ifnull 함수만 nvl 로 바꾸면 문법은 ANSI 표준이므로 동일합니다.

아. 하나 더 있네요. FROM 절의 AS a, AS b 에서 AS 빼세요.

-- 1. ANSI 표준 유지 --
SELECT a.loca_name
     , a.loca_id
     , NVL(b.surv_id, 0) cnt              -- IFNULL > NVL
  FROM location a                         -- AS 제거
  LEFT OUTER JOIN
       (SELECT loca_id
             , COUNT(surv_id) AS surv_id
          FROM survey_his
         GROUP BY loca_id
        ) b                               -- AS 제거
    ON (a.loca_id = b.loca_id)
-- GROUP BY a.loca_id                     -- 불필요한 부분
 ORDER BY a.loca_id                       -- 정렬 필요
;

-- 2. 서브쿼리(인라인뷰) 제거 --
SELECT a.loca_name
     , a.loca_id
     , COUNT(b.surv_id) cnt
  FROM location a
  LEFT OUTER JOIN survey_his b
    ON a.loca_id = b.loca_id
 GROUP BY a.loca_id, a.loca_name
 ORDER BY a.loca_id
;

-- 3. Oracle 아우터 조인
SELECT a.loca_name
     , a.loca_id
     , COUNT(b.surv_id) cnt
  FROM location a
     , survey_his b
 WHERE a.loca_id = b.loca_id(+)           -- 아우터조인(+)
 GROUP BY a.loca_id, a.loca_name
 ORDER BY a.loca_id
;