As others have said, there are several areas where the hard sciences can benefit you heavily and give you a major advantage in your field:

1. Statistics. Specifically, you want statistics with calculus. The lack of a requirement of calculus and calculus-based statistics in many social sciences results in people in those fields using inherently flawed models. Furthermore a proper knowledge of statistics allows you to avoid debacles such as the infamous entanglement between Salzman and Ralph Mathisen, who as Chief Programmer for the Space Medicine Laboratory and author of original research publications on statistical analysis of human life expectancy, absolutely annihilated Salzman's math.
2. Information Systems/Programming/etc. The world of history is moving in the direction of digital humanities, and I've seen the research presentations of grad students in Classics developing programs to assess/break down translations by context and compare multiple ones side by side for users to develop the most accurate understanding of the original phrase (basically it's trying to address the "lost in translation" bit of translating). And that's just one example of how programming is becoming very, very widely utilized in the world of classics/history/humanities.
3. Chemistry/Materials Science. Archaeology uses a variety of methods you'll learn in chemistry or better yet materials science. Scanning Electron Microscopy, X-Ray Diffraction, etc. etc. A lot of techniques from the world of electrochemistry are also used in surveying to analyze soil or the ocean floor for archaeological sites. When I took my archaeology courses and they started talking about these techniques I already knew how to do pretty much all of them from my research in nanomaterials and supercapacitors. Chemistry and Materials science also have the added benefit of giving you the necessary background needed in experimental archaeology (also associated with "living history" and "reenactment"), or reconstructing past techniques and objects through modern experimentation using archaeological evidence. E.g. John Haldon's reconstruction of Greek Fire, or better yet the famous Ulfberth swords. Many areas of experimental archaeology requires an extensive knowledge of metallurgy, etc., the foundations of which are found in chemistry and materials science.

Anyways, those are my recommendations. I'm actually a double-degree student myself (B.S. Chemistry, B.A. History).

Statistics definitely if you wish to enter quantitative history. It is history told according to the numbers, so statistics is required.

Physics/chemistry/engineering if you want to dovetail into archaeology. These fields are important to conservation, reconstruction, drafting, and designing new instruments.

Computer Science and skills in Geographic Information Systems. Field mapping and global positioning are rapidly becoming used for all sorts of datasets. Throw in some programming and coding skills and you should be set.

It depends what subjects you plan to study. If you include anthropology as well as pure history, quite a few can help.

Historical linguistics is surprisingly hard and useful.

Soil geology, zoology to examine animal remains, botany to examine plant remains and/or pollen, and so on can help reconstruct archaeological sites.

Paleoclimatology.

Geology can also help identify when and how much soil erosion took place. Some scholars such as Michael Jones in The End of Roman Britain have attrbute mass famines and civilization collapse to soil erosion.

Zoology may also help understand the extent of the Plague of Justinian, since there have been disputes over the presence of Black Rats in Britain and other parts of northern Europe.

Anything which can shed light on demography can be useful.

Anything which can shed light on trade routes. Tin was widely traded in the Bronze Age. Amber continued much later, but if you can identify Baltic vs. North Sea Amber, that can help understand European trade routes.

I did my undergrad in genetics and grad school in history, and it’s been incredibly useful for a lot of applications. Biogeographical ancestry inference (eg Ancestry.com nonsense), historical plagues and epidemics, race as a cultural vs biological lens, etc.