First things first:
No one knows exactly how Google picks one page over the other when displaying results in the SERPs. One thing is clear, the selection process is somewhat complicated by various selection processes while Google re-ranks the results of the various search queries (and yes Google does more than one search query every time you search and blends the results). We do know somethings about Googles selection process in this regard. Certainly enough has been shared over the years and therefore well understood.
As well, any PageRank of a page does not necessarily ensure that one page will perform higher in the SERPs. People do good work to make a page rank higher, of course, by adding inbound back links and that is good. Inbound links are one of a short list of significant factors. However, people often do not understand how this works and often overly simplify the process with link counts and apparent PageRank of the page that links to their site. There is much more to the link ranking process than just two factors.
I took a look at just a few on-page elements of both pages and quite frankly, they are pretty close. For example:
Fruit Picking jobs in Queensland, Australia | Fruit Picking jobs in
Queensland, for backpackers and travellers
Fruit Picking jobs Australia | Fruit Picking jobs for backpackers and
Queensland appears 39 times.
Queensland appears 33 times.
I looked at several other elements- this is just an example.
As well, I do not mean to imply that keyword density is a good thing or even a factor. It is not. In fact, Google uses keyword density as a negative factor and not a positive one in a direct sense.
There are some SEO changes that can be recommended. However that does not answer the question. So I will now address the question directly.
From the very beginning when Google was a research project, semantics was a major part of the PageRank algorithm and a significant consideration for link text. In at least 2002 and certainly in place in 2003, link semantics not only included the link text, but also the content block the link resides in (if at all) of the source page and the content of the target page.
As well, trust and authority caps can and often do exist for any page that is not of low value thus creating a curve in the algorithm that corrects too much value being passed. As part of this, link values also exist. The model of a PR 6 page with two links passing PR 3 to two target pages, was never right. With authority caps in place, the value of any link is evaluated and given a score of less than 1 with 1 conceptually being a perfect score and passing full value. A low value link would be .1 and a high value link would be .9. The value of a link has several factors such as placement of the link, the content block or blocks that the link resides in (if it does), the link text, and other factors. This means that the value passed through a link is based upon any existing authority cap where it applies calculated against the value of the link. This means that link counts mean very little in building the value of a target page. It is link quality and use that matters.
Keyword density is a myth. Sorta. It always was. Admittedly, at one point very early in the game, Google felt that the number of times any term was used was a strong indicator of the page topic. Semantics was always a part of the calculation and so density in of itself was not really a factor, however, a factor in the semantics calculation which pushed the density metric way down the line. As a result, some level of density made sense, however, that number was always rather low. Google did use density as a factor in that it was also an indicator of a pages attempt to be manipulative and should be discounted. Again, this was very early in the game.
Any pages PageRank is only a part of the process. PageRank is just one of all of the other many factors Google uses in deciding which pages should appear in the SERPs in response to the search intent. It is not a linear calculation of PageRank, but rather a long series of calcualtions against factors and scores that decide where a page should appear in the SERPs.
The most valuable indications of how a page should rank for search terms is the title tag, link text and value, the link source page content, and the pages content. Period. Yes there are other factors at play, however, just getting these simple factors right is enough to outrank most if not all of your competition which can also be your own sites pages.
As a disclaimer, I have not looked at your links. I did not need to.
When Google goes through the process of ranking pages to be displayed in the SERPs there are many factors that are weighed. Of this semantics comes into play much more than people realize.
In your case, the two pages do not distinguish themselves are being all that different. Your title tags, description meta-tags (and yes matches are made against this), and content are similar enough that Google has to make a decision. Google will always chose one page over another for search intent and the strongest matching page will get a boost to ensure survival.
I used https://www.textrazor.com/demo to examine the semantic scoring of your two pages. For the term Queensland, both pages score a 1 for the topic which is a perfect score. This means that for the topic of Queensland, in the semantics realm, one does not outrank the other. Keep in mind that semantics is not based upon keyword density, but rather scores usage of terms in relation to other terms and proximity as well as other factors. I could use another tools for deeper analysis, but it is not necessary in this case.
All things considered, one page likely does not outrank the other or if it does, it is not by much.
I want to also stress one other thing people forget about in regards to SEO. If one page satisfies more searches than another generally, it will have a higher score or sets of scores in Googles index as a result. It is not just the CTR of the SERP links over a series of searches over time, it is also the semantic scoring of the page. Your /jobs/fruit-picking-jobs/ page has a broader semantic (topical) score than your /jobs/fruit-picking-jobs/queensland/ page. This means two things: one, since the semantic topics of /jobs/fruit-picking-jobs/ are more numerous, the topics are also related and likely scores higher overall than /jobs/fruit-picking-jobs/queensland/ thus satisfying more search queries generally which is an important factor; and two, the /jobs/fruit-picking-jobs/queensland/ possesses a shorter list of topics and would match a smaller number of search queries directly. Content topic diversity is an important factor. Too much spells disaster in that the content is not focused and too little garners less attention which I suspect is the case here.