lukeecomod / kustaa Goto Github PK

KUSTAA is a simple tool to estimate loading of suspended solids, total nitrogen and phosphorus from diffuse and point-sources in a catchment. The repository containst two versions:

Easy-to-use spreadsheet-tool in MS Excel, located in \excel folder. This follows Launiainen et al. 2015 Suomen Ympäristö 33 (In Finnish)

Python 3.x implementation for more advanced use, without GUI. Used in Bhattacharjee et al. 2021 Sci. Tot. Env.

References:

Launiainen, S., Sarkkola, S., Laurén, A., Puustinen, M., Tattari, S., Mattsson, T., Piirainen, S., Heinonen, J., Alakukku, L. ja Finér, L. 2015. KUSTAA-työkalu valuma-alueen vesistökuormituksen laskentaan. Suomen Ympäristökeskuksen Raportteja 33, 55p., ISBN 978-952-11-4374-8 (PDF)

Palviainen, M., Laurén, A., Launiainen, S. and Piirainen, S. 2016. Predicting the export and concentrations of organic carbon, nitrogen and phosphorus in boreal lakes by catchment characteristics and land use: a practical approach. Ambio (2016), doi:10.1007/s13280-016-0789-2

Bhattacharjee J., Marttila H., Launiainen S., Lepistö A., Kløve B. 2021. Combining Landsat image analysis, land-use statistics and land-use-specific export coefficient to predict river water quality after large scale peatland drainage. Science of the Total Environment, 779 146419, https://doi.org/10.1016/j.scitotenv.2021.146419

Minimalistic User-guide for the Python-version:

• Tested in Python 3.6; requires numpy and pandas.

• download and unzip to folder.

• in spyder or python prompt set working dir to folder

• sandbox.py gives examples how to use

• example input data (annual areas or polluter units) in .csv-files in \data

• note that all columns in input data must have (case-sensitive) keys in export_coefficient -dict

• export load coefficients defined in export_coefficients - modify according to your need

• note that column units in data (ha, ton, pieces, ...) must meet the way how export coefficients are defined.

• stores results in Kustaa-object fields; and prints them also to -csv -files. The logic is: 1) plain annual loads (and their variances) per source are stored in self.N_load and self.N_var -dataframes (and same for P and SS) by self.compute_loads(). These results can then be massaged in many ways to compute different results, 2) long-term sums and contributions to total load over given period are computed in self.summarize() similarly to Kustaa Excel -tool. 3) simple graphs can be produces also.

• two different approaches to compute forestry load: 1) Kalle - follows Finer et al. 2010 Suomen Ympäristö, and disributes loading to 10 year period after operation (works now only with default_coeff or at least needs group 'FORESTRY'). 2) 'annual' or whatever: assigns whole 10year load to the operation year

• for export coefficient values, see Excel-version and pdf-document(s)

• export coeffients for e.g. N are given in list: 'N': [ave, min, max], where min - max is typical variability range and assumed to correspond 90% of total variability later in the code.