.3V arduino ?

@jremington I will measure power when I receive some parts and put them together. I understand that is difficult to plan ahead and I'm not looking for exact calculations. Just a rough estimation so I can buy some parts and do field testing. It usually takes about a month for the parts to arrive from China and at the moment I'm only looking for compatibility not so much optimization.

So if we only talk about voltage not capacity.
Would arduino board with 3.3V be sufficient or better to go with 5v?
Is it ok to use 6V solar panel or is there a preferred option?
What about the battery. What kind of voltage should I get?

If you use the Sparkfun (or similar) version of the 3.3v Arduino Pro Mini, there is a jumper which you can cut to isolate that voltage regulator and the power led. You can then use a stabilised voltage source (say a battery) of between 3 and 5 volts applied on the pins Vcc and GND .

The "+5V" end of the pullup resistor can also be connected to Vcc on the Arduino board which you select.

My project is almost done. Some test were made and the station seems to run well.
Code is aveliable on github GitHub - jaka87/vetercek_WS: Vetercek.com weather station

If would be great is someone would find this useful or even do some commits.