deletions | additions       diff --git a/introduction.tex b/introduction.tex  index 1a4fe69..3f6fcc0 100644  --- a/introduction.tex  +++ b/introduction.tex 
...
Variation in surface water quality at the catchment scale in the boreal landscape is mainly controlled by landscape heterogeneity \cite{humborg_nutrient_2004}. A major influence on surface water pH is proportion of peatlands in the catchment through the release of organic acids \cite{buffam_landscape-scale_2007} . Peatland cover also reduces the nitrite concentration in surface waters \cite{sponseller_patterns_2014}. Despite the clear effect of landscape characteristics on water chemistry this aspect has received little attention when examining the effect of wildfire on water chemistry. For example, wildfire can cause severe disturbance to peatlands and potentially increase oxidation of S with subsequent leaching as a result. Similary, nitrite may not be retained in peatlands after fire.   WHAT INFO IS MISSING?  -BEFORE AFTER EXP  -LONG TERM DATA  -FACTORS DETERMINING FIRE EFFECTS  -EARLY MEASUREMENTS A wildfire in Sweden in 2014 created the oppertunity to study the effect of wildfire on water chemistry in a managed landscape with a high cover of peatlands. To quantify the effects of wildfire on water quality, and to understand the drivers behind variation in water quality responses to fire, before-after data is needed and catchment needs to be replicated. The burnt area consists of multiple catchments allowing us to investigate local variation in post-fire responses. One of the catchments is included in a national water monitoring network enabling comparison with long-term trends in water chemistry. This before-after approach is complimented by comparing data with nearby long-term monitored catchments. Hence, compared to most studies, our study does not rely on only post-fire data and a few reference sites (see Mast 2013 and Betts and Jones, 2009 for other before-after studies).  THIS PAPERS AIM AND NOVELTY  A wildfire in Sweden in 2014 gave the oppertunity to study the effect of wildfire on water chemistry in a managed landscape with a high cover of peatlands. The burnt area consists of multiple catchments allowing us to investigate local variation in post-fire responses. One of the catchments is included in a national water monitoring network enabling comparison with long-term trends in water chemistry. This before-after approach is complimented by comparing data with nearby monitored catchments. Thus, our study can quantify the effect of the fire without relying on only post-fire data and a few reference sites (see Mast 2013 and Betts and Jones, 2009 for other examples).   -  The overarching goal of this study is to investigate the short-term (12 months) effects of the 2014 Swedish wildfire on stream and lake water chemistry. Downstream data from seven burned watersheds and two reference (unburned) watersheds are presented together with data from ten lakes. In addition, pre-fire data exist for two of the streams in the burned area, making this a unique opportunity to quantify the impact of the wildfire on water chemistry. Furthermore, we want explore if catchment characteristics can be associated with the post-fire water chemistry. In particular we tested the variables: i) the proportion of severely burned upland, ii) proportion drained peatlands (i.e. treed peatlands), iii) proportion of open (i.e. undrained) peatlands. Overall classification of burn severity in uplands and peatlands are determined by remote sensing techniques.