PROJECT FOR INTERCOMPARISON OF LANDSURFACE PARAMETERIZATION SCHEMES -APPLICATION OF SOME STRUCTURAL COMPLEXITY METRICS

Initiated in 1992, the international PILPS project aims to evaluate an d intercompare land-surface parameterization packages, destined for em bedding into atmospheric general circulation models. The Project for I ntercomparison of Landsurface Parameterization Schemes (PILPS) involve s 27 numerical submodels to describe the interaction of the land surfa ce with the overlying atmosphere. This project offers the opportunity of not only comparing the physical basis and simulation results of the se land-surface codes, but also for collecting software engineering me trics on the codes themselves. The existing PILPS infrastructure suppo rted the data collection of measures of the pieces of FORTRAN code in an organized fashion. A number of questions were included in a data ga thering exercise, via questionnaire, regarding the structural complexi ty of the codes. Even for this parsimonious set of metrics, adequate d ata were returned for only 7 of the 27 land-surface parameterization s chemes involved in the PILPS intercomparison. Results from these seven data sets are analyzed here in terms of control flow complexity and s ize. A second experiment is also described briefly. This was conducted to evaluate, subjectively, the overall ''complexity'' of four of the PILPS codes. Eight senior climate researchers, all of whom are also es tablished FORTRAN programmers, were asked to evaluate the code listing s using a questionnaire. These data were evaluated and their relations hip to the objective measures assessed. A surprisingly good correlatio n was found between many of the standard, objective metrics and subjec tive assessments of overall ''complexity.''